Thyristors and Triacs Power Semiconductor Applications

Philips Semiconductors
CHAPTER 6
Power Control with Thyristors and Triacs
6.1 Using Thyristors and Triacs
6.2 Thyristor and Triac Applications
6.3 Hi-Com Triacs
485
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Using Thyristors and Triacs
487
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
6.1.1 Introduction to Thyristors and Triacs
Brief summary of the thyristor family
The term thyristor is a generic name for a semiconductor
switch having four or more layers and is, in essence, a
p-n-p-n sandwich. Thyristors form a large family and it is
helpful to consider the constituents which determine the
type of any given thyristor. If an ohmic connection is made
to the first p region and the last n region, and no other
connection is made, the device is a diode thyristor. If an
additional ohmic connection is made to the intermediate n
region (n gate type) or the intermediate p region (p gate
type), the device is a triode thyristor. If an ohmic connection
ismade toboth intermediate regions, the device isa tetrode
thyristor. All such devices have a forward characteristic of
the general form shown in Fig. 1.
There are three types of thyristor reverse characteristic:
blocking (as in normal diodes), conducting (large reverse
currents at low reverse voltages) and approximate mirror
imageof theforward characteristic (bidirectional thyristors).
Reverse blocking devices usually have four layers or less
whereas reverse conducting and mirror image devices
usually have five layers.
The simplest thyristor structure, and the most common, is
the reverse blocking triode thyristor (usually simply referred
to as the ’thyristor’ or SCR ’silicon controlled rectifier’). Its
circuit symbol and basic structure are shown in Fig. 2.
The most complex common thyristor structure is the
bidirectional triode thyristor, or triac. The triac (shown in
Fig. 3) isable topass current bidirectionally andis therefore
an a.c. power control device. Its performance is that of a
pair of thyristors in anti-parallel with a single gate terminal.
The triac needs only one heatsink, but this must be large
enough to remove the heat caused by bidirectional current
flow. Triac gate triggering circuits must be designed with
careto ensurethat unwantedconduction, ie. loss of control,
does not occur when triggering lasts too long.
Thyristors and triacs are both bipolar devices. They have
very lowon-state voltages but, becausethe minority charge
carriers in the devices must be removed before they can
block an applied voltage, the switching times are
comparatively long. This limits thyristor switching circuits to
low frequency applications. Triacs are used almost
exclusivelyat mains supply frequenciesof 50or 60Hz, while
in some applications this extends up to the 400Hz supply
frequency as used in aircraft.
The voltage blocking capabilities of thyristors and triacs are
quite high: the highest voltage rating for the Philips range
is 800V, while the currents (I
T(RMS)
) range from0.8A to 25A.
The devices are available as surface mount components,
or as non-isolated or isolated discrete devices, depending
on the device rating.
Fig. 1 Thyristor static characteristic
Fig. 2 Thyristor circuit symbol and basic structure
Fig. 3 Triac circuit symbol and basic structure
On-state
characteristic
Off-state
characteristic
Avalanche
breakdown
region
Reverse
characteristic
Reverse
current
Forward
current
Reverse
voltage
Forward
voltage
I
L
I
H
V
(BO)
I = 0
G
I > 0
G
Anode Anode
Gate
Gate
Cathode Cathode
p
n
p
n
J1
J2
J3
MT1
MT2
Gate
Gate
MT1
MT2
n
n
n
n
p
p
489
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Thyristor operation
Theoperationof thethyristor canbeunderstoodfromFig. 2.
When the thyristor cathode is more positive than the anode
then junctions J1 and J3 are reverse biased and the device
blocks. When the anode is more positive than the cathode,
junctions J1 and J3 are forward biased. As J2 is reverse
biased, then the device still blocks forward voltage. If the
reverse voltage across J2 is made to reach its avalanche
breakdown level then the device conducts like a single
forward-biased junction.
The ’two transistor’ model of Fig. 4 can be used to consider
the p-n-p-n structure of a thyristor as the interconnection of
an npn transistor T
1
and a pnp transistor T
2
. The collector
of T
1
provides the base current for T
2
. Base current for T
1
is provided by the external gate current in addition to the
collector current from T
2
. If the gain in the base-collector
loop of T
1
and T
2
exceeds unity then the loop current can
be maintained regeneratively. When this condition occurs
then both T
1
and T
2
are driven into saturation and the
thyristor issaidtobe’latched’. Theanodetocathodecurrent
is then only limited by the external circuit.
Fig. 4 ’Two transistor’ model of a thyristor
There are several mechanisms by which a thyristor can be
latched. The usual method is by a current applied to the
gate. This gate current starts the regenerative action in the
thyristor and causes the anode current to increase. The
gains of transistors T
1
and T
2
are current dependent and
increase as the current through T
1
and T
2
increases. With
increasing anode current the loop gain increases
sufficiently such that the gate current can be removed
without T
1
and T
2
coming out of saturation.
Thus a thyristor can be switched on by a signal at the gate
terminal but, because of the way that the current then
latches, the thyristor cannot be turned off by the gate. The
thyristor must be turned off by using the external circuit to
break the regenerative current loop between transistors T
1
and T
2
. Reverse biasing the device will initiate turn-off once
the anode current drops below a minimum specified value,
called the holding current value, I
H
.
Thyristor turn-on methods
Turn-on by exceeding the breakover
voltage
When the breakover voltage, V
BO
, across a thyristor is
exceeded, the thyristor turns on. The breakover voltage of
a thyristor will be greater than the rated maximum voltage
of the device. At the breakover voltage the value of the
thyristor anode current is called the latching current, I
L
.
Breakover voltage triggering is not normally used as a
triggeringmethod, andmost circuit designs attempt toavoid
its occurrence. When a thyristor is triggered by exceeding
V
BO
the fall time of the forward voltage is quite low (about
1/20th of the time taken when the thyristor is
gate-triggered). As a general rule, however, although a
thyristor switches faster with V
BO
turn-on than with gate
turn-on, the permitted di/dt for breakover voltage turn-on is
lower.
Turn-on by leakage current
As the junction temperature of a thyristor rises, the leakage
current also increases. Eventually, if the junction
temperature is allowed to rise sufficiently, leakage current
would become large enough to initiate latching of the
regenerative loop of the thyristor and allow forward
conduction. At a certain critical temperature (above T
j(max)
)
the thyristor will not support any blocking voltage at all.
Turn-on by dV/dt
Any p-n junction has capacitance - the larger the junction
area the larger the capacitance. If a voltage ramp is applied
across the anode-to-cathode of a p-n-p-n device, a current
will flow in the device to charge the device capacitance
according to the relation:
If the charging current becomes large enough, the density
of moving current carriers in the device induces switch-on.
Turn-on by gate triggering
Gate triggering is the usual method of turning a thyristor on.
Application of current to the thyristor gate initiates the
latching mechanismdiscussed in the previous section. The
characteristic of Fig. 1 showed that the thyristor will switch
toits on-state condition with forward bias voltages less than
V
BO
when the gate current is greater than zero. The gate
current and voltage requirements which ensure triggering
of a particular device are always quoted in the device data.
As thyristor triggering characteristics are temperature
dependant, the amplitude and duration of the gate pulse
must be sufficient to ensure that the thyristor latches under
all possible conditions.
T1
T2
Anode
Cathode
Gate
i
A
i
G
i
C
· C.
dv
dt
(1)
490
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
During gate turn-on, the rate of rise of thyristor anode
current dI
F
/dt is determined by the external circuit
conditions. However, the whole active area of the thyristor
(or triac) cannot be turned on simultaneously: the area
nearest to the gate turns on first, followed by the remainder
of the device. At turn-on it is important that the rate of rise
of current does not exceed the specified rating. If dI
F
/dt is
excessive then only a limited area of the device will have
been turned on as the anode current increases. The
resulting localised heating of the device will cause
degradation and could lead to eventual device failure.
A suitably high gate current and large rate of rise of gate
current (dI
G
/dt) ensures that the thyristor turns on quickly
(providing that the gate power ratings are not exceeded)
thus increasing the thyristor turn-on di/dt capability. Once
the thyristor has latched then the gate drive can be reduced
or removed completely. Gate power dissipation can also
be reduced by triggering the thyristor using a pulsed signal.
Triac operation
The triac can be considered as two thyristors connected in
antiparallel as shown in Fig. 5. The single gate terminal is
common to both thyristors. The main terminals MT1 and
MT2 are connected to both p and n regions of the device
and the current path through the layers of the device
depends upon the polarity of the applied voltage between
the main terminals. The device polarity is usually described
with reference to MT1, where the term MT2+ denotes that
terminal MT2 is positive with respect to terminal MT1.
Fig. 5 Anti parallel thyristor representation of a triac
The on-state characteristic of the triac is similar to that of a
thyristor and is shown in Fig. 6. Table 1 and Fig. 7
summarise the different gate triggering configurations for
triacs.
Due to the physical layout of the semiconductor layers in a
triac, the values of latching current (I
L
), holding current (I
H
)
and gate trigger current (I
GT
) vary slightly between the
different operating quadrants. In general, for any triac, the
latching current is slightly higher in the second (MT2+, G-)
quadrant than the other quadrants, whilst the gate trigger
current is slightly higher in fourth (MT2-, G+) quadrant.
Fig. 6 Triac static characteristic
Quadrant Polarity of MT2 wrt MT1 Gate polarity
1 (1+) MT2+ G+
2 (1-) MT2+ G-
3 (3-) MT2- G-
4 (3+) MT2- G+
Table 1. Operating quadrants for triacs
Fig. 7 Triac triggering quadrants
For applications where the gate sensitivity is critical and
wherethe devicemust trigger reliablyandevenlyfor applied
voltages in both directions it may be preferable to use a
negative current triggering circuit. If the gate drive circuit is
arranged so that only quadrants 2 and 3 are used (i.e. G-
operation) then the triac is never used inthe fourth quadrant
where I
GT
is highest.
On-state
Off-state
Reverse
current
Forward
current
Reverse
voltage
Forward
voltage
I
L
I
H
V
(BO)
I = 0
G I > 0
G
On-state
Off-state
L
I
H
I
(BO)
V
I = 0
G
I > 0
G
T2-
T2+
Quadrant 1
Quadrant 2
Quadrant 4
Quadrant 3
G+ G-
MT2+
MT2-
I
G
I
G
I
G
I
G
+ +
-
-
+
-
+
-
MT2
MT1
491
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
For some applications it is advantageous to trigger triacs
with a pulsating signal and thus reduce the gate power
dissipation. To ensure bidirectional conduction, especially
with a very inductive load, the trigger pulses must continue
until theendof eachmains half-cycle. If singletrigger pulses
are used, one-way conduction (rectification) results when
the trigger angle is smaller than the load phase angle.
Philips produce ranges of triacs having the same current
and voltage ratings but with different gate sensitivities. A
devicewitharelativelyinsensitivegatewill bemoreimmune
to false triggering due to noise on the gate signal and also
will be more immune to commutating dv/dt turn-on.
Sensitive gate triacs are used in applications where the
device is driven from a controller IC or low power gate
circuit.
The diac
It is also worthwhile to consider the operation and
characteristics of thediac in the context of multilayer bipolar
devices. The diacismorestrictlyatransistor thanathyristor,
but has an important role in many thyristor and triac
triggering circuits. It is manufactured by diffusing an n-type
impurity into both sides of a p-type slice to give a two
terminal device with symmetrical electrical characteristics.
As shown in the characteristic of Fig. 8, the diac blocks
applied voltages in either direction until the breakover
voltage, V
BO
is reached. The diac voltage then breaks back
to a lower output voltageV
O
. Important diac parameters are
breakover voltage, breakover current and breakback
voltage as shown in the figure.
Fig. 8 Diac static characteristic and circuit symbol
Gate requirements for triggering
To a first approximation, the gate-to-cathode junction of a
thyristor or triac acts as a p-n diode. The forward
characteristic is as shown in Fig. 9. For a given thyristor
type there will be a spread in forward characteristics of gate
junctions and a spread with temperature.
Fig. 9 Thyristor gate characteristic
The gate triggering characteristic is limited by the gate
power dissipation. Figure 9 also shows the continuous
power rating curve (P
G(AV)
=0.5W) for a typical device and
the peak gate power curve (P
GM(max)
=5W). When designing
a gate circuit to reliably trigger a triac or thyristor the gate
signal must lie on a locus within the area of certain device
triggering. Continuous steadyoperation woulddemand that
the 0.5W curve be used to limit the load line of the gate
drive circuit. For pulsed operation the triggering locus can
be increased. If the 5Wpeak gate power curve is used, the
duty cycle must not exceed
At theother endof the scale, thelevel belowwhichtriggering
becomes uncertain is determined by the minimum number
of carriers needed in the gate-cathode junction to bring the
thyristor into conduction by regenerative action. The trigger
circuit load line must not encroach into the failure to trigger
region shown in Fig. 9 if triggering is to be guaranteed. The
minimumvoltageandminimumcurrent totrigger all devices
(V
GT
and I
GT
) decreases with increasing temperature. Data
sheets for Philips thyristors and triacs showthe variation of
V
GT
and I
GT
with temperature.
Thyristor commutation
A thyristor turns off by a mechanism known as ’natural
turn-off’, that is, when the main anode-cathode current
drops belowthe holding value. It is important to remember,
however, that the thyristor will turn on again if the reapplied
forward voltage occurs before a minimum time period has
elapsed; this is because the charge carriers in the thyristor
at the time of turn-off take a finite time to recombine.
Thyristor turn-off is achieved by two main methods - self
commutation or external commutation.
Gate voltage, V
G
(V)
Gate current, I
G
(A)
P = 5W
= 0.1
= 1.0
V
GT
I
GT
Failure
to trigger
GM(max)
P = 0.5W
G(AV)
Gate power
ratings
Gate-cathode
characteristic
δ
max
·
P
G(AV)
P
GM
·
0.5
5
· 0.1 (2)
Reverse
current
Forward
current
Reverse
voltage
Forward
voltage
V
(BO)
(BO)
V
I
(BO)
I
(BO)
V
O
V
O
Breakback
voltage
492
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Self Commutation
In self-commutation circuits the thyristor will automatically
turn off at a predetermined time after triggering. The
thyristor conduction period is determined by a property of
the commutation circuit, such as the resonant cycle of an
LC-circuit or the Volt-Second capability of a saturable
inductor. The energy needed for commutation is delivered
by a capacitor included in the commutation circuit.
LC circuit in series with the thyristor
When the thyristor is triggered, the resulting main current
excites the resonant circuit. After half a resonant cycle, the
LC circuit starts to reverse the anode current and turns the
thyristor off. The thyristor conduction interval is half a
resonant cycle. It is essential for proper commutation that
the resonant circuit be less than critically damped. Fig. 10
shows the circuit diagram and the relevant waveforms for
this arrangement.
LC Circuit in parallel with the thyristor
Initially the capacitor charges to the supply voltage. When
the thyristor is triggered the load current flows but at the
same time the capacitor discharges through the thyristor in
the forward direction. When the capacitor has discharged
(i.e. after one resonant half-cycle of the LCcircuit), it begins
to charge in the opposite direction and, when this charging
current is greater than the thyristor forward current, the
thyristor turns off. The circuit diagram and commutation
waveforms are shown in Fig. 11.
External commutation
If the supply is an alternating voltage, the thyristor can
conduct only during the positive half cycle. The thyristor
naturally switches off at the end of each positive half cycle.
The circuit and device waveforms for this method of
commutation are shown in Fig. 12. It is important to ensure
that the duration of a half cycle is greater than the thyristor
turn-off time.
Reverse recovery
In typical thyristors the reverse recovery time is of the order
of a few micro-seconds. This time increases with increase
of forward current andalso increases as the forward current
decay rate, dI
T
/dt, decreases. Reverse recovery time is the
period during which reverse recovery current flows (t
1
to t
3
in Fig. 13) and it is the period between the point at which
forward current ceases and the earliest point at which the
reverse recovery current has dropped to 10% of its peak
value.
Fig. 10 Commutation using a series LC circuit
Fig. 11 Commutation using a parallel LC circuit
R
R
leakage
L
C
E
+
I
thyristor
R
L
C
E
I
R
+
I
thyristor
493
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Reverse recovery current can cause high values of turn-on
current in full-wave rectifier circuits (where thyristors are
used as rectifying elements) and in certain inverter circuits.
It should also be remembered that, if thyristors are
connected in series, the reverse voltage distribution can be
seriously affected by mismatch of reverse recovery times.
Fig. 12 Thyristor commutation in an a.c. circuit
Fig. 13 Thyristor turn-off characteristics
Turn-off time
Turn-off time is the interval between the instant when
thyristor current reverses andthe point at whichthe thyristor
can block reapplied forward voltage (t
1
to t
4
in Fig. 13). If
forward voltage is applied to a thyristor too soon after the
main current has ceased to flow, the thyristor will turn on.
The circuit commutated turn-off time increases with:
-junction temperature
-forward current amplitude
-rate of fall of forward current
-rate of rise of forward blocking voltage
-forward blocking voltage.
Thus the turn-off time is specified for defined operating
conditions. Circuit turn-off time is the turn-off time that the
circuit presents tothe thyristor; it must, of course, be greater
than the thyristor turn-off time.
Triac commutation
Unlike the thyristor, the triac can conduct irrespective of the
polarity of the applied voltage. Thus the triac does not
experience a circuit-imposed turn-off time which allows
each anti-parallel thyristor to fully recover from its
conducting state as it is reverse biased. As the voltage
across the triac passes through zero and starts to increase,
then the alternate thyristor of the triac can fail to block the
applied voltage and immediately conduct in the opposite
direction. Triac-controlled circuits therefore require careful
design in order to ensure that the triac does not fail to
commutate (switch off) at the end of each half-cycle as
expected.
It is important to consider the commutation performance of
devices in circuits where either dI/dt or dV/dt can be large.
In resistive load applications (e.g. lamp loads) current
surges at turn-on or during temporary over-current
conditions may introduce abnormally high rates of change
of current which may cause the triac to fail to commutate.
In inductive circuits, such as motor control applications or
circuits where a dc load is controlled by a triac via a bridge
rectifier, it is usually necessary to protect the triac against
unwanted commutation due to dv
(com)
/dt.
The commutating dv
(com)
/dt limit for a triac is less than the
static dv/dt limit because at commutation the recently
conducting portion of the triac which is being switched off
has introduced stored charge to the triac. The amount of
stored charge depends upon the reverse recovery
characteristics of the triac. It is significantly affected by
junction temperature and the rate of fall of anode current
prior to commutation (dI
(com)
/dt). Following high rates of
change of current the capacity of the triac to withstand high
reapplied rates of change of voltage is reduced. Data sheet
specifications for triacs give characteristics showing the
R
i
thyristor
I
T
I
R
V
D
V
R
dI
T
dt
dV
D
dt
t
0
t
1
t
2
t
3
t
4
494
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
maximum allowable rate of rise of commutating voltage
against device temperature and rate of fall of anode current
which will not cause a device to trigger.
Fig. 14 Inductive load commutation with a triac
Consider the situation when a triac is conducting in one
direction and the applied ac voltage changes polarity. For
the case of an inductive load the current in the triac does
not fall to its holding current level until some time later. This
is shown in Fig. 14. At the time that the triac current has
reached the holding current the mains voltage has risen to
some value and so the triac must immediately block that
voltage. The rate of rise of blocking voltage following
commutation (dv
(com)
/dt) can be quite high.
The usual method is to place a dv/dt-limiting R-C snubber
inparallel with the triac. Additionally, because commutating
dv/dt turn-on is dependent upon the rate of fall of triac
current, then in circuits with large rates of change of anode
current, the ability of a triac to withstand high rates of rise
of reapplied voltage is improved by limiting the di/dt using
a series inductor. This topic is discussed more fully in the
section entitled ’Using thyristors and triacs’.
Conclusions
This article has presented the basic parameters and
characteristics of triacs and thyristors and shown how the
structure of the devices determines their operation.
Important turn-on and turn-off conditions and limitations of
the devices have been presented in order to demonstrate
the capabilities of the devices and showthe designer those
areas which require careful consideration. The device
characteristics which determine gate triggering
requirements of thyristors and triacs have been presented.
Subsequent articles in this chapter will deal with the use,
operation and limitations of thyristors and triacs in practical
applications, and will present some detailed design and
operational considerations for thyristors and triacs in phase
control and integral cycle control applications.
V
DWM
-dI/dt
dV
com
/dt
Time
Time
Time
Supply
voltage
Load
current
Voltage
across
triac
Trigger
pulses
Current
495
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
6.1.2 Using Thyristors and Triacs
This chapter is concerned with the uses, operation and
protection of thyristors andtriacs. Two types of circuit cover
the vast majority of applications for thyristors and triacs:
static switching circuits and phase control circuits. The
characteristics and uses of these two types of circuit will be
discussed. Various gate drive circuits and protection
circuits for thyristor and triacs are also presented. The use
of thesecircuits will enabledesignerstooperatethedevices
reliably and within their specified limits.
Thyristor and triac control techniques
There are two main techniques of controlling thyristors and
triacs - on-off triggering (or static switching) and phase
control. In on-off triggering, the power switch is allowed to
conduct for a certain number of half-cycles and then it is
kept off for a number of half-cycles. Thus, by varying the
ratio of "on-time" to "off-time", the average power supplied
to the load can be controlled. The switching device either
completely activates or deactivates the load circuit. In
phase control circuits, the thyristor or triac is triggered into
conduction at some point after the start of each half-cycle.
Control is achieved on a cycle-by-cycle basis by variation
of the point in the cycle at which the thyristor is triggered.
Static switching applications
Thyristors and triacs are the ideal power switching devices
for many high power circuits such as heaters, enabling the
load to be controlled by a low power signal, in place of a
relay or other electro-mechanical switch.
Ina high power circuit where the power switch may connect
or disconnect the load at any point of the mains cycle then
large amounts of RFI (radio frequency interference) are
likely to occur at the instants of switching. The large
variations in load may also cause disruptions to the supply
voltage. The RFI and voltage variation produced by high
power switching in a.c. mains circuits is unacceptable in
many environments and is controlled by statutory limits.
The limits depend upon the type of environment (industrial
or domestic) and the rating of the load being switched.
RFI occurs at any time when there is a step change in
current caused by the closing of a switch (mechanical or
semiconductor). The energy levels of this interference can
bequitehighin circuits suchas heatingelements. However,
if the switch is closed at the moment the supply voltage
passes through zero there is no step rise in current and
thus no radio frequency interference. Similarly, at turn-off,
a large amount of high frequency interference can be
caused by di/dt imposed voltage transients in inductive
circuits.
Circuit-generated RFI can be almost completely eliminated
by ensuring that the turn-on switching instants correspond
to the zero-crossing points of the a.c. mains supply. This
technique is known as synchronous (or zero voltage)
switching control as opposed to the technique of allowing
the switching points to occur at any time during the a.c.
cycle, which is referred to as asynchronous control.
In a.c. circuits using thyristors and triacs the devices
naturally switch off when the current falls below the device
holding current. Thus turn-off RFI does not occur.
Asynchronous control
In asynchronous control the thyristor or triac may be
triggered at a point in the mains voltage other than the zero
voltage crossover point. Asynchronous control circuits are
usually relatively cheap but liable to produce RFI.
Synchronous control
In synchronous control systems the switching instants are
synchronised with zero crossings of the supply voltage.
They also have the advantage that, as the thyristors
conduct over complete half cycles, the power factor is very
good. This method of power control ismostly usedtocontrol
temperature. The repetition period, T, is adjusted to suit the
controlled process (within statutory limits). Temperature
rippleiseliminated whenthe repetitionperiodismade much
smaller than the thermal time constant of the system.
Figure 1 shows the principle of time-proportional control.
RFI andturn-ondi/dt are reduced, andthe best power factor
(sinusoidal load current) is obtained by triggering
synchronously. The average power delivered to a resistive
load, R
L
, is proportional to t
on
/T (i.e. linear control) and is
given by equation 1.
where: T is the controller repetition period
t
on
is controller ’on’ time
V
(RMS)
is the rms a.c. input voltage.
Elsewhere in this handbook the operation of a controller i.c.
(the TDA1023) is described. This device is specifically
designed to implement time-proportional control of heaters
using Philips triacs.
P
out
·
V
(RMS)
2
R
L
.
t
on
T
(1)
497
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 1 Synchronous time-proportional control
Phase control
Phase control circuits are used for low power applications
such as lamp control or universal motor speed control,
where RFI emissions can be filtered relatively easily. The
power delivered to the load is controlled by the timing of
the thyristor (or triac) turn-on point.
The two most common phase controller configurations are
’half wave control’, where the controlling device is a single
thyristor and’full wavecontrol’, where the controllingdevice
is a triac or a pair of anti-parallel thyristors. These two
control strategies are considered in more detail below:
Resistive loads
The operation of a phase controller with a resistive load is
the simplest situation to analyse. Waveforms for a full wave
controlled resistive load are shown in Fig. 2. The triac is
triggered at angle δ, and applies the supply voltage to the
load. The triac then conducts for the remainder of the
positivehalf-cycle, turningoff whentheanodecurrent drops
below the holding current, as the voltage becomes zero at
θ=180˚. The triac is then re-triggered at angle (180+δ)˚, and
conducts for the remainder of the negative half-cycle,
turning off when its anode voltage becomes zero at 360˚.
The sequence is repeated giving current pulses of
alternating polarity which are fed to the load. The duration
of each pulse is the conduction angle α, that is (180-δ)˚.
The output power is therefore controlled by variation of the
trigger angle δ.
For all values of α other than α=180˚ the load current is
non-sinusoidal. Thus, because of the generation of
harmonics, the power factor presented to the a.c. supply
will be less than unity except when δ=0.
For a sinusoidal current the rectified mean current, I
T(AV)
,
and the rms current, I
T(RMS)
, are related to the peak current,
I
T(MAX)
, by equation 2.
Fig. 2 Phase controller - resistive load
where
From equation 2 the ’crest factor’, c, (also known as the
’peak factor’) of the current waveform is defined as:
The current ’form factor,’ a, is defined by:
Thus, for sinusoidal currents:
For the non-sinusoidal waveforms which occur in a phase
controlled circuit, the device currents are modified due to
thedelay whichoccurs beforethe power deviceis triggered.
Thecrest factor of equation4andtheformfactor of equation
5 can be used to describe variation of the current
waveshape from the sinusoidal case.
t
ON
T
Input
voltage
Trigger
signal
Output
current
Trigger
Conduction
angle,
Voltage
Current
Supply
voltage
Triac
Triac
Device
triggers
Trigger
angle,
O = wt
O = wt
O = wt
I
T(AV)
·
2.I
T(MAX)
π
· 0.637 I
T(MAX)
I
T(RMS)
·
I
T(MAX)
√2
· 0.707 I
T(MAX)
(2)
I
T(MAX)
·
V
T(MAX)
R
L
·
√2V
(RMS)
R
L
(3)
Crest factor, c ·
I
T(MAX)
I
T(RMS)
(4)
Form factor, a ·
I
T(RMS)
I
T(AV)
(5)
a ·
I
T(RMS)
I
T(AV)
· 1.111; c ·
I
T(MAX)
I
T(RMS)
· 1.414 (6)
498
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Half wave controller
Figure 3a) shows the simplest type of thyristor half-wave
phase controller for a resistive load. The load current
waveformis given in Fig. 3b). The variation of average load
current, I
T(AV)
, rms load current, I
T(RMS)
and load power over
the full period of the a.c mains, with trigger angle are given
in equation 7.
N.B. When using equation 7 all values of α must be in
radians. For each case the maximum value occurs when
α=180˚ (α=π radians).
At α=180˚ the crest factor and form factor for a half wave
controller are given by:
Full wave controller
Figure 4 shows the circuit and load current waveforms for
a full-wave controller using two antiparallel thyristors, or a
triac, as the controlling device. The variation of rectified
mean current, I
T(AV)
, rms current, I
T(RMS)
, and load power with
trigger angle are given by equation 9.
N.B. When using equation 9 all value of α must be in
radians. For each case the maximum value occurs when
α=180˚ (α=π radians).
Fig. 3 Half wave control
Fig. 4 Full wave control
The variation of normalised average current, I
T(AV)
/I
T(AV)max
,
rms current I
T(RMS)
/I
T(RMS)max
, and power, P
(out)
/P
(out)max
, for
equations 7 and 9 are plotted in Fig. 5.
Figure 6 shows the variation of current form factor with
conduction angle for the half wave controller and the full
wave controller of Figs. 3 and 4.
Fig. 5 Current and power control using conduction
angle
a) b)
Trigger
Voltage
Current
Supply
voltage
Thyristor
Thyristor
I
T(MAX)

I
T(AV)
· I
T(AV) max
.
(1 − cosα)
2
I
T(RMS)
· I
T(RMS) max
.
¸

¸
α−
1
2
sin2α
π
_

,
1
2
P
(out)
· P
(out) max
.
¸

¸
α−
1
2
sin2α
π
_

,
I
T(AV) max
·
I
T(MAX)
π
I
T(RMS) max
·
I
T(MAX)
2
P
(out) max
·
I
T(MAX)
2
R
L
4
(7)
a) b)
Trigger
Voltage
Current
Supply
voltage
Triac
Triac
I
T(MAX)
I
T(MAX)
a ·
I
T(RMS)
I
T(AV)
· 1.571; c ·
I
T(MAX)
I
T(RMS)
· 2.0 (8)
I
T(AV)
I
T(AV)max
I
T(RMS)
I
T(RMS)max
P
(OUT)
P
(OUT)max
A
m
p
l
i
t
u
d
e
Conduction angle
0 30 60 90 120 150 180
0
0.2
0.4
0.6
0.8
1

I
T(AV)
· I
T(AV) max
.
(1 − cosα)
2
I
T(RMS)
· I
T(RMS) max
.
¸

¸
α−
1
2
sin2α
π
_

,
1
2
P
(out)
· P
(out) max
.
¸

¸
α−
1
2
sin2α
π
_

,
I
T(AV) max
·
2I
T(MAX)
π
I
T(RMS) max
·
I
T(MAX)
√2
P
(out) max
·
I
T(MAX)
2
R
L
2
(9)
499
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 6 Variation of form factor with conduction angle
Inductive loads
The circuit waveforms for a phase controller with an
inductive load or an active load (for example, a motor) are
more complex than those for a purely resistive load. The
circuit waveforms depend on the load power factor (which
may be variable) as well as the triggering angle.
For a bidirectional controller (i.e triac or pair of anti-parallel
thyristors), maximumoutput, that is, sinusoidal loadcurrent,
occurs when the trigger angle equals the phase angle.
When the trigger angle, δ, is greater than the load phase
angle, ϕ, then the load current will become discontinuous
andthetriac (or thyristor) will block someportionof the input
voltage until it is retriggered.
If the trigger angle is less than the phase angle then the
load current in one direction will not have fallen back to zero
at the time that the device is retriggered in the opposite
direction. This is shown in Fig. 7. The triac fails to be
triggered as the gate pulse has finished and so the triac
then acts as a rectifier. In Fig. 7 the triac is only triggered
by the gate pulses when the applied supply voltage is
positive (1+ quadrant). However, the gate pulses which
occur one half period later have no effect because the triac
is still conducting in the opposite direction. Thus
unidirectional current flows in the main circuit, eventually
saturating the load inductance.
This problem can be avoided by using a trigger pulse train
as shown in Fig. 8. The triac triggers on the first gate pulse
after the load current has reached the latching current I
L
in
the 3+ quadrant. The trigger pulse train must cease before
the mains voltage passes through zero otherwise the triac
will continue to conduct in the reverse direction.
Fig. 7 Triac triggering signals - single pulse
Fig. 8 Triac triggering signals - pulse train
Gate circuits for thyristors and triacs
As discussed in the introductory article of this chapter, a
thyristor or triac can be triggered into conduction when a
voltageof theappropriatepolarityisappliedacrossthemain
terminals and a suitable current is applied to the gate. This
can be achieved using a delay network of the type shown
in Fig. 9a). Greater triggering stability and noise immunity
can be achieved if a diac is used (see Fig. 9b). This gives
a trigger circuit which is suitable for both thyristors and
triacs.
Figure 10 shows several alternative gate drive circuits
suitable for typical triac and thyristor applications. In each
circuit the gate-cathode resistor protects the device from
false triggering due to noise - this is especially important
for sensitivegate devices. Inadditionopto-isolated thyristor
and triac drivers are available which are compatible with
the Philips range of devices.
Form
factor
I
T(RMS)
I
T(AV)
Conduction angle
0 30 60 90 120 150 180
0
1
2
3
4
5
6
Half-wave rectifier
Full-wave rectifier
Trigger
Device fails
to trigger
Conduction
angle
Voltage
Current
Supply
voltage
Triac
Triac
Inductor
iron core
saturation
Device
triggers
Trigger
Conduction
angle
Voltage
Current
Supply
voltage
Triac
Triac
Device
triggers
Device
triggers
Fails to
trigger
500
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 10 Alternative triac triggering circuits
1k0
220R
BT145
Load
1k0
BT145
Load
180R
10k
12V
BC337
1k0
Load
10k
12V
1k0
Load
10k
12V
BC337
4k7
100nF
BT145
BT145 BAW62
2:1
a)
b)
Fig. 9 Basic triac triggering circuits
In some applications it may be necessary to cascade a
sensitive gate device with a larger power device to give a
sensitive gate circuit with a high power handling capability.
A typical solution which involves triggering the smaller
device (BT169) from a logic-level controller to turn on the
larger device (BT151) is shown in Fig. 11.
Figure 12 shows an isolated triac triggering circuit suitable
for zero voltage switching applications. This type of circuit
is also known as a solid state relay (SSR). The function of
the Q1/R2/R3 stage is that the BC547 is on at all instants
in time when the applied voltage waveformis high and thus
holds the BT169 off. If the BT169 is off then no gate signal
is applied to the triac and the load is switched off.
Fig. 11 Master-slave thyristor triggering circuit
Fig. 12 Opto-isolated triac triggering circuit
R
E
I
R
+
R
E
I
R
+
BT151 BT169
R
I
R
+
-
R1
R3
R2 R4
BC547
BT169
100R
1K0
BT138
100R
100nF
Q1
501
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
If the input signal is switched high then the photo-transistor
turns on. If this occurs when the mains voltage is high then
Q1 remains on. When the line voltage passes through its
next zero crossing in either direction the photo transistor
ensures that Q1 stays off long enough for the BT169 to
trigger. This then turns the triac on. Once the thyristor turns
on, the drive circuit is deprived of its power due to the lower
voltage drop of the BT169. The triac is retriggered every
half cycle.
Voltage transient protection
Thereare three major sources of transient which may affect
thyristor and triac circuits:
-the mains supply (e.g. lightning)
-other mains and load switches (opening and closing)
-the rectifying and load circuit (commutation)
In order to ensure reliable circuit operation these transients
must be suppressed by additional components, removed
at source or allowed for in component ratings.
Three types of circuit are commonly employed to suppress
voltage transients - a snubber network across the device,
a choke between the power device and external circuit or
an overvoltage protection such as a varistor.
Series line chokes
A series choke may be used to limit peak fault currents to
assist in the fuse protection of thyristors and triacs. If the
choke is used in conjunction with fuse protection, it must
retain its inductance to very large values of current, and so
for this reason it is usually an air-cored component.
Alternatively, if thechokeis onlyrequiredtoreducethedv/dt
across non-conducting devices then the inductance needs
onlyto be maintained up toquite lowcurrents. Ferrite-cored
chokes may be adequate provided that the windings are
capable of carrying the full-load current. Usually only a few
microhenries of inductance are required to limit the circuit
di/dt to an acceptable level. This protects the devices from
turning on too quickly and avoids potential device
degradation.
For instance, a 220V a.c. supply with 20µH source
inductance gives a maximum di/dt of (220√2)/20=16A/µs.
Chokes used to soften commutation should preferably be
saturable so as to maintain regulation and avoid
deterioration of the power factor. As their impedance
reduces at high current, they have very little effect on the
inrush current.
The addition of di/dt limiting chokes is especially important
in triac circuits where the load is controlled via a bridge
rectifier. At the voltage zero-crossing points the conduction
transfers between diodes in the bridge network, and the
rate of fall of triac current is limited only by the stray
inductance in the a.c. circuit. The large value of
commutating di/dt may cause the triac to retrigger due to
commutating dv
(com)
/dt. A small choke in the a.c circuit will
limit the di
(com)
/dt to an acceptable level. An alternative
topology which avoids triac commutation problems is to
control the load on the d.c. side.
Snubber networks
Snubber networks ensure that the device is not exposed to
excessive rates of change of voltage during transient
conditions. This is particularly important when considering
the commutation behaviour of triacs, which has been
discussed elsewhere.
Fig. 13 Triac protection
The followingequations canbe used to calculatethe values
of the snubber components required to keep the reapplied
dv/dt for a triac within the dv
(com)
/dt rating for that device.
The parameters which affect the choice of snubber
components are the value of load inductance, frequency of
the a.c. supply and rms load current. The value of the
snubber resistor needs to be large enough to damp the
circuit andavoidvoltageovershoots. Thesnubber capacitor
should be rated for the full a.c. voltage of the system. The
snubber resistor needs to be rated at 0.5W.
For circuits where the load power factor, cosϕ, ≥ 0.7 the
snubber values are given approximately by:
where: L is the load inductance
f is the supply frequency
I
T(RMS)
is the rms device current
dv
(com)
/dt is the device commutating dv/dt rating.
The presence of a snubber across the device can improve
the turn-on performance of the triac by using the snubber
capacitor discharge current in addition to the load current
to ensure that the triac latches at turn-on. The value of the
snubber resistor must be large enough to limit the peak
capacitor discharge current through the triac to within the
turn-on di/dt limit of the device.
Load
Snubber Varistor
Choke
C ≥ 25L
¸

¸
f I
T(RMS)
dV
(com)
/dt
_

,
2
R ·

3L
C
(9)
502
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Varistor
The use of a metal oxide varistor (MOV), as shown in
Fig. 13, protects the device from transient overvoltages
which may occur due to mains disturbances.
Overcurrent protection
Like all other semiconductor devices, triacs have an infinite
life if they are used within their ratings. However, they
rapidly overheat when passing excessive current because
the thermal capacitance of their junction is small.
Overcurrent protective devices (circuit breakers, fuses)
must, therefore, be fast-acting.
Inrush condition
Motors, incandescent lamp or transformer loads give rise
to an inrush condition. Lamp and motor inrush currents are
avoided by starting the control at a large trigger angle.
Transformer inrush currents are avoided by adjusting the
initial trigger angletoavalueroughlyequal totheloadphase
angle. No damage occurs when the amount of inrush
current is below the inrush current rating curve quoted in
the device data sheet (see the chapter ’Understanding
thyristor and triac data’).
Short-circuit condition
Fuses for protecting triacs should be fast acting, and the
amount of fuse I
2
t to clear the circuit must be less than the
I
2
t rating of the triac. Because the fuses open the circuit
rapidly, they have a current limiting action in the event of a
short-circuit. High voltage fuses exhibit low clearing I
2
t but
the fuse arc voltage may be dangerous unless triacs with
a sufficiently high voltage rating are used.
Conclusions
This paper has outlined the most common uses and
applications of thyristor and triac circuits. The type of circuit
used depends upon the degree of control required and the
nature of the load. Several types of gate circuit and device
protection circuit have been presented. The amount of
device protection required will depend upon the conditions
imposed on the device by the application circuit. The
protection circuits presented here will be suitable for the
majority of applications giving a cheap, efficient overall
design which uses the device to its full capability with
complete protection and confidence.
503
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
6.1.3 The Peak Current Handling Capability of Thyristors
The ability of a thyristor to withstand peak currents many
times the size of its average rating is well known. However,
thereis little informationabout the factors affectingthe peak
current capability. This section will investigate the effect of
pulse duration on the peak current capability of thyristors.
Data sheets for thyristors always quote a figure for the
maximum surge current that the device can survive. This
figureassumes a half sine pulsewith awidthof either 10 ms
or 8.3 ms, which are the conditions applicable for 50/60 Hz
mains operation. This limit is not absolute; narrow pulses
with much higher peaks can be handled without damage
but little information is available to enable the designer to
determine howhigh this current is. This section will discuss
some of the factors affecting a thyristor’s peak current
capability and reviewthe existing prediction methods. It will
go on to present the results of an evaluation of the peak
current handling capabilities for pulses as narrow as 10 µs
for the BT151, BT152 and BT145 thyristors. It will also
propose a method for estimating a thyristor’s peak current
capability for ahalf sine pulsewith adurationbetween10 µs
and 10 ms from its quoted surge rating.
Energy Handling
Inaddition tothe maximumsurge current, datasheets often
quote a figure called "I
2
t for fusing". This number is used to
select appopriate fuses for device protection. I
2
t represents
the energy that can be passed by the device without
damage. In fact it is not the passage of the energy which
causes damage, but the heating of the crystal by the energy
absorbed by the device which causes damage.
If the period over which the energy is delivered is long, the
absorbed energy has time to spread to all areas of the
device capable of storing it - like the edges of the crystal,
the plastic encapsulation, the mounting tab and for very
long times the heatsink - therefore the temperature rise in
the crystal is moderated. If, however, the delivery period is
short - say a single half sine pulse of current with a duration
of <10 ms - the areas to which the energy can spread for
the actual duration of the pulse are limited. This means that
the crystal keeps all the energy giving a much bigger
temperature rise. For very short pulses (<0.1 ms) and large
crystal, the problem is even worse because not all of the
active area of a thyristor crystal is turned on simultaneously
- conduction tends to spread out from the gate area - so
the current pulse passes through only part of the crystal
resulting in a higher level of dissipation and an even more
restricted area for absorbing it.
Expected Results
I
2
t is normally quoted at 10 ms, assuming that the surge is
ahalf sine pulse, andis derived fromthe surge current from:
This calculates the RMS current by dividing by
Under the simplest of analyses I
2
t would be assumed to be
constant so a device’s peak current capability could be
calculated from:
where I
pk
is the peak of a half sine current pulse with a
duration of t
p
. However, experience and experiments have
shown that such an approach is inaccurate. To overcome
this, other ’rules’ have been derived.
One of these ’rules’ suggests that it is not I
2
t which is
constant but I
3
t or I
4
t. Another suggestion is that the
’constancy’ continuously changes from I
2
t to I
4
t as the
pulses become shorter. All these rules are expressed in the
general equation:
whereis Nis either constant or afunction of the pulsewidth,
for example:
The graph shown in Fig. 1 shows what several of these
’rules’ predict would happen to the peak current capability
if they were true. Unfortunately little or no real information
currently exists to indicate the validity of these rules. Tests
have been performed on three groups of devices - BT151,
BT152 and BT145 - to gather the data which would,
hopefully, decide which was correct.
Test Circuit
The technique chosen to measure the peak current
capability of the devices was the stepped surge method. In
this test, the deviceis subjectedto aseries of current pulses
of increasing magnitude until it receives a surge which
causes measurable degradation.
I
2
t ·
¸

¸
I
TSM
√2
_

,
2
0.01
√2 I
TSM
I
pk
· I
TSM
¸

¸
0.01
t
p
_

,
1
2
I
pk
· I
TSM
¸

¸
0.01
t
p
_

,
1
N
N · log
¸

¸
1
t
p
_

,
505
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 1 Predicted I
TSM
multiplying factors
Circuit Description
The circuits used to perform the required measurements
were of the form shown in Fig. 2. They produce half sine
pulses of current from the resonant discharge of C via L.
Triggering of the device under test (DUT) itself is used to
initiate the discharge. The gate signal used for all the tests
was a 100 mA / 1 µs pulse fed from a pulse generator in
single-shot mode.
The magnitude of the current pulse is adjusted by changing
the voltage to which C is initially charged by varying the
output of the PSU. The pulse is monitored by viewing the
voltage across R3 on an digital storage oscilloscope. R1
and D protect the power supply. R1 limits the current from
the supply when DUT fails and during the recharging of C.
D attempts to prevent any high voltage spikes being fed
back into the PSU.
Fig. 2 Surge current test circuit
Pushbutton S1 and resistor R2 are a safety feature. R2
keeps C discharged until S1 is pressed. The trigger pulse
needs a button on the pulse generator to be pressed which
means both hands are occupied and kept away from the
test circuit high voltages.
Choice of L & C
The width of the half sine pulse from an LC circuit is:
and the theoretical peak value of the current is:
These equations assume that the circuit has no series
resistance to damp the resonant action which would result
in a longer but lower pulse. Minimising these effects was
considered to be important so care was taken during the
building of the circuits to keep the resistance to a minimum.
To this end capacitors with low ESR were chosen, the
inductors were wound using heavy gauge wire and the loop
C / L / DUT / R3 was kept as short as possible.
It was decided to test the devices at three different pulse
widths - 10 µs, 100 µs and 1 ms - so three sets of L and C
were needed. The values were selected with the help of a
’spreadsheet’ program running on an PC compatible
computer. The values which were finally chosen are shown
in Table 1. Also given in Table 1 are the theoretical peak
currents that the L / C combination would produce for a
initial voltage on C of 600 V.
Test Procedure
As mentionedearlier, the test methodcalled for eachdevice
to be subjected to a series of current pulses of increasing
amplitude. The resolution with which the current capability
is assessed is defined by the size of each increase in
current. It was decided that steps of approximately 5%
would give reasonable resolution.
Experimentation indicated that the clearest indication of
device damage was obtained by looking for changes in the
off-state breakdown voltage. So after each current pulse
the DUT was removed fromthe test circuit and checked on
a curve tracer. This procedure did slow the testing but it
was felt that it would result in greater accuracy.
Pulse Width C (µF) L (µH) Ipeak (A)
10 µs 13.6 0.75 2564
100 µs 100 10 1885
1 ms 660 154 1244
Table 1. Inductor and Capacitor Values
It was also decided that, since this work was attempting to
determinethe current that adevice couldsurvive- not which
killedit, thefigure actuallyquotedinthe resultsfor adevice’s
current capability would be the value of the pulse prior to
the one which caused damage.
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
10us 100us 1ms 10ms
Width of Half Sine Pulse
I t = const.
I t = const.
I t = const.
I t = const.
log(1/t)
2
3
4
P
e
a
k

C
u
r
r
e
n
t

M
u
l
t
i
p
l
y
i
n
g

F
a
c
t
o
r
t
pulse
· π √ L C
I
peak
· V

C
L
DC PSU
0-600V
D
R1
L
C
R2
DUT
R3
S1
Vak
Trigger
Ia
Pulse
506
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 3 Peak current capability measurements
Test Results
Figure 3 is a graph showing the measured current
capabilities of all of the tested devices. Table 2 summarises
the measurements by giving the mean of the results for the
three device types at each of the pulse widths. Table 3
expresses the mean values as factors of the device I
TSM
rating. This table also gives the factors that the various
’rules’ would have predicted for the various pulse widths.
Mean Peak Current Capability (Amps)
Pulse Width BT151 BT152 BT145
10 µs 912 1092 1333
100 µs 595 1021 1328
1 ms 264 490 697
Table 2. Measured Current Capability
Measured Predicted Factor
Factor (by I
n
t rule)
Pulse BT BT BT n=2 n=3 n=4 n=
Width 151 152 145 log(1/t)
10 µs 9.1 5.5 4.4 31.6 10.0 5.6 4.0
100 µs 6.0 5.1 4.4 10.0 4.6 3.2 3.2
1 ms 2.6 2.4 2.3 3.2 2.2 1.8 2.2
Table 3. Measured and Predicted I
TSM
Multiplication
Factors
Interpretation of Results
It had been hoped that the measurements would give clear
indication of which of the ’rules’ would give the most
accurate prediction of performance. However, an
inspection of Table 3 clearly shows that there is no
correlation between any of the predicted factors and the
measured factors. In fact the variation in the factors
between the various device types would indicated that no
rule based on an I
n
t function alone can give an accurate
prediction. This implies that something else will have to be
taken into account.
Further study of Fig. 3revealsthat thedifference inthepeak
current capability of the threedevice types is becomingless
as the pulses become shorter. This could be explained by
a reduction in the active area of the larger crystals, making
them appear to be smaller than they actually are. This is
consistent with theknown fact that not all areas of athyristor
turn on simultaneously - the conduction region tends to
spread out from the gate. If the pulse duration is less than
the time it takes for all areas of the device to turn on, then
the current flows through only part of the crystal, reducing
the effective size of the device. If the rate at which the
conduction area turns on is constant then the time taken
for a small device to be completely ON is shorter than for
a large device. This would explain why the performance
increase of the BT145 starts falling off before that of the
BT151.
Proposed Prediction Method
The above interpretation leads one to believe that the
original energy handling rule, which says that I
2
t is a
constant, may still be correct but that the performance it
predicts will ’roll off’ if the pulse duration is less than some
critical value. The equation which was developed to have
the necessary characteristics is:
which simplifies to:-
where t
crit
is proportional to - but not necessarily equal to -
the time taken to turn on all the active area of the crystal
and is calculated from:-
where: A = crystal area
R = constant expressing the rate at which the area
is turned on.
Preferably, A should be the area of the cathode but this
information is not always available. As an alternative the
total crystal area can be used if the value of R is adjusted
accordingly. This will inevitably introduce an error because
cathode and crystal areas are not directly proportional, but
it should be relatively small.
##
#
#
#
###
###
##
#
##
#
#
@@
@
@
@
@@@@@
@
@@@
@
*
*
* **
*
***
*
*
*
*
*
**
*
10us 100us 1ms 10ms
100
1000
Width of half-sine pulse
P
e
a
k

C
u
r
r
e
n
t

(
a
m
p
s
)
# BT151 @ BT152 * BT145
I
pk
· I
TSM
¸

¸
0.01
t
p
_

,
1
2
¸

¸
t
p
t
p
+ t
crit
_

,
1
2
I
pk
· I
TSM

¸

¸
0.01
t
p
+ t
crit
_

,
t
crit
·
A
R
507
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Rwasdeterminedempirically tobeapproximately 0.02m
2
/s
Using this value of R gives the values of t
crit
shown in Table
3. Using these values in the above equation predicts that
the peak current handling capability of the BT151, BT152
and BT145 would be as shown in Fig. 4.
Device t
crit
BT151 148 µs
BT152 410 µs
BT145 563 µs
Table 3. Calculated Values of t
crit
Conclusions
The first conclusion that can be drawn fromthis work is that
a thyristor, with average rating of only 7.5A, is capable of
conducting, without damage, a peak current greater than
100times this valueinashort pulse. Furthermore the power
required to trigger the device into conducting this current
can be <1 µW. This capability has always been known and
indeedthe surge rating givenin the datasheet gives avalue
for it at pulse widths of around 10 ms. What has been
missing is a reliable method of predicting what the peak
current capability of a device is for much shorter pulses.
The results obtained using the test methods indicate that
the previously suggested ’rules’ fail to take into account the
effect that crystal size has on the increase in performance.
In this section, an equation has been proposed which takes
crystal size into account by using it to calculate a factor
called t
crit
. This time is then used to ’roll off’ the performance
increase predicted by the original energy handling
equation - I
2
t = constant. This results in what is believed
to be a more accurate means of estimating the capability
of a device for a half sine pulse with a duration between
10 µs and 10 ms.
Fig. 4 Predicted peak current handling using ’Rolled-off
I
2
t’ rule
Width of half-sine pulse
P
e
a
k

C
u
r
r
e
n
t

(
a
m
p
s
)
BT151 BT152 BT145
10us 100us 1ms 10ms
100
1000
508
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
6.1.4 Understanding Thyristor and Triac Data
The importance of reliable and comprehensive data for
power semiconductor devices, together with the
advantages of the absolute maximum rating system, is
clear. This present article describes the data sheet
descriptions of Philips thyristors and triacs, and aims to
enable the circuit designer to use our published data to the
full and to be confident that it truly describes the
performance of the devices.
A brief survey of short-form catalogues is an insufficient
method of comparing different devices. Published ratings
and characteristics require supporting information to truly
describe the capabilities of devices; thus comparisons
between devices whose performance appears to be similar
should not be made on economic grounds alone.
Manufacturers have been known to quote ratings in such
a way as to give a false impression of the capabilities of
their devices.
Ratings and characteristics given in published data should
always be quoted with the conditions to which they apply,
and these conditions should be those likely to occur in
operation. Furthermore, it is important to define the rating
or characteristic being quoted. Only if data is both complete
andunambiguouscan atrue comparison be made between
the capabilities of different types.
Thyristors
Thyristor isagenerictermfor asemiconductor devicewhich
has four semiconductor layers and operates as a switch,
having stable on and off states. A thyristor can have two,
three, or four terminals but common usage has confined
the term thyristor to three terminal devices. Two-terminal
devices are known as switching diodes, and four-terminal
devices are known as silicon controlled switches. The
common, or three-terminal, thyristor is also known as the
reverse blocking triode thyristor or the silicon controlled
rectifier (SCR). Fig. 1 shows the circuit symbol and a
schematic diagramof the thyristor. All Philips thyristors are
p-gate types; that is, the anode is connected to the metal
tab.
The thyristor will conduct a load current in one direction
only, as will a rectifier diode. However, the thyristor will only
conduct this load current when it has been ’triggered’; this
is the essential property of the thyristor.
Fig. 2 shows the static characteristic of the thyristor. When
a small negative voltage is applied to the device, only a
small reverse leakage current flows. As the reverse voltage
is increased, the leakage current increases until avalanche
breakdown occurs. If a positive voltage is applied, then
again a small forward leakage current flows which
increases as the forward voltage increases. When the
forward voltage reaches the breakover voltage V
(BO)
,
turn-onisinitiated by avalanchebreakdown andthe voltage
across the thyristor falls to the on state voltage V
T
.
However, turn-on can occur when the forward
(anode-to-cathode) voltage is less than V
(BO)
if the thyristor
is triggered by injecting a pulse of current into the gate. If
the device is to remain in the on state, this trigger pulse
must remain until the current through the thyristor exceeds
the latching current I
L
. Once the on state is established, the
holding current I
H
is the minimum current that can flow
through the thyristor and still maintain conduction. The load
current must be reduced to below I
H
to turn the thyristor off;
for instance, by reducing the voltage across the thyristor
and load to zero.
Fig. 1 Thyristor circuit symbol and basic structure
Fig. 2 Thyristor static characteristic
Thyristors are normally turned on by triggering with a gate
signal but they can also be turned on by exceeding either
the forward breakover voltage or the permitted rate of rise
Anode Anode
Gate
Gate
Cathode Cathode
p
n
p
n
On-state
characteristic
Off-state
characteristic
Avalanche
breakdown
region
Reverse
characteristic
Reverse
current
Forward
current
Reverse
voltage
Forward
voltage
I
L
I
H
V
(BO)
509
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
of anode voltage dV
D
/dt. However, these alternative
methods of switching to the conducting state should be
avoided by suitable circuit design.
Triacs
The triac, or bidirectional triode thyristor, is a device that
can be used to pass or block current in either direction. It
is therefore an a.c. power control device. It is equivalent to
twothyristors in anti-parallel with a commongate electrode.
However, it only requires one heatsink compared to the two
heatsinks required for the anti-parallel thyristor
configuration. Thus the triac saves both cost and space in
a.c. applications.
Figure 3 shows the triac circuit symbol and a simplified
cross-section of the device. The triac has two main
terminals MT1and MT2 (the load connections) and asingle
gate. The main terminals are connected to both p and n
regions since current can be conducted in both directions.
The gate is similarly connected, since a triac can be
triggered by both negative and positive pulses.
Fig. 3 Triac circuit symbol and basic structure
Fig. 4 Triac static characteristic
The on state voltage/current characteristic of a triac
resembles that of a thyristor. The triac static characteristic
of Fig. 4 shows that the triac is a bidirectional switch. The
condition whenterminal 2of thetriac is positivewith respect
to terminal 1 is denoted in data by the term’T2+’. If the triac
is not triggered, the small leakage current increases as the
voltage increases until the breakover voltage V
(BO)
is
reached and the triac then turns on. As with the thyristor,
however, the triac can be triggered below V
(BO)
by a gate
pulse, provided that the current through the device exceeds
the latching current I
L
before the trigger pulse is removed.
Thetriac, likethethyristor, has holdingcurrent valuesbelow
which conduction cannot be maintained.
When terminal 2 is negative with respect to terminal 1 (T2-)
the blocking and conducting characteristics are similar to
those in the T2+ condition, but the polarities are reversed.
The triac can be triggered in both directions by either
negative (G-) or positive (G+) pulses on the gate, as shown
in Table 1. The actual values of gate trigger current, holding
current and latching current may be slightly different in the
different operating quadrants of the triac due to the internal
structure of the device.
Quadrant Polarity of T2 wrt T1 Gate polarity
1 (1+) T2+ G+
2 (1-) T2+ G-
3 (3-) T2- G-
4 (3+) T2- G+
Table 1. Operating quadrants for triacs
Device data
Anode to cathode voltage ratings
The voltage of the a.c. mains is usually regarded as a
smooth sinewave. In practice, however, there is a variety
of transients, some occurring regularly and others only
occasionally (Fig. 5). Although some transients may be
removed by filters, thyristors must still handle anode to
cathode voltages in excess of the nominal mains value.
The following reverse off-state voltage ratings are given in
our published data:
V
RSM
: the non-repetitive peak reverse voltage. This is the
allowable peak value of non-repetitive voltage transients,
and is quoted with the maximum duration of transient that
can be handled (usually t < 10ms).
V
RRM
: the repetitive peak reverse voltage. This is the
allowable peak value of transients occurring every cycle.
V
RWM
: the peak working reverse voltage. This is the
maximum continuous peak voltage rating in the reverse
direction, neglecting transients. It corresponds to the peak
negative value (often with a safety factor) of the sinusoidal
supply voltage.
MT1
MT2
Gate
Gate
MT1
MT2
n
n
n
n
p
p
Reverse
current
Forward
current
Reverse
voltage
Forward
voltage
I
H
I
L
V
(BO)
Blocking
Blocking
I
L
I
H
V
(BO)
T2+
T2-
510
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 5 Diagrammatic voltage waveform showing device
anode voltage ratings
The forward off-state voltages corresponding to V
RSM
, V
RRM
and V
RWM
are listed below.
V
DSM
: the non-repetitive peak off-state voltage applied in
the forward direction.
V
DRM
: the repetitive peak off-state voltage applied in the
forward direction.
V
DWM
: the peak working off-state voltage applied in the
forward direction.
Both the repetitive and non-repetitive voltage ratings are
determined partly by the voltage limit that prevents the
thyristor being driven into forward or reverse breakdown,
and partly by the instantaneous energy (resulting from an
increase in leakage current) that can be dissipated in the
device without exceeding the rated junction temperature.
Whenathyristor istooperatedirectlyfromthemains supply,
it is advisable to choose a device whose repetitive peak
voltageratings V
RRM
andV
DRM
are at least 1.5times the peak
value of the sinusoidal supply voltage. This figure forms
part of the device type number; for example BT151-650R,
where 650 corresponds to V
DRM
, V
RRM
=650V and the final
R (for Reverse) indicates that the anode of the device is
connected to the metal tab.
Anode-to-cathode current ratings
The following current ratings, described by the waveforms
shown in Fig. 6, are given in our published data. Note that
the suffix
T
implies that the thyristor is in the on state.
I
T(AV)
: the average value of the idealised mains current
waveformtaken over one cycle, assuming conduction over
180˚. For devices mounted on heatsinks, the I
T(AV)
rating
should be quoted for a particular mounting-base
temperature T
mb
; our devices are generally characterised
at a mounting-base temperature of at least 85˚C. A device
can have an artificially high current rating if the
mounting-base temperature is unrealistically low; ratings
with no associated mounting-base temperature should be
regarded with suspicion.
I
T(RMS)
: the rms on-state current. This rating gives the
maximum rms current that the thyristor can handle. It is
important for applications when the device current
waveformis describedby ahigh value formfactor. For such
conditions the rms current rather than the average current
may be the limiting rating.
I
TRM
: the repetitive peak forward current. This rating is the
peak current that can be drawn each cycle providing that
the average and rms current ratings are not exceeded.
I
TSM
: the non-repetitive (surge) peak forward current. This
rating is the peak permitted value of non-repetitive
transients, and depends on the duration of the surge. Our
published data quotes the I
TSM
rating for t=10ms, the
duration of a half-cycle of 50Hz mains. However, some
manufacturers quote I
TSM
for t=8.3ms (half-cycle of 60Hz
mains), and thus surge ratings for devices quoted at
t=8.3ms should be approximately downrated (multiplied by
0.83) before comparing them with t=10ms surge ratings.
The surge rating also depends on the conditions under
which it occurs. Our data sheets quote I
TSM
rating under the
worst probable conditions, that is, T
j
=T
j(max)
immediately
prior to the surge, followed by reapplied V
RWM(max)
immediately after the surge. An unrealistically high I
TSM
rating could be quoted if, for example, T
j
<T
j(max)
prior to the
surge and then the full rated voltage is not reapplied.
Published data also includes curves for I
TSM
against time
which showthe maximumallowable rms current which can
occur during inrush or start-up conditions. The duration of
the inrush transient and the mounting base temperature
prior to operation determine the maximum allowable rms
inrush current.
Fig. 6 Diagrammatic current waveform showing device
anode current ratings
V
D
V
DSM
V
DRM
V
DWM
V
RWM
V
RRM
V
RSM
V
R
Time
Mains
waveform
I
T
I
TSM
I
TRM
I
T(RMS)
I
T(AV)
Time
511
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
dI/dt: the rate of rise of on-state current permissible after
triggering. An excessive rate of rise of current causes local
heating and thus damage to the device. The rate of rise of
current is determined by both the supply and load
impedances, and can be limited by additional series
inductance in the circuit.
Fig. 7 Non-repetitive surge current as a function of time
I
2
t: a dimensional convenience specifying the capability of
a thyristor to absorb energy. This rating is required for the
selection of fuses to protect the thyristor against excessive
currents caused by fault conditions. It is normally only valid
over the range 3 to 10ms. In our published data, a value is
quoted for 10ms, in which case:
The user should match the minimum I
2
t capability of the
thyristor to the worst case I
2
t let-through of a range of
nominally rated fuses in order to select a fuse that will
protect the device under worst probable conditions.
Values of I
2
t other than those quoted for 10ms can be
estimated by referring to the appropriate published curves
of non-repetitive surge current against time. For example,
Fig. 7 is the non repetitive surge current curve for a thyristor
whose I
2
t at 10ms is 800A
2
s. From Fig. 7, I
TS(RMS)
at 3ms is
470A and therefore I
2
t at 3ms is given by:
To summarise, when selecting an appropriate fuse the
following conditions must be taken into account.
1. The fuse must have an rms current rating equal to, or
less than, that of the thyristor it is to protect.
2. The I
2
t at the rms working voltage must be less than
that of the thyristor taken over the fuse operating time.
3. The arc voltage of the fuse must be less than the V
RSM
rating of the thyristor.
Gate-to-cathode ratings
The following gate-to-cathode ratings are given in the
published data.
V
RGM
: the gate peak reverse voltage.
P
G(AV)
: the mean gate power, averaged over a 20ms period.
P
GM
: the peak gate power dissipation.
The gate-to-cathodepower ratings should not be exceeded
if over-heatingof the gate-cathodejunctionis tobeavoided.
Temperature ratings
Two temperature ratings are given in the published data.
T
stg
: the storagetemperature. Bothmaximumandminimum
values of the temperature at which a device can be stored
are given.
T
j
: the junction temperature. This is one of the principal
semiconductor ratings since it limits the maximum power
that a device can handle. The junction temperature rating
quoted in our published data is the highest value of junction
temperature at which the device may be continuously
operated to ensure a long life.
Thermal characteristics
The following thermal resistances and impedances are
given in our data.
R
th(j-a)
: the thermal resistance between the junction of the
device and ambient (assumed to be the surrounding air).
R
th(j-mb)
: the thermal resistance between the junction and
mounting base of the device.
R
th(mb-h)
: the thermal resistance between the mounting base
of the device and the heatsink (contact thermal resistance).
Z
th(j-mb)
: the transient thermal impedance between the
junction and mounting-base of the device. The value given
in the published data is for non-repetitive conditions and a
particular pulse duration. Under pulse conditions, thermal
impedances rather than thermal resistances should be
considered. Higher peak power dissipation is permitted
under pulse conditions since the materials in a thyristor
have a definite thermal capacity, and thus the critical
junction temperature will not be reached instantaneously,
even when excessive power is being dissipated in the
device. The published data also contains graphs of Z
th(j-mb)
against time (for non-repetitive conditions) such as those
shown in Fig. 8.
0.001 0.003 0.01 0.03 0.1 0.3 1 3 10
0
100
200
300
400
500
600
Duration (s)
ITS(RMS)
I
2
t ·


i
2
.dt (1)
·
¸

¸
I
TSM

2
_

,
2
× 10.10
−3
(A
2
s)
I
2
t (3ms) · I
TS(RMS)
2
× t
· 470
2
× 3.10
−3
· 662.7A
2
s
512
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 8 Thermal impedance between the junction and
mounting-base as a function of time
The values of the various thermal resistances between the
thyristor junction and the surroundings must be considered
to ensure that the junction temperature rating is not
exceeded. The heat generated in a semiconductor chip
flows by various paths to the surroundings. Fig. 9 shows
the various thermal resistances to be taken into account in
this process. With no heatsink, the thermal resistance from
the mounting-base to the surroundings is given by R
th(mb-a)
.
When a heatsink is used, the heat loss direct to the
surroundings from the mounting-base is negligible owing
to the relatively high value of R
th(mb-a)
and thus:
Fig. 9 Heat flow paths
R
th(mb-a)
= R
th(mb-h)
+ R
th(h-a)
(2)
Where appropriate, our published data contains power
graphs such as that in Fig. 10. These characteristics relate
the total power P dissipated in the thyristor, the average
forward current I
T(AV)
, the ambient temperature T
a
, and the
thermal resistance R
th(mb-a)
, with the form factor, a, as a
parameter. They enable the designer to work out the
required mounting arrangement from the conditions under
which the thyristor is to be operated.
1E-05 0.0001 0.001 0.01 0.1 1 10
0.001
0.003
0.01
0.03
0.1
0.3
1
3
10
Time (s)
Z th(j-mb) (K/W)
T
j
T
mb
T
h
T
a
R
th(j-mb)
R’
th(mb-a)
R
th(mb-h)
R
th(h-a)
Fig. 10 Derivation of the appropriate R
th(mb-a)
for a given value of I
T(AV)
, a and T
amb
0 5 10 15 20
0
5
10
15
20
25
30
35
40
45
50
0 20 40 60 80 100 120
125
120
115
110
105
100
95
90
85
80
75
a=4
a=2.8 2.2 1.9 1.6
R
th(mb-a)
=0.5K/W 1K/W 2K/W
3
4
6
10
P(W) T
mb
(
o
C)
I
T(AV)
(A) T
a
(
o
C)
513
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Usually, the characteristics are designed for use in 50Hz
sinusoidal applications, when the procedure below should
be followed.
1. Determine the values of I
T(AV)
and I
T(RMS)
for the relevant
application.
2. Determine the form factor, which is given by:
3. Starting from the appropriate value of I
T(AV)
on a curve
such as Fig. 10, move vertically upwards to intersect
the appropriate form factor curve (interpolating if
necessary).
4. This intersection gives the power dissipated in the
thyristor on the left-hand axis of the combined graph
and the mounting base temperature on the right hand
axis.
5. Moving horizontally across from this intersection to the
appropriate value of ambient temperature gives the
required mounting base to ambient thermal resistance
R
th(mb-a)
.
6. The required heatsink thermal resistance R
th(h-a)
can
now be calculated from Equation 2 since the mounting
base to heatsink thermal resistance R
th(mb-h)
is given in
the published data.
Example
The thyristor to which Fig. 10 applies is operated at an
average forward current I
T(AV)
of 12A and an rms forward
current I
T(RMS)
of 19.2A. The maximum anticipated ambient
temperature is 25˚C. Now, Equation 3 gives,
Figure 10 gives the power as P=20W and the
mounting-base temperature as T
mb
=105˚C. Also, at this
power and ambient temperature of 25˚C, Fig. 10 gives the
value of R
th(mb-a)
to be 4˚C/W. The published data gives the
valueof R
th(mb-h)
(using aheatsink compound) to be 0.2˚C/W
and then Equation 2 gives
Mounting torque
Two values of mounting torque are given in the published
data. A minimum value is quoted below which the contact
thermal resistance rises owing to poor contact, and a
maximum value is given above which the contact thermal
resistance again rises owing to deformation of the tab or
cracking of the crystal.
Fig. 11 Forward current vs. forward voltage
The surface of a device case and heatsink cannot be
perfectly flat, and thus contact will take place on several
points only, with a small air-gap over the rest of the contact
area. The use of a soft substance to fill this gap will lower
the contact thermal resistance. We recommend the use of
proprietary heatsinking compounds which consist of a
silicone grease loaded with an electrically insulating and
good thermal conducting powder such as alumina.
Anode-to-cathode characteristics
The following anode-to-cathode characteristics are
included in the published data.
I
R
: the reverse current. This parameter is given for the worst
probable conditions; that is, the reverse voltage
V
R
=V
RWM(max)
and a high T
j
.
I
D
: the off-state current. This parameter is again given for
the worst probable conditions; that is, the forward voltage
V
D
=V
DWM(max)
and a high T
j
.
I
L
: the latching current (Fig. 2). This parameter is quoted at
a particular value of junction temperature.
I
H
: the holding current (Fig. 2). This parameter is quoted at
a particular value of junction temperature.
V
T
: the forward voltage when the thyristor is conducting.
This parameter is measured at particular values of forward
current andjunction temperature. The junction temperature
is usually low (T
j
=25˚C, for example) since this is the worst
case. The measurement must be performed under pulse
conditions to maintain the low junction temperature. The
published data also contains curves of forward current
against forward voltage, usually for two values of the
junction temperature: 25˚C and T
j(max)
(Fig. 11).
0 0.2 0.4 0.6 0.8 1.0
0
1.0
2.0
3.0
IF(A)
VF(V)
MAX
TYP
25 C
150 C
a ·
I
T(RMS)
I
T(AV)
(3)
a ·
19.2
12
· 1.6
R
th(h − a)
· 4 − 0.2 · 3.8 °C/W
514
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 12 Definition of rate of rise of off-state voltage
dV
D
/dt
dV/dt: the rate of rise of off-state voltage that will not trigger
any device. This parameter is given at maximum values of
junction temperature T
j(max)
and forward voltage
V
D
=V
DRM(max)
.
The values of dV
D
/dt quoted in our published data are
normally specified assuming an exponential waveform.
This facilitates the design of RC snubber circuits for device
protection when required. Fig. 12 illustrates the definition
of dV
D
/dt. The final voltage applied to the device V
DM
is
chosenasV
DRM(max)
andthejunctiontemperatureisT
j
=T
j(max)
.
Fig. 12 shows that dV
D
/dt is given by the expression:
where T is the exponential time constant.
The dV
D
/dt capability of a thyristor increases as the junction
temperature decreases. Thus curves such as those shown
in Fig. 13a) are provided in the published data so that
designers can uprate devices operated at lower junction
temperatures.
The dV
D
/dt characteristic can also be increased by
operating the device at a low supply voltage. Thus the
published data also contains curves such as Fig. 13b)
which shows how dV
D
/dt increases as the ratio V
DM
/V
DRM
max decreases. Note that V
DM
is unlikely to be greater than
2
/
3
V
DRM(max)
(usually owing to the restriction of V
DWM(max)
) and
therefore the fact that dV
D
/dt approaches zero as V
DM
increases above the value of
2
/
3
V
DRM(max)
does not cause
problems.
dV
D
dt
V
DM
0.63V
DM
T 2T 3T 4T Time
dV
D
dt
·
0.63V
DM
T
·
0.63 × 2/3V
DRM(max)
T
·
0.42V
DRM(max)
T
(V/µs)
a) Junction temperature, T
j
b) Applied voltage, V
D
Fig. 13 Derating of maximum rate of rise of off-state voltage
0 20 40 60 80 100
0
200
400
600
800
1,000
1,200
1,400
Exp
l
dV
D
dt
Exp
l
dV
D
dt
Rating
point
Rating
point
T
j
(
o
C)
V
D
/V
DRM(max)
(%)
20 40 60 80 100 120 140
0
250
500
750
1,000
1,250
1,500
1,750
2,000
515
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
a) Minimum V
GT
that will trigger all devices b) Minimum I
GT
that will trigger all devices
Fig. 14 Gate characteristics vs. junction temperature
-50 0 50 100 150
Tj ( C)
VGT (V)
3
2
1
0
0
50
100
150
0 50 100 150 -50
IGT (mA)
Tj ( C) Tj ( C)
Gate-to-cathode characteristics
The following gate-to-cathode characteristics are given in
the published data.
V
GT
: the gate-to-cathode voltagethat will trigger all devices.
This characteristic should be quoted for particular values
of applied voltage V
D
and low junction temperature.
I
GT
: the gate-to-cathode current that will trigger all devices.
This characteristic should be quoted for the same
conditions given above.
A gate drive circuit must be designed which is capable of
supplyingat least the requiredminimumvoltageandcurrent
without exceeding the maximum power rating of the gate
junction. Curves such as those shown in Fig. 14 (which
relate the minimumvalues of V
GT
and I
GT
for safe triggering
to the junction temperature) are provided in data. The
following design procedure is recommended to construct a
gate drive circuit load-line on the power curves shown in
Fig. 15.
1. Determine the maximum average gate power
dissipation P
G(AV)
from the published data (normally
0.5W, 1.0W, or 2.0W) and then use the appropriate
choice of x-axis scaling in Fig. 15.
2. Estimate the minimum ambient temperature at which
the device will operate, and then determine the
minimum values of V
GT
and I
GT
from curves such as
Figs. 14a) and 14b) in the published data. Note that it
is assumed that at switch-on T
j
=T
a
.
3. Determine the minimum open-circuit voltage of the
trigger pulse drive circuit: this is the first co-ordinate on
the load line at I
G
=0.
4. Using the appropriate horizontal scaling for the device
(P
G(AV)
=0.5W, 1.0Wor 2.0W), plot a second point on the
power curve whose co-ordinates are given by V
GT(min)
and 5×I
GT(min)
. Construct a load line between these two
points. The slope of this load gives the maximum
allowable source resistance for the drive circuit.
5. Check the power dissipation by ensuring that the load
line must not intersect the curve for the maximum peak
gatepower P
GM(max)
whichis the outermost (δ=0.1) curve
of Fig. 15. Theloadlinemust alsonot intersect thecurve
which represents the maximum average gate power
P
G(AV)
modified by the pulse mark-space ratio, where:
For instance, in Fig. 15, for a thyristor with P
G(AV)
=1W,
the δ=0.25 curve can be used for a gate drive with a 1:3
mark-space ratio giving an allowable maximum gate
power dissipation of P
GM(max)
=4W.
An illustration of howthe above design procedure operates
to give an acceptable gate drive circuit is presented in the
following example.
Example
A thyristor has the V
GT
/T
j
and I
GT
/T
j
characteristics shown
in Fig. 14 and is rated with P
G(AV)
=0.5W and P
GM(max)
=5W. A
suitable trigger circuit operating with δ
max
=0.25,
V
GT(min)
=4.5V, I
GT(max)
=620mA and T
a(min)
=-10˚C is to be
designed. Determine its suitability for this device.
P
GM(max)
·
P
AV
δ
(5)
516
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 15 Gate circuit design procedure - power curves
0
2.5
5.0
7.5
10.0
12.5
15.0
17.5
20.0
22.5
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 3.2 3.6 4.0
= 0.100
= 0.111
= 0.125
= 0.143
= 0.167
= 0.200
= 0.250
= 0.333
= 0.500
= 1.000
Gate current, I
G
(A)
Gate voltage, V
G
(V)
(P
G(AV)
=0.5W)
(P
G(AV)
=1.0W)
(P
G(AV)
=2.0W)
A
B
C
1. Select the top x-axis scale of Fig. 15 (P
G(AV)
=0.5W).
2. From Fig. 14, V
GT(min)
=1.75V, and IGT(min)=66mA.
3. At minimum supply voltage, the open-circuit gate
voltage is 4.5V, giving point ’A’ in Fig. 15. Point B is
plotted at the co-ordinates V
GT(min)
and 5xI
GT(min)
, that is
at 1.75V and 330mA, and load line ABC is constructed
as shown. Note that point C is the maximum current
required at I
G
=570mA and is within the capability of the
drive circuit.
4. As required the load line does not intersect the P
G(max)
(δ=0.1). The gate drive duty cycle, δ, is 0.25. Therefore
P
GM(max)
= P
G(AV)
/δ = 0.5/0.25 = 2W. As required, the load
line ABC does not intersect the δ=0.25 curve.
Switching characteristics
Two important switching characteristics are usually
includedin our publisheddata. Theyare the gate-controlled
turn-on time t
gt
(divided into a turn-on delay time, t
d
, and a
rise time, t
r
) and the circuit-commutated turn-off time, t
q
.
Gate-controlled turn-on time, t
gt
Anode current does not commence flowing in the thyristor
at the instant that the gate current is applied. There is a
period which elapses between the application of the trigger
pulse and the onset of the anode current which is known
as the delay time t
d
(Fig. 16). The time taken for the anode
voltage to fall from 90% to 10% of its initial value is known
as the rise time t
r
. The sum of the delay time and the rise
time is known as the gate-controlled turn-on time t
gt
.
The gate controlled turn-on time depends on the conditions
under which it is measured, and thus the following
conditions should be specified in the published data.
-Off-state voltage; usually V
D
=V
DWM(max)
.
-On-state current.
-Gate trigger current; high gate currents reduce t
gt
.
-Rate of rise of gate current; high values reduce t
gt
.
-Junction temperature; high temperatures reduce t
gt
.
Circuit-commutated turn-off time
When a thyristor has been conducting and is
reverse-biased, it does not immediately go into the forward
blocking state: minority charge carriers have to be cleared
away by recombination and diffusion processes before the
device can block reapplied off-state voltage. The time from
the instant that the anode current passes through zero to
the instant that the thyristor is capable of blocking reapplied
off-state voltage is the circuit-commutated turn-off time t
q
(Fig. 17).
517
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 16 Thyristor gate-controlled turn-on characteristics
Fig. 17 Thyristor turn-off characteristics
The following conditions should be given when t
q
is quoted.
-On-state current; high currents increase t
q
.
-Reverse voltage; low voltages increase t
q
.
-Rate of fall of anode current; high rates increase t
q
.
-Rate of rise of reapplied off-state voltage; high rates
increase t
q
.
-Junction temperature; high temperatures increase t
q
.
-Gate bias; negative voltages decrease t
q
.
Triac ratings
The ratings and characteristics of the triac are similar to
those of the thyristor, except that the triac does not have
any reverse voltage ratings (a reverse voltage in one
quadrant is the forward voltage in the opposite quadrant).
However, one characteristic requires special attention
when choosing triacs; the rate of re-applied voltage that the
triac will withstand without uncontrolled turn-on.
If a triac is turned off by simply rapidly reversing the supply
voltage, the recovery current in the device would simply
switchit onintheoppositedirection. Toguaranteereduction
of the current below its holding value, the supply voltage
must be reduced to zero and held there for a sufficient time
to allow the recombination of any stored charge in the
device. To ensure turn-off, the rate of fall of current during
the commutation interval (turn-off period) and the rate of
rise of re-applied voltage after commutation must both be
restricted. Anexcessive rate of fall of current createsalarge
number of residual charge carriers which are then available
to initiate turn-on when the voltage across the triac rises.
With supply frequencies up to around 400Hz and a
sinusoidal waveform, commutation does not present any
problemswhenthe loadis purely resistive, sincethe current
and voltage are in phase. As shown in Fig. 18 the rate of
fall of on-state current -dI/dt, given by Equation 6, and the
rate of rise of commutating voltage dV
com
/dt, given by
equation 7, are sufficiently low to allow the stored charge
in the device to fully recombine. The triac is thus easily able
to block the rising reapplied voltage dV
com
/dt.
(6)
(7)
Fig. 18 Triac commutation waveforms (resistive load)
Fig. 19 Triac commutation waveforms (inductive load)
90%
10%
10%
V
D
I
GT
I
T
t
r
t
d
t
gt
dI/dt · 2πf .

2I
T(RMS)
t
q
I
T
I
R
V
D
V
R
dI
T
dt
dV
D
dt
dV
com
/dt · 2πf .

2V
(RMS)
V
DWM
-dI/dt
dV
com
/dt
Time
Time
Time
Supply
voltage
Load
current
Voltage
across
triac
Trigger
pulses
Current
V
T
V
DWM
-dI/dt
dV
com
/dt
Time
Time
Time
Supply
voltage
Load
current
Voltage
across
triac
Trigger
pulses
Current
518
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 20 Rate of rise of commutating voltage with rate of fall of on-state current and temperature
20 40 60 80 100 120 0
1
10
100
1000
dV/dt (V/us)
5.1 3.9 3.0 2.3 1.8 1.4 -dI/dt=
dVD/dt limit
Tj (C)
However, with an inductive load (Fig. 19) the current lags
behind the voltage and consequently commutation can
present special difficulties. When the on-state current has
fallen to zero after a triac has been conducting in one
direction the supply voltage in the opposite direction will
have already reached a significant value. The rate of fall of
triac current will still be given by Equation 6 but the rate of
rise of reapplied voltage, dV
com
/dt will be very large. The
triac may switch on immediately unless dV/dt is held less
than that quoted in the published data by suitable circuit
design. Alternatively, the circuit design can remain simple
if Hi-Com triacs are employed instead. Sections 6.3.1 and
6.3.2 explainthe advantages of using Hi-Comtriacs in such
inductive circuits.
The maximum rate of rise of commutating voltage which
will not cause the device totrigger spuriously is an essential
part of the triac published data. However, dV
com
/dt is
meaninglessunless theconditions whichare applicableare
provided, particularly the rate of fall of on-state current
-dI
T
/dt. Our published data also contains graphs such as
Fig. 20 which relate dV
com
/dt to junction temperature with
-dI
T
/dt as a parameter. The characteristic dV
com
/dt is
specified under the worst probable conditions, namely:
-mounting base temperature, T
mb
=T
mb(max)
-reapplied off-state voltage, V
D
=V
DWM(max)
-rms current, I
T(RMS)
= I
T(RMS)(max)
.
In order that designers may economise their circuits as far
as possible, we offer device selections with the same
current ratings but with different values of dV
com
/dt (at the
same value of -dI
T
/dt) for some of our triac families. The
dV/dt capability canbetradedoff against the gatesensitivity
(I
GT(max)
) of the device. Sensitive gate triacs (i.e. those which
require only a small amount of gate current to trigger the
device) have less ability to withstand high values of dV
com
/dt
before sufficient current flows within the device to initiate
turn-on. These different device selections are differentiated
by suffices which are added to the device type number eg.
BT137-600F.
Detailed design considerations for dV
com
/dt limiting in
inductive circuits when using triacs are considered in
separate articles in this handbook.
519
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Thyristor and Triac Applications
521
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
6.2.1 Triac Control of DC Inductive Loads
The problem of inductive loads
This publication investigates the commutation problem
encountered when triacs are used in phase control circuits
with inductive loads. Commutation failure is likely to occur
owing to circuit inductance imposing a sudden rise of
voltage on the triac after conduction. Control of
transformers supplying an inductively loaded bridge
rectifier is particularly troublesome because of the added
effect of rapid current decay during commutation. For a
better understanding of the nature of the problem, the
commutation behaviour is summarised here.
Triacs are bipolar power control elements that may turn on
with either polarity of voltage applied between their main
terminals. Unlike thyristors there is no circuit-imposed
turn-off time. To ensure commutation the decay rate of
current before turn-off and the rate of rise of reapplied
voltage must both be held below specified limits. An
excessive current decay rate has a profound effect on the
maximum rate of rise of voltage that can be sustained, as
then a large amount of stored charge is available to initiate
the turn-on in the next half cycle.
Figure 1 shows the condition for a triac controlled
transformer followed by a rectifier with inductive load. The
load inductance forces the rectifier diodes into conduction
whenever the instantaneous dc output voltage drops to
zero. The transformer secondary is thus shorted for some
time after the zero transitions of the mains voltage and a
reversevoltage is appliedtothe triac, turningit off. Because
of transformer leakage inductance the triac does not turn
off immediately but continues to conduct over what is called
the commutation interval (see Fig. 1).
During the commutation interval a high rate of decay of
current (dI
com
/dt) results for two reasons. Firstly the rate of
fall of current is high because the leakage inductance of
most transformers is low. This is necessary to achieve a
small dc output voltage loss (represented by the shaded
areas in the voltage waveform of Fig. 1) in the transformer.
Secondly, with an inductive rectifier load a substantial
current flows when commutation starts to occur.
The large value of dI
com
/dt results in a high rate of rise of
voltage, dv/dt. Since the current decays rapidly the peak
reverse recovery current I
RRM
is fairly large. Upon turn-off,
I
RRM
is abruptly transferred to the snubber elements R and
Csothevoltageabruptlyrisestothelevel R.I
RRM
(Cisinitially
discharged). Owing to the high value of both dI
com
/dt and
dv/dt, loss of control follows unless measures are taken to
prevent it.
Fig. 1 Triac control of transformer suppling rectifier with
an inductive load (α = trigger angle)
Obtaining reliable commutation
A saturable choke in series with the transformer primary
proves effective in achieving reliable commutation (Fig. 2).
Saturation should occur at a fraction of the rated load
current so that the loss in the rectifier output voltage is
minimised. At lowcurrents thetotal inductance islarge, thus
softening the commutation and eliminating transients. The
choke delays the rise in voltage so a quiescent period of a
few tens of microseconds is introduced, during which time
523
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
the triac can recover. There is usually no difficulty in
designing a choke such that the decay rate of current
(dI
com
/dt) andtherate of riseof voltage(dv/dt) aresufficiently
reduced to ensure reliable control.
Fig. 2 Use of a saturable choke to ensure commutation
Circuit analysis
Over the commutation interval the transformer secondary
is shorted as the load inductance keeps the rectifier diodes
in conduction, so the simplified diagram of Fig. 3 applies.
If the load time constant is much larger than the mains
period then the load current can be assumed to be purely
dc. The waveforms of triac voltage and current are given in
Fig. 4. The mains voltage is given by v
i
= Vsinωt. As the
commutation interval is a fraction of the ac period then the
rate of change of voltage during the commutation interval
can be assumed to be linear, giving:
Over the period 0 to t
2
the voltage across the saturable
choke L
s
and leakage inductance L
leak
is equal to v
i
(assuming the triac on-state voltage to be negligible).
Assuming for this analysis that L
s
remains in saturation
(dashed portion in i
t
waveform) then if L
sat
is the saturated
inductance, the following expression can be derived:
where di/dt is the rate of change of triac current.
Fig. 3 Equivalent circuit diagram
Fig. 4 Commutation voltage and current waveforms
(Dashed portion of I
t
shows current waveform if L
s
remains saturated. Period t1 to t2 is shown expanded)
Integrating equation (2) gives:
where I
t
is the current prior to commutation.
(L
leak
+ L
sat
).di /dt · −
ˆ
Vωt (2)
i
t
· I
t

ˆ
Vωt
2
2(L
leak
+ L
sat
)
(3)
v
i
· −
ˆ
Vωt (1)
524
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
At time t
1
, current i
t
passes through zero, so, from (3):
At t
1
the mains voltage has attained the value V
1
which is
found by combining equations (1) and (4) to give:
Choke L
s
comes out of saturation at low current levels so
the triac turn-off point is delayed to time t
2
. Since in a
practical circuit the delay is only of the order of 50µs, the
mains voltage V
2
at the instant of turn-off is very nearly
equal to V
1
. Thus from equation 5:
The triac conducts until time t
2
. Denoting the value of
unsaturated inductance as L
unsat
, the current decay rate at
zero current is given by:
The initial rate of rise of off-state voltage, dv
com
/dt, can now
be derived. This parameter is decisive for the behaviour of
the triac, since a much greater dv/dt can be sustained after
carrier recombination, that is, when the off-state voltage
has reached a substantial value.
At time t
2
the triac turns off but the voltage across it is still
zero. The voltage drop across L
s
and L
leak
is equal to V
2
and
the rate of rise of current carried by these inductances,
di
L
/dt, is givenin equation(7). The rate of riseof triac voltage
dv/dt is determined by di
L
/dt and the values of the snubber
components R and C.
When the interval t
1
to t
2
is long enough, the triac has fully
recovered at time t
2
, and so the current i to be taken over
by the parallel RC snubber network is zero. At time t
2
, dv/dt
is equal to the initial rate of rise of voltage dv
0
/dt. From
equations (7) and (8):
In circuits where no transformer is interposed between the
triac and rectifier, some series inductance is still needed to
restrict turn-on di/dt. In that case Equations (7) and (9) are
still valid by omitting L
leak
.
Example - DC motor load
The motor control circuit of Fig. 5 illustrates the use of the
design method proposed in the previous section. Since the
motor has a fairly high inductance it may be considered as
a constant current source, giving a severe test condition for
triac commutation.
Fig. 5 DC Motor test circuit.
Choke L
unsat
=2.25mH, 30 turns on 36x23x10mm
3
toriod core
Transformer 220V/150V, 6kVA, 0.9mH leakage
inductance
Motor Series wound DC motor, Leakage
inductance = 30mH
With L
leak
= 0.9mH, L
unsat
= 2.25mH and L
sat
<<L
leak
the circuit
conditions can be calculated for a triac current of I
t
= 20A
anda 220V, 50Hz supply. Using equations (7) and (9) gives
di
c
/dt = -18.3A/ms and dv
0
/dt = 0.6V/µs. These values can
be compared with the commutation limits of the device to
ensure that reliable commutation can be expected.
The inductance in the ac circuit also restricts turn-on di/dt
which, for a continuous dc load current is:
where v
i
is the instantaneous ac input voltage, t
on
is the
turn-on time of the triac and R is the snubber resistance.
Maximum turn-on di/dt occurs at the peak value of input
voltage, v
i
. The initial rise of on-state current depends on
the snubber discharge current through R as well as the
limiting effect of the circuit inductance.
t
1
·

2I
t
(L
leak
+ L
sat
)
ˆ

(4)
V
1
· −


ˆ
VI
t
(L
leak
+ L
sat
) (5)
V
2
≈ −


ˆ
VI
t
(L
leak
+ L
sat
) (6)
di
c
dt
·
V
2
(L
leak
+ L
unsat
)
· −


ˆ
VI
t
(L
leak
+ L
sat
)
L
leak
+ L
unsat
(7)
dv
dt
· R.
di
L
dt
+
i
C
(8)
di
on
dt

v
i
t
on
R
+
v
i
L
leak
+ L
unsat
(10)
dv
0
dt
·
R
L
leak
+ L
unsat


ˆ
VI
t
(L
leak
+ L
sat
) (9)
525
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
The oscillograms of Figs. 6 to 10 illustrate circuit
performance. With no choke added a large dv/dt was
observed(Figs. 6and7) andso consequently commutation
failed when motor current was increased to around 9A. As
seen from Figs. 8 to 10 the choke softens commutation so
that dependable control results even at 23A motor current.
At this current (Fig. 10) the quiescent interval is about 30µs,
which is adequate time for the triac to recover.
Fig. 6 Triac voltage and current. No series choke.
7A motor current. Timebase: 2ms/div
Upper trace: Triac voltage, v
t
(100V/div)
Lower trace: Triac current, i
t
(5A/div)
Fig. 7 Triac voltage and current. No series choke.
7A motor current. N.B. Snap-off current
Timebase: 100µs/div
Upper trace: Triac voltage, v
t
(20V/div)
Lower trace: Triac current, i
t
(1A/div)
Fig. 8 Triac voltage and current. Series choke added.
7A motor current. Timebase: 100µs/div
Upper trace: Triac voltage, v
t
(20V/div)
Lower trace: Triac current, i
t
(1A/div)
Fig. 9 Triac voltage and current. Series choke added.
23A motor current. Timebase: 100µs/div
Upper trace: Triac voltage, v
t
(10V/div)
Lower trace: Triac current, i
t
(5A/div)
Fig. 10 Triac voltage and current.Series choke added
23A motor current. N.B. slight reverse recovery current
Timebase: 50µs/div
Upper trace: Triac voltage, v
t
(10V/div)
Lower trace: Triac current, i
t
(1A/div)
526
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
6.2.2 Domestic Power Control with Triacs and Thyristors
The increasing demand for more sophisticated domestic
products can, in part, be met by providing the user with
some form of electronic power control. This control can be
used, for example, to adjust the suction of a vacuum
cleaner, the brightness of roomlightingor the speed of food
mixers and electric drills.
It might be assumed that the cost of the electronics would
be high, but this is not necessarily the case. With triacs and
thyristors it is possible to produce high performance mains
controllers which use only a few simple components. The
following notes give details of some typical control circuits
and highlight areas for special attention when adapting the
designs for specific applications.
Vacuum cleaner suction control
The competitive nature of the vacuum cleaner market has
led to the development of a wide variety of machine types
and accessories. In many cases, the speed of the motor
remains constant and, if suction control is attempted, it
consists merely of an adjustable vent in the air flow path.
Electronicsuctioncontrol sounds somewhat expensiveand
unnecessarily complicated for such an elementary
application. In fact, by using a BT138 triac, a simple but
nevertheless effective and reliable suction control circuit
(Fig. 1) can be constructed very economically, and is
suitable for all types of cleaner with a power consumption
of up to 900W.
The heart of the circuit is the BT138. This is a glass
passivated triac which can withstand high voltage
bidirectional transients and has a very high thermal cycling
performance. Furthermore its very low thermal impedance
minimizes heatsink requirements.
Fig. 1 Vacuum cleaner suction control circuit
Circuit Description
In Fig. 1 the BT138 is the power control element. Its action
is controlled by a diac which is switched on by a charge on
C
1
under the control of potentiometer R
2
. The resistance of
the diac is virtually infinite as long as the voltage across it
remains within the breakover voltage limits, -V
BO
to +V
BO
.
During each half cycle of the mains sinewave, C
1
charges
until the voltage across it exceeds the diac breakover
voltage. The diac then switches on and C
1
discharges itself
into the gate of the triac and switches it on. Diodes D
1
and
D
2
stabilise the supply voltage to the charging circuit so that
its operation is independent of mains voltage fluctuations.
If -V
BO
and +V
BO
are equal and opposite, the triac will be
triggered at the same time after the start of either a positive
or negative half cycle. The conduction angle, and therefore
the speed of the motor and the cleaner suction, is
determined by the adjustment of R
2
. Preset potentiometer
R
3
is used to set the minimum suction level. The width and
amplitude of the trigger pulses are kept constant by gate
resistor R
4
. The zinc oxide voltage dependent resistor (U)
minimises the possibility of damage to the triac due to very
high voltage transients that may be superimposed on the
mains supply voltage. Figure 2 shows the current and
voltage waveforms for the triac when the conduction angle
is 30˚.
a) Triac Voltage (100 V/div. 5ms/div.)
b) Triac Current (1 A/div. 5ms/div.)
Fig. 2 Vacuum cleaner - Triac waveforms
Circuit Performance
A laboratory model of the circuit has been tested to
determine the range of control that it has over the suction
power of a typical vacuumcleaner. For the test, the cleaner
was loaded with a water column. The result of the test is
shown graphically in Fig. 3. The measured range of water
column height (100 to 1100 mm) translates into a wide air
flow range - from little more than a whisper to full suction.
527
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 3 Suction power as a function of motor speed
As suction power is a function of the speed of the vacuum
cleaner motor, a second test was carried out to determine
the range of motor speed control under conditions of
minimum and maximum air flow (i.e. with the suction
blocked and unrestricted). This test also checked the motor
speed variation due to t10% variation of a nominal 220 V
AC mains supply. The initial test conditions were:
unrestricted flow; mains supply 198 V (220 V - 10%); R
2
at
maximum resistance, and R
3
set so that the motor just ran.
Table 1 shows the results of the test. N
min
is the speed at
which the motor just runs and N
max
is the speed of the motor
with R
2
set at minimum resistance.
mains blocked air flow unrestricted air flow
voltage N
min
N
max
N
min
N
max
(V) (rpm) (rpm) (rpm) (rpm)
198 5300 17100 4300 15400
220 6250 19000 5000 17100
242 7400 20000 6000 18200
Table 1. Motor speed figures for circuit of Fig. 1
The table shows that the speed setting range is wide. The
ratio of N
max
to N
min
is 3.42:1, for 220 V mains and
unrestricted airflow. The variation of motor speed due to
variation of the mains input is quite small and represents a
negligible change of suction. If D
1
and D
2
are omitted from
thecircuit, the speed settingratio is reduced to1.82:1under
the same conditions. The table also shows that the
difference between the N
min
for minimum and maximum air
flow is quite small. This implies that speed stabilisation is
unnecessary.
Special Design Considerations
The circuit shown in Fig. 1 has been shown to work well in
a typical vacuum cleaner application. But motors and
environments do vary, so some aspects of the design
should be looked at carefully before it is finalised.
Circuit positioning
The siting of the circuit, within the case of the cleaner, is
particularly important. In some areas within the cleaner the
temperature can be quite high. The circuit, and in particular
the triac and its heatsink, should not be placed in one of
these areas if the designer is to avoid problems keeping
the temperature of the triac below T
jmax
.
Starting current
Another factor that may lead to thermal problems is that of
inrush current. The starting current of a vacuum cleaner
motor is typically as shown in Fig. 4. The rms current during
the first 20 ms could be 20 A or more. The current decays
to its steady state value in about 1 s. To ensure that the
triac does not overheat, reference should be made to the
inrush current curves in the triac data sheet, the curve for
the BT138 is reproduced in Fig. 5.
Fig. 4 Starting current (20 A/div, 50 ms/div.)
cycle time peak rms ’limit’
number. (ms) current current current
(A) (A) (A)
1 20 49 22 24
2 40 41 18 21
3 60 35 13 19.5
4 80 32 14 18.5
5 100 29 13 18
10 200 20 9 15.5
20 400 14 6.3 14
Table 2. Currents during starting
The first step in checking for a problem is to estimate the
mounting base temperature, T
mb
, prior to starting. A
reasonable figure would be the worst case steady state
value of T
mb
during normal running.
528
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 5 Inrush current curve for BT138
Step 2 is to calculate the rms value of one cycle of the
starting current at several times during start up and step 3
is to compare these figures with the values taken from the
appropriate line of the inrush current curve.
As an example consider the performance of the BT138
driving a motor whose starting current is shown in Fig. 4.
Direct measurement indicated that during normal running
the T
mb
of a BT138 mounted on a particular heatsink, would
be no more than 22˚C above ambient. From other
measurements it was estimated that the ambient
temperature would not exceed 73˚C. These figures give a
worst case steady state T
mb
of 95˚C. It can be assumed that
this isthe highest temperaturethat the mountingbase could
be, prior tostarting - areasonableassumption which covers
the case where the motor has been running for a long time,
is turned off and then started again before there has been
any cooling.
The rms values of cycles 1 to 5, 10 and 20 of the starting
current are given in Table 2. Since the current is not an
ideal sine wave these have been calculated from the peak
current by assuming a crest factor (peak to rms) of 2.23.
Also shown are the relevant I
O(RMS)
figures from the 95˚C
line of Fig. 5. Since ’actual’ inrush current is always less
than the ’allowed’ current it is safe to use the BT138 under
the proposed conditions to control the motor. It should be
noted that because the crest factor is >√2 the dissipation
of theBT138 will beless thanassumed bythe inrushcurrent
curves of Fig. 5.
Commutation
The circuit shown in Fig. 1 has no RC snubber. This was
because the values of dI/dt and dV/dt generated by the
circuit were well within the capability of the BT138. This will
often be the case with vacuum cleaner motors for two
reasons:
- these motors introduce only a small phase shift in the
current, so the voltage step is small and the dV/dt is low,
- the steady state value of the current is much less than
the maximumrating of the BT138, this amounts to a dI/dt
well within the capability of the BT138.
However care must be taken to ensure that this is true in
all applications. Inparticular, careshouldbetakentoensure
that the triac switches correctly even during starting. If a
snubber is found to be necessary then a 100 Ω 0.5 W
resistor in series with a 0.1µF capacitor will be more than
adequate in most circumstances.
Interference
It is, of course, necessary to check that the overall
equipment complies with local regulations for conducted
and radiated interference. However, the measures taken to
suppress the electrical ’noise’ of the motor combined with
the motor itself will often be more than sufficient to
overcome the interference generated by the switching of
the triac but this must be checked in all applications.
529
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Domestic lamp dimmer
The use of light dimmers, once the prerogative of
entertainment centres, has now become widespread in the
home. It is necessary to ensure that the component parts
of these units are simple and reliable so that they are
compatible with the domestic environment.
The glass passivated BT138 triac meets these
requirements. Firstly, it has a peak non-repetitive on-state
current handling capability of up to 90 Awhich means it can
easily withstand the inrush current that occurs when a cold
lamp is switched on. It can also withstand high voltage
bidirectional transients and its low thermal impedance
minimizes heatsink requirements.
Fig. 6 Lamp dimmer circuit
Circuit Description
A simple circuit of a light dimmer using the BT138 is given
inFig. 6. The BT138 is the power control element, triggered
viathe diac. The setting of potentiometer R
2
determines the
phase difference between the mains sine wave and the
voltageacross C
2
. This in turn sets the triac triggering angle
and the lamp intensity.
The resistanceof the diac is very high as longas the voltage
across it remains within its breakover voltage limits, -V
BO
to
+V
BO
. Each half cycle of the mains charges C
2
via R
1
, R
2
and R
3
until the voltage being applied to the diac reaches
one of its breakover levels. The diac then conducts and C
2
discharges into the gate of the triac, switching it on. If -V
BO
and +V
BO
are equal and opposite, the triac will be triggered
at the same time after the start of either a positive or
negative half cycle. If C
1
were not included in the circuit,
thevoltageacrossC
2
wouldchange abruptly after triggering
and cause the phase relationship between the mains
voltage and voltage across C
2
to progressively alter. This
would cause an undesirable hysteresis effect. The voltage
across C
1
partially restores the voltage across C
2
after
triggering and thereby minimizes the hysteresis effect. The
width and amplitude of the trigger pulses are kept constant
by gate resistor R
4
. The VDR minimizes the possibility of
thetriac beingdamaged by highvoltagetransients that may
be superimposed on the mains supply voltage.
Special Design Considerations
Circuit rating
TheBT138 has an rms current rating of 12 A. It is, therefore,
capable of controlling loads with a rating of 2 kW or more.
However, theloadof this circuit must be restrictedtoamuch
lower level. There are two reasons for this. The first is to
keep mains distortion within the allowed limits, without the
necessity of expensive filter networks. The second reason
is to limit dissipation. If, as is likely, the circuit is to be
mounted in the wall in place of a conventional switch, then
air circulation is going to be very restricted and the ambient
temperature around the circuit will be quite high. It is
important for reliability reasons to ensure that the
temperature of the BT138 never exceeds T
jmax
, so the
dissipation of the triac must be kept to a low level.
Fig. 7 Interference on mains supply
Interference
Regulations concerning conducted and radiated
interference vary considerably form country to country but
it is likely that some formof filter will be needed. The simple
LCfilter shown within the dashed-lined box in Fig. 6 is often
all that is needed. The values of the filter components will
vary, but a combination of 0.15 µF capacitor and a low Q
inductor of 2.5 µH was found to be sufficient for the circuit
to meet the C.I.S.P.R. limits. This is illustrated by the plots
shown in Fig. 7. Curves (a) and (b) show the level of noise
on the mains supply for the circuit, without filter, when
530
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
controlling 550 W and 25 W loads respectively. Curves (d)
and (e) are for the circuit with filter connected showing that
the C.I.S.P.R. limit, which is curve (c), has been met.
Filter inductor
Having selected the value of filter inductor, the designer
has then to decide how to make it. Construction will not be
too critical - it is not necessary to achieve a high Q - and
there will be considerable room for reducing its size.
However, care must be taken to ensure that the inductor
does not saturate when the inrush current of a cold lamp
flows through it. If the inductor does saturate then the filter
capacitor will, effectively, be shorted out by the triac. In this
case the triac current could rise faster than the dI/dt rating
allows. This could cause progressive damage to the triac
resulting in premature failure.
Speed control for food mixers and electric
drills
Food mixers and electric hand drills are products whose
useability is improved by the addition of electronic speed
control. But they are products wherecosts have to betightly
controlled so the choice of circuit is very important. This
decision is made harder by the need to have a good speed
regulation under the widely varying loads that these
products are subjected to.
The circuits to be described provide continuous control of
motor speed over a wide speed range by adjusting the
conduction angle of a BT151 thyristor. They compensate
for load variation by adjusting the firing angle when there
is a change in the motor speed - as indicated by a change
in its back EMF.
Back EMF Feedback Circuits
Asimplemotor speedcontrol circuit that employs backEMF
tocompensatefor changes in motor load andmains voltage
is shown in Fig. 8(a). The resistor chain R
1
, R
2
,R
3
and diode
D
1
provide a positive going reference potential to the
thyristor gate via diode D
2
. Diode D
1
is used to reduce the
dissipation in the resistor chain by some 50%and diode D
2
isolates the trigger circuit with the thyristor in the on-state.
When the thyristor is not conducting the motor produces a
back EMF voltage across the armature proportional to
residual flux and motor speed. This appears as a positive
potential at the thyristor cathode.
A thyristor fires when its gate potential is greater than
cathode potential by some fixed amount. Depending on the
waveform shape and amplitude at the gate, the circuit may
function in several modes.
a) Basic Controller
b) Improved Low Speed Controller
c) Improved Low and High Speed Controller
Fig. 8 Thyristor Speed Control Circuit Using Back EMF
Feedback
R1
5k6 6W
R2
1k0 1W
R3
150
D1
D2 BT151
R4
500
R1
5k6 6W
R2
2k0 1W
R3
150
D1
D2 BT151
R4
500
C1
25uF 25V
R1
5k6 6W
R2
2k0 1W
R3
150
D2 BT151
R4
500
D1
C1
50uF 40V
531
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 9 Waveforms with DC Gate Supply
If, for example, during positive half cycles a constant DC
potential was applied at the gate (see Fig. 9), the thyristor
would continue to fire at the beginning of each cycle until
the back EMF was large enough to prevent firing. Thyristor
firing would then continue intermittently at the beginning of
the positive cycles to maintain some average motor speed.
Referring to Fig. 8(a) the waveform appearing at the
thyristor gate will approximate to a half sine wave,
Fig. 10(a). As a result it is impossible for the firing angle to
be later than 90˚ - the most positive value of the trigger
potential. At lower motor speeds the firing angle might need
tobe 130˚ for smooth operation. If the maximumfiring angle
is limited to 90˚ then intermittent firing and roughness of
motor operation will result.
If, however, the waveform at the gate has a positive slope
value to an angle of at least 130˚ then it will be possible to
have a stable firing point at low speeds. Such a waveform
can be produced if there is some phase shift in the trigger
network.
Stable Firing at Small Conduction Angles
The trigger network of the circuit shown in Fig. 8(b) has
been modified by the addition of a capacitor C
1
and diode
D
1
. The diode clamps the capacitor potential at zero during
the negative going half cycles of the mains input. The
waveform developed across the capacitor has a positive
slope to some 140˚, allowing thyristor triggering to be
delayed to this point.
a) Basic Controller
b) Improved Low Speed Controller
c) Improved Low and High Speed Controller
Fig. 10 Gate Voltage Waveforms
v
Mains
Voltage
back
EMF
0 90 180
3
2
1
v
1
2
3
v
Mains
Voltage
0 90 180
3
2
1
v
1
2
3
v
Mains
Voltage
0 90 180
3
2
1
v
1
2
3
v
Mains
Voltage
0 90 180
532
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
As the slider of R
2
is moved towards R
1
, the peak of the
waveform at the gate will move towards 90˚ as shown in
Fig. 10(b). As the speed increases, the no load firing angle
will also advance by a similar amount so stability will be
maintained. This circuit will give smoother and more stable
performance than the circuit of Fig. 8(a). It will, however,
give a marginally greater speed drop for a given motor
loading at low speed settings. At the maximum speed
settings the circuit of Fig. 8(a) approximates to that of
Fig. 8(b).
Fig. 11 Simplified Firing Circuit
Improved Motor Performance With Stable
Firing
Both the circuits so far discussed have gate voltage
waveforms that are of near linear slope fromthe zero point
of each positive half cycle, see Figs. 10(a) and (b). This
means that the only time that the thyristor can be fired early
in the mains cycle, say at 20˚, is when the back EMF and
hence motor speed is very low. This effect tends to prevent
smooth running at high speeds and high loads.
Stable triggering, at lowangles, can be achieved if the gate
voltageramp starts each cycle at a small positive level. This
means that the time to reach the minimum trigger voltage
is reduced. The circuit of Fig. 8(c) is one way of achieving
this. In this circuit capacitor C
1
is charged during positive
half cycles via resistor R
1
and diode D
1
. During negative
half cycles the only discharge path for capacitor C
1
is via
resistors R
2
and R
3
.
Diode D
1
also prevents C
1
from being discharged as the
thyristor switches off by the inductively generated pulse
from the motor. As the value of resistor R
2
is increased,
capacitor C
1
is discharged less during negative half cycles
but its charging waveform remains substantially
unchanged. Hence the result of varying R
2
is to shift the DC
level of the ramp waveform produced across C
1
.
Diode D
2
isolates the triggering circuit when the thyristor is
ON. Resistor R
4
adjusts minimumspeed, and by effectively
bleeding a constant current, in conjunction with the gate
current from the triggering circuit, it enables resistor R
2
to
give consistent speed settings.
Fig. 12 Calculated Gate Waveforms
Circuit Design
If the speed controller is to be effective it must have stable
thyristor firing angles at all speeds and give the best
possiblespeed regulation with variations of motor load. The
circuit of Fig. 8(c) gives a motor performance that satisfies
both of the above requirements.
There are two factors that are important in the circuit
operation in order to obtain the above requirements.
- The value of positive slope of the waveform appearing at
the thyristor gate.
- The phase angle at which the positive peak gate voltage
is reached during a positive half cycle of mains input.
R1
R2
D2
D1
C1
v
i
i1 i2
0 50 100 150 200 250
0
5
10
15
20
25
30
35
Phase Angle
Gate Voltage
C1=32uF C1=50uF C1=64uF
R2=1500
R2=800
R2=200
533
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
As previously described the charging of capacitor C
1
by
resistor R
1
determines the rate of rise of voltage at the
thyristor gate during the positive half cycle. However,
resistor R
1
must also have a value such that several times
the maximum thyristor gate current passes through the RC
network to D
1
. This current will then give consistent speed
settings with the spread of thyristor gate currents when the
minimum speed is set by resistor R
4
.
The positive slope value of the thyristor gate voltage will
have to be fixed according to the motor used. A motor that
gives a smooth back EMF voltage will allow a low slope
value to be used, giving good torque speed characteristics.
Some motors have coarser back EMF waveforms, with
voltage undulations and spikes, and a steeper slope of
thyristor gate voltage must be used in order to obtain stable
motor operation. The value of capacitor C
1
is chosen to
provide the required positive slope of the thyristor gate
voltage.
b) measurement circuit
b) typical waveform.
Fig. 13 Back EMF Measurement Arrangement
Some calculations have been made on the circuit of
Fig. 8(c) simplified to the form of Fig. 11, where it is
assumed that current flowing to the thyristor gate is small
compared with the current flowing through resistor R
1
. An
expression has been derived for the voltage that would
appear at the anode of D
2
in terms of R
1
, R
2
and C
1
and is
given later. Component values have been substituted into
the expression to give the thyristor gate waveforms shown
in Fig. 12.
In order to adjust the circuit to suit a given motor, the back
EMF of the motor must be known. This may be measured
using the arrangement shown in Fig. 13. The voltage
appearing across the motor is measured during the period
when the series diode is not conducting (period A). The
voltage so obtained will be the motor back EMF at its top
speed on half wave operation, and corresponds to the back
EMF that would be obtained fromthe unloaded motor at its
highest speed when thyristor controlled. In practice, since
the mains input is a sine wave, there is little increase in the
’noload’ speedwhen the firing angle is reduced to less than
about 70˚.
The value of resistor R
2
in Fig. 8(c) determines the motor
’no load’ speed setting. The waveforms of Fig. 12 may be
used as a guide to obtaining the value of this resistor. It
must be chosen so that at 70˚ and at its highest value, the
gate voltage is higher than the measured back EMF by
about 2 V- theforwardgate/cathodevoltageof the thyristor.
The thyristor is turned ONwhen a trigger waveform, shown
in Fig. 12, exceeds the back EMF by the gate/cathode
voltage. So, if the back EMF varies within a cycle then there
will be acycle to cycle variation in the firing angle. Normally,
random variations of the firing angle by 20˚ are tolerable.
If, for example, there were variations in the back EMF of
1 V, then with a firing angle of 70˚and a capacitor of 32 µF,
the variation of firing angle would be about 12˚. With
capacitor values of 50 µF and 64 µF the firing angles
variations would be 19˚ and 25˚ respectively. Therefore, a
capacitor value of 50 µF would be suitable.
Performance
The torque speed characteristics of the three circuits, when
used to drive an electric drill, are compared in Fig. 14. It
may be seen that the circuit of Fig. 8(b) has a poorer
performance than the two other circuits. That of Fig. 8(c)
may be seen to give a similar performance to the circuit of
Fig. 8(a) at low speeds but, at high speeds and torques, it
is better. It should be noted that the circuits of Figs. 8(b)
and (c) provide low speed operation free from the
intermittent firing and noise of the Fig. 8(a) circuit. Figure
15 compares the circuits of Fig. 8(a) and 8(c) when the load
is a food mixer motor.
to
oscilloscope
back
EMF
period
diode
conductiong
period
B
A
back EMF
534
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 14 Performance with Hand Drill Load
Fig. 15 Performance with Food Mixer Load
Circuit Calculations
The following analysis derives an expression for voltage ’v’
at the anode of D
2
. This expression can be used to produce
the gate voltage waveforms shown in Fig. 12. The analysis
assumes that the current drawn by the thyristor gate is
negligible in comparison with the current flowing in R
1
.
The charging current i
1
for capacitor C
1
in Fig. 11, is given
by:
and
Representing a mains half sine wave by where is the
peak mains voltage.
therfore,
where i, i
1
, i
2
are instantaneous currents.
Simplifying:-
Fourier analysis of a half sinewave gives:-
neglecting terms of the Fourier series with n > 2, then
then
simplifies to
where A, B, D, X, Y are constants.
Put
where a, b, c, d are constants.
0 0.5 1 1.5 2 2.5
0
500
1,000
1,500
2,000
2,500
3,000
Motor Torque (Nm)
Motor Speed (rpm)
Series
diode
Fig.8(a)
circuit
Fig.8(b)
circuit
Fig.8(c)
circuit
i
1
·
dq
dt
· C
1
dv
dt
i
2
·
v
R
2
f(E) E
i ·
f(E) − v
R
1
· i
1
+ i
2
f(E) − v
R
1
· C
1
dv
dt
+
v
R
2
C
1
dv
dt
+ v
¸

¸
1
R
1
+
1
R
2
_

,
·
f(E)
R
1
0 0.1 0.2 0.3 0.4 0.5 0.6
0
2,000
4,000
6,000
8,000
10,000
12,000
Motor Torque (Nm)
Motor Speed (rpm)
series
diode
Fig.8(c)
circuit
Fig.8(a)
circuit
f(E) · E
¹
'
¹
1
π
+
1
2
sin(θ) −
2
π

n · 2, 4, 6...
n · 0
cos(nθ)
n
2
− 1
¹
;
¹
C
1
dv
dt
+ v
¸

¸
1
R
1
+
1
R
2
_

,
·
E
R
1
¹
'
¹
1
π
+
1
2
sin(ωt ) −
2

cos(2ωt )
¹
;
¹
C
1
dv
dt
+ v
¸

¸
1
R
1
+
1
R
2
_

,

E
R
1
π
·
E
R
1
¹
'
¹
1
2
sin(ωt ) −
2

cos(2ωt )
¹
;
¹
(1)
A
dv
dt
+ Bv − D · X sin(ωt ) − Y cos(2ωt ) (2)
v · a sin(ωt ) + b cos(ωt )
+ c sin(2ωt ) + d cos(2ωt ) +
D
B
(3)
535
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
substituting (3) and (4) in equation (2) and equating terms
in , then
substituting for the constants in equation (2) gives:
This may be simplified since
Sothe voltage that the trigger circuit would apply tothe gate
(assuming the gate draws no current) is given by:
Solving this equation for a different values of C
1
and
positions of R
2
gives the curves shown in Fig. 12.
Conclusions
The addition of electronic control can enhance the overall
useability of many domestic products. Cost and
performance requirements are major factors when
determining the type of control circuit to be used in these
applications. It is possible, using thyristors and triacs, to
construct a range of phase control circuits which can meet
many of these cost and operational requirements.
Althoughthesecircuits arenot complexanduseonlysimple
components, it isstill important todesign withcaretoensure
that the best performanceis achieved. This report has given
examples of some of these circuits and has highlighted the
areas of their design requiring particular care.
dv
dt
· aωcos(ωt ) − bωsin(ωt )
+ 2cωcos(2ωt ) − 2dωsin(2ωt ) (4)
v ·
R
2
E
π(R
1
+ R
2
)
+
R
2
E

2
C
1
2
R
1
2
R
2
2
{(R
1
+ R
2
)sin(ωt )
− ωC
1
R
1
R
2
cos(ωt )

2

ωC
1
R
1
R
2
sin(2ωt )

1

(R
1
+ R
2
)cos(2ωt )}
cos(ωt ), cos(2ωt ), sin(ωt ), sin(2ωt )
v ·
BX
A
2
ω
2
+ B
2
. sin(ωt ) −
AωX
A
2
ω
2
+ B
2
. cos(ωt )

2AωY
4A
2
ω
2
+ B
2
. sin(2ωt )

BY
4A
2
ω
2
+ B
2
. cos(2ωt ) +
D
B
v · R
2
E.
(R
1
+R
2
) sin(ωt) −ωC
1
R
1
R
2
cos(ωt)

2
C
1
2
R
1
2
R
2
2
+(R
1
+R
2
)
2
+R
2
E.
1
π(R
1
+R
2
)
−R
2
E.
4ωC
1
R
1
R
2
sin(2ωt) +4(R
1
+R
2
) cos(2ωt)
3π[4ω
2
C
1
2
R
1
2
R
2
2
+(R
1
+R
2
)
2
]
(R
1
+ R
2
)
2

2
C
2
R
1
2
R
2
2
536
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
6.2.3 Design of a Time Proportional Temperature Controller
Electronic temperature control is no longer new: phase and
on/off controls for heaters have been widely used toreplace
mechanical switches. However, both phase control and
on/off control have disadvantages. Conventional phase
control allows fully-proportional control of the power
dissipatedintheload, but thehighrates of change of current
and voltage cause RFI and transients on the mains supply.
Because of this effect, phase control is not allowed to be
used for domestic heaters. Simple on/off control with
zero-voltage switching avoids generation of RFI but the
amount of hysteresis required to prevent temperature
oscillations does not give the required control accuracy.
The principle of time-proportional control
Time proportional control combines the zero-voltage
switching of on/off control with the accuracy of proportional
control and so eliminates the disadvantages of these two
alternative systems. Time-proportional control regulates
the load power such that there will be no overshoot or
undershoot of the desired temperature as is the case with
normal on/off systems. The TDA1023 has been designed
to provide time-proportional control for room heaters and
electric heating elements using a minimum number of
external components. It incorporates additional features to
provide fail-safe operation and fine control of the
temperature.
There are three states of operation when using
time-proportional control:
• load switched fully off,
• loadpower proportional tothe differencebetween actual
and desired temperatures,
• load switched fully on.
Figure 1 illustrates the principle; the load is switched on
once and off once in a fixed repetition period, the ratio of
the on and off periods providing the proportional control.
This method of control can cause mains flicker; the mains
voltage changes slightly each time the load is switched on
or off.
CENELEC, the European Committee for Electro-technical
Standardisation, has published rules which limit the rate at
which domestic heating apparatus may be switched on and
off. Table 1 gives the minimumrepetition period for a range
of load powers and common mains voltages from
CENELEC publication EN50.006.
Fig. 1 Duty cycle control
Appliance Repetition period, t
o
(s)
Power (W) 220V 240V 380V
600 0.2 0.2
800 0.8 0.3 0.1
1000 2.0 1.0 0.2
1200 4.6 2.0 0.2
1400 7.0 4.3 0.2
1600 10.0 6.3 0.3
1800 16.0 8.9 0.5
2000 24.0 13.0 0.9
2200 32.0 17.0 1.3
2400 40.0 24.0 1.9
2600 31.0 2.6
2800 3.6
Table 1. CENELEC minimum repetition periods for
Domestic Heater Applications
Description of the TDA1023
The TDA1023 is a 16-pin dual in-line integrated circuit
designed to provide time-proportional power control of
electrical heating elements. The TDA1023 is ideally suited
for the control of:
• Panel heaters
• Cooker elements
• Electric irons
• Water heaters
• Industrial applications, e.g. temperature controlled oil
baths, air conditioners.
The TDA1023 Incorporates the following functions:
• A stabilised power supply. The TDA1023 may be
connected directly to the AC mains using either a
dropping resistor or capacitor. It provides a stabilised
reference voltagefor the temperature-sensing network.
ON
OFF
Input
Output
t
0
t
ON
Temperature
reference
Ramp voltage
537
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 2 Triac trigger pulse width requirement
• A zero-crossing detector to synchronise the output
trigger pulses to the zero-crossings of the mains supply.
The detector produces a pulse, the duration of which is
determined by an external resistor, centred on the
zero-crossing of the mains voltage.
• A comparator with adjustable hysteresis, preventing
spurious triggering of the output. This compares a
thermistor voltage, a function of the room temperature,
with the voltage from the temperature selection dial.
• A voltage translation circuit for the potentiometer input.
Normally, the relatively small temperature variation in a
room (5˚C to 30˚C) corresponds to a narrow angle of
rotation of a potentiometer shaft. Use of this circuit
doubles the angle of rotation of the potentiometer shaft
for the same temperature range.
• A sensor fail-safe circuit to prevent triggering if the
thermistor input becomes open or short-circuited.
• Atiminggenerator with an adjustableproportional band.
This allows a full 100% control of the load current over
a temperature range of only 1˚C or 5˚C. The repetition
period of the timing generator may be set by an external
capacitor to conformto the CENELECspecifications for
mains load switching.
• An output amplifier with a current-limited output. The
amplifier has an output current capability of at least
200mA and is stabilised to 10V while the current limit is
not exceeded.
• Input buffers, to isolate the voltage translation circuit
and comparator from external influences.
• A control gate circuit to activate the output if there is a
mains zero-crossing, the comparator is ON and the
fail-safe comparator is OFF.
Althoughdesigned specificallyfor time proportional control,
the TDA1023 is also suitable for applications requiring
on/off control if the timing generator is not used.
Required Duration of Triac Trigger Pulse
The main advantage of triggering at the instant when the
applied voltage passes through zero is that this mode of
operation renders the use of RF suppression components
unnecessary. For time-proportional control, continuous
conduction of the triac may be required for many cycles of
the mains supply. To maintain conduction while the load
current is approaching the zero-crossing, the trigger pulse
must last from the time when the load current falls to the
value of the triac holding current (I
H
), until the time when
the load current reaches the triac latching current (I
L
).
I
L
I
H
I
L
t
p(min)
I
T
Triac current
Trigger pulse
Fig. 3 TDA1023 block diagram and external components
1
3
13 4 5 7 8 12
9
6
11 14 10 16
AC line
AC line
D1
RD
RS
R1
RP
CS
RNTC
Load
RG
CT
C1
Varistor
Input
buffers
Timing
generator
Voltage
translation
Zero
crossing
detector
Power
supply
Output
amplifier
Comparator
TDA1023
T1
538
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 4 Minimum pulse width as a function of supply voltage and latching current
In general, the latching current of a triac is higher than the
holding current, so the minimumtrigger pulse duration may
be taken as twice the time for the load current in the triac
(I
T
) torisefromzerotothetriac’slatchingcurrent. SeeFig. 2.
The current passed by the triac is a function of its on-state
voltage, the load resistance, and the applied voltage. The
trigger pulse width is therefore a function of:
• triac latching current (I
L
)
• applied AC voltage (v = Vsinωt)
• load resistance (R)
• on-state voltage of the triac (V
T
) at I
L
.
The load resistance is related to the nominal load power,
P and nominal supply voltage, V
s
by R=V
s
2
/P. Assuming
that the load resistance has a tolerance of 5% and the AC
voltage variation is 10%, the minimumrequired width of the
trigger pulse in the worst case can be calculated. The
graphs of Fig. 4 show t
p(MIN)
as a function of P for four
common mains voltages with values of 30mA, 60mA,
100mA, and 200mA for the triac latching current I
L
and a
maximum on-state voltage of V
T
=1.2V at I
L
.
Selection of external components
The external components required by the TDA1023
determine the operation of the device. The following
paragraphs describe the selection of these components to
ensure reliable operation under worst-case conditions.
Synchronisation Resistor, R
S
A current comparator is used as a zero-crossing detector
to provide trigger pulse synchronisation. It compares the
current through the synchronisation resistor (R
S
) with a
reference current. As the supply voltage passes through
539
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
zero, the current in the synchronisation resistor becomes
less than the reference current and a trigger pulse is given
until the current in R
S
increases above the reference level.
Thus, the duration of the trigger pulse depends upon the
rate of change of current in R
S
at the supply voltage
zero-crossing point. This rate of change is affected by:
• -the AC supply voltage
• the supply frequency
• the value of the synchronisation resistor.
Fig. 5 shows the value of R
S
as a function of trigger pulse
width, with the AC supply voltage as a parameter.
Fig. 5 Synchronisation resistor values as a function of
pulse width and AC voltage
Gate Resistor R
G
The guaranteed minimum amplitude of the output trigger
pulse of the TDA1023 is specified as 10V at an output
current less than 200mA. The output stage is protected
against damage due to short-circuits by current-limiting
action when the current rises above 200mA.
Although the output is current-limited, it is still
advantageous to include a gate series resistor in the circuit.
Inclusion of a gate resistor to limit the gate current to the
minimum value required reduces the overall current
consumption and the power dissipation in the mains
dropping resistor. Furthermore, the point at which current
limiting occurs is subject to considerable variation between
samples of the TDA1023: a gate resistor will reduce the
effect of this in production circuits.
The rectangular output V/I characteristic of the TDA1023
is shown in Fig. 6. Load lines for various values of gate
resistor have been plotted on this diagram so that the
maximumvalue of gate resistor can be selected by plotting
horizontal and vertical lines to represent the required
minimum gate current and voltage. The following example
illustrates the use of Fig. 6.
Thetriac tobetriggeredisaPhilipsBT139. At 0˚Cthetrigger
pulse requirements for a standard BT139 are:
I
GT
= 98 mA
V
GT
= 1.6 V
These figures are for triggering with a positive gate pulse
when MT
2
is negative with respect to MT
1
. The lines
representing V
GT
= 1.6V and I
GT
=98 mA cross the load line
for a gate resistor value of 82 Ω. The maximum value of
gate resistor is therefore 82 Ω.
Fig. 6 Gate voltage as a function of output current and
gate resistor values
Gate Termination Resistor R
PD
The TDA1023 has a resistor approximately 1.5kΩbetween
Pin 1 and Pin 13. This is intended for use as a pull-down
resistor when sensitive triacs are being used.
The Proportional Band Resistor R
5
The proportional band is the input voltage range that
provides control of 0% to 100% of the load power. The
TDA1023 has a built-in proportional band of V
pb
=80mV
(corresponding to about 1˚C) which can be increased by
the addition of resistor R
5
between Pin 5 and ground. The
maximum proportional band of 400mV is obtained by
shorting Pin 5 to ground.
Hysteresis Resistor R
4
The comparator of the TDA1023 is designed with built-in
hysteresis to eliminate instability and oscillation of the
output which would cause spurious triggering of the triac.
Apart from providing a stable two-state output, the
hysteresis gives the comparator increased noise immunity
and prevents half-waving.
Figure 7 shows the application of hysteresis to the
comparator and the transfer characteristic obtained. The
built-inhysteresis is20mV; this may be increasedbyadding
a resistor (R
4
) from Pin 4 to ground which increases the
current I
H
. Pin 4 shorted to ground gives a maximum of
320mV. Table 2 gives the value of R
4
for a range of
hysteresis settings.
0 50 100 150 200
0
2
4
6
8
10
RG=22R
27
33
39
47
56 68 82 100 120 180 270
RG=
390R
IG (mA)
VGT (V)
0 100 200 300 400
0
100
200
300
400
500
600
Vs=380V 240V 220V
110V
tp (us)
R5 (kOhm)
540
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 7 Temperature control hysteresis characteristic
When the proportional band (V
pb
) is increased, it may be
necessary to increase the hysteresis voltage (V
h
). Table 2
alsoshowsarangeof proportional bandsettings, thevalues
of R
5
required for these, the corresponding minimum
hysteresis voltage and the maximum value of hysteresis
resistor R
4
.
Proportional R
5
(kΩ) R
4
(kΩ) Hysteresis
Band (mV) band (mV)
80 - - 20
160 3.3 9.1 40
240 1.1 4.3 60
320 0.43 2.7 80
400 0.0 1.8 100
Table 2. Choice of components R
4
and R
5
Voltage AC DC Catalogue
(V) rating (µF) value (µF) Number
25 47 68 2222 016 90129
40 33 47 2222 016 90131
25 22 33 2222 015 90102
40 15 22 2222 015 90101
25 10 15 2222 015 90099
40 6.8 10 2222 015 90098
Table 3. Preferred capacitors for use with TDA1023
Power CENELEC C
T
(DC) t
0(nom)
t
0(min)
t
0(max)
(W) t
0
(s) (µF) (s) (s) (s)
2000 24.0 68 41 22 65
1800 16.0 47 28 15 45
1600 10.0 33 20 11 32
1400 7.0 22 13 7 21
1200 4.6 15 9 4.8 14
1000 2.0 10 6 3.2 9.6
800 0.8 10 6 3.2 9.6
600 0.3 10 6 3.2 9.6
Table 4. Timing capacitor values for 220V operation
Smoothing Capacitor, C
S
The smoothing capacitor is required to provide the supply
current to the TDA1023 during the negative half cycles of
the mains voltage waveform. As the TDA1023 possesses
an internal voltage stabilization circuit, a high input ripple
voltage can be tolerated. A practical preferred value of C
S
is 220µF, 16V.
Timing Capacitor, C
T
The minimum repetition period required for a particular
application was given in Table 1. This timing is selected
using the external capacitor C
T
. Typical electrolytic
capacitors have wide tolerances: up to -10% to +50%.
Moreover, the effective DCcapacitanceis different fromthe
marked (AC) value, usually greater. Thus, the use of
standard capacitors may lead to repetition periods far in
excess of those required. A range of electrolytic capacitors
has been developed for use with the TDA1023 (Table 3).
All further references to C
T
assume the use of the preferred
capacitors which have the following advantages:
• DC capacitance is known for each marked AC value.
• Tolerance for the DC capacitance is t20%.
• Very low leakage current (<1µA).
• Long lifetime (>100,000 hours at 40˚C).
Fig. 8 Temperature-sensing bridge circuit
Fig. 9 Temperature-sensing voltage translation circuit
V7
V6
Increasing temperature
V
OUT Ih.Rh
V9
V6
R1
R
NTC
RP
V11
Buffer
Translation
circuit
Buffer
Comparator
Fail-safe
Comparator
0.95V
11
6
7
8 9
TDA1023
V
11
RP
R
V
11
R1
NTC
541
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 10 Average gate current as a function of R
G
with R
S
and AC voltage as parameters
The timing circuit
The TDA1023 employs a triangular waveform for timing
purposes. The advantages of using a triangular waveform
are that for a given capacitor value the triangular waveform
provides twice the repetition period that the sawtooth gives.
This allows the use of smaller capacitors and minimises the
effects of the capacitor leakage current thus reducing the
spread in repetition periods.
The publisheddatafor the TDA1023specifies the repetition
period as 0.6 s t 0.2 s/µF. Table 4 shows the minimum
preferred value of C
T
(DC value) to provide the required
minimum repetition time for a range of appliance powers
operating at 220V AC. The resulting nominal, minimum,
and maximum repetition times are also given.
Input voltage translation circuit
Figure 8 shows a temperature sensing network which
requires a minimum of components and eliminates
performance spreads due to potentiometer tolerances. For
applications where the input voltage variation is very much
less than the available voltage then the required
temperature will be controlled by a small angle of rotation
of the potentiometer shaft. The TDA1023 voltage
translationcircuit allowstheuseof 80%of thepotentiometer
rotation giving accurate control of the temperature. If the
voltage translation circuit is not used then pins 9 and 11
must be shorted together to disable the circuit. A block
diagram of the translation circuit is shown in Fig. 9.
Fail-safe circuits
The TDA1023 is fail-safe for both short-circuit and
open-circuit conditions. Either of these conditions will
prevent production of trigger pulses for the triac.
Short-circuit sensing is automatically obtained from the
normal temperature sensing circuit. When the thermistor
input voltage is zero, the triac will never be triggered
because the potentiometer slider voltage will be higher. To
sense the open-circuit thermistor condition, an extra
comparator is used. This fail-safe comparator will inhibit
output pulses if the thermistor input voltage rises above a
542
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
reference value (see Fig. 9).
Determination of required supply current
Before any calculations concerning the required supply
current can be made, the maximumaverage output current
of the TDA1023 must be determined. The minimumsupply
current required is the sum of the following currents:
• the maximum average output current
• the current drawn by the temperature-sense circuit
• the current required by the integrated circuit.
For worst-case conditions, a 5% tolerance for R
S
and R
G
and a 10% variation of the mains is assumed. Figure 10
shows graphs of I
G(AV)max
as a function of R
G
and R
S
for four
50Hz supply voltages. Below R
G
=22Ω there is no further
increase in I
G
as the output current is limited. The current
drawn by the temperature-sensing circuit must not be
greater than 1mA. The current consumption of the
TDA1023 depends upon the hysteresis and proportional
bandsettings. Figure11 shows theminimumsupply current
as a function of the average output current for limit settings.
Fig. 11 Maximum required input current as function of
gate current for limits of hysteresis band settings
5 10 15 20 25 30
0
5
10
15
20
25
Gate
current
I3 (mA)
Supply current I16 (mA)
I4=0
I5=0
V4=0
V5=0
Fig. 12 Input current as a function of R
D
and power dissipation (with and without series diode)
543
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
The mains dropping resistor, R
D
The value of the mains dropping resistor must be chosen
such that the average supply current to the input of the
TDA1023 is at least equal to the required minimum. The
value of the resistor R
D
is defined by the maximum current
that can flowinto Pin 16, the maximumpeak mains voltage,
andthe minimumvoltage at Pin 16. Table 5 shows practical
values for R
D(min)
for four common mains supply voltages
Supply voltage V
s
(V) R
D(min)
(kΩ)
110 2.0
220 3.9
240 4.3
380 7.5
Table 5. Mains dropping resistor values
Fig. 13 Use of a mains dropping capacitor, C
D
The power dissipated by the dropping resistor has been
computed for four mains voltages as a function of R
D
and
the results plotted on the graphs of Fig. 12. The power
dissipated in R
D
may be considerably reduced by the
addition of a series diode as in Fig. 14. In this case there
is no conduction through R
D
during the negative half-cycle
of the supply voltage, giving a reduction of more than 50%
of the power dissipated in R
D
.
Use of a mains dropping capacitor
It is possible to replace the mains dropping resistor and
series diode with a capacitor, Fig. 13, and thereby reduce
the power dissipation in the voltage reduction components
still further. However, for mains voltages below 200V, the
power dissipated by the dropping resistor is comparatively
small and the use of a capacitor is not considered to be
necessary. For mains voltages above 240V, the additional
cost of the required high-voltage capacitor is not justified.
For these reasons, it is recommended that capacitive
voltage reduction is only used with mains supplies of
200V(RMS) or 240V(RMS).
When selectinga capacitor for mains voltage reduction, the
following points must be considered:
• AC voltage rating
• Suppression of mains-borne transients - A
voltage-dependent resistor must be connected
across the mains input to limit mains borne transients.
For R
SD
=390Ω this yields a maximum transient
voltage of about 740V. For 220V operation, a VDR
(catalogue number 2322 594 13512) will limit the
supply voltage to the required level during current
transients of up to about 200A. For 240V operation,
a VDR (catalogue number 2322 594 13912) will limit
the supply voltage to the required level during current
transients of up to about 80A.
• Limit of Inrush current - The capacitor C
D
must not
be chosen so large that the input current to the
TDA1023 violates the absolute maximumspecified in
the published data. A practical value for C
D
is 680nF.
Resistor R
SD
must also limit the peak value of the
inrush current to less than 2A under worst case
operating conditions. With a 240V(+10%) supply, the
value of 390Ω (-5%) will limit the worst case peak
value of the inrush current to:
Triac protection
If the mains dropping circuit consists of capacitor C
D
and
resistor R
SD
, a VDR must be included in the circuit as
describedabove. This VDRwill alsoprotect thetriac against
current surges in the mains supply. If the mains dropping
circuit consists of resistor R
D
and diode D1, the VDR may
be connected directly across the triac, giving improved
protectionduetotheseriesresistanceof theheater. Current
surges in the supply will not harm the TDA1023 as the
dropping resistor will limit the current to a safe level.
Application examples
The TDA1023 is intended primarily for room temperature
control usingelectric panel heaters. The controllable heater
power range is from 400W to 2000W, although the upper
limit may be increased by suitable choice of triacs and/or
heatsinks. The TDA1023 may also be used as a time
proportional switch for cooker elements and similar
devices, giving 100% control of the power dissipation.
1. Domestic panel heater controller
Figure 3 showed the design for a time proportional heater
control using the TDA1023. Economies may be gained by
the use of smaller or lower power components and so two
versions are described in Table 6. Version A, for heaters
from400Wto1200W, uses aBT138 triac anda15µF timing
capacitor, version B, for heaters from 1200W to 2000W,
uses a BT139 triac and a 68µF timing capacitor. Table 6
gives the necessary component values under worst case
conditions for each of these versions for use with mains
supplies of 220V, 50Hz.
13
16
C
D
R
SD
Varistor TDA1023
240 × 1.1
0.95 × 390
√2 · 1.01A
544
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
The capacitor C1 has been included in the circuit of Fig. 3
to minimise sensor line interference pick-up. This is only
necessary when the sensor is remote from the control
circuit. The built-in hysteresis and proportional band
provides optimum performance for panel heaters so pins 4
and 5 are not connected.
Component Version A Version B
400W - 1200W 1200W - 2000W
T
1
BT138-500 BT139-500
VDR
1
350V, 1mA 350V, 1mA
D
1
BYX10G BYX10G
R
1

2
18.7kΩ 18.7kΩ
R
NTC

3
R25=22kΩ,B=4200k R25=22kΩ, B=4200k
R
P
22kΩ 22kΩ
R
D
4.3kΩ 6.2kΩ
R
G
82Ω 82Ω
R
S
430kΩ 180kΩ
C
1
47nF 47nF
C
S
220µF, 16V 220µF, 16V
C
T
15µF (DC) 68µF (DC)
C
D

4
680nF 470nF
R
SD

4
390Ω 390Ω
Notes: 1. Cat. No. 2322 594 13512
2. 1% tolerance
3. Cat. No. 2322 642 12223
4. Only required if used in place of D
1
and R
D
Table 6. 220V, 50Hz temperature controller components
2. Temperature control of 2kW load.
For a load power of 2kWthe BT139 triac must be used. The
circuit is also that shown in Fig. 3. Table 7 gives a summary
of the required component values.
Component Value Remarks
T
1
BT139-500
VDR 350V, 1mA No. 2322 594 13912
D
1
BYX10G
R
1
18.7kΩ 1% tolerance
R
NTC
R25=22kΩ, B=4200k No. 2322 642 12223
R
P
22kΩ
R
D
6.8kΩ
R
G
82Ω
R
S
150kΩ
C
1
47nF
C
S
220µF, 16V
C
T
47µF (DC) No. 2222 016 90129
Table 7. 2000W, 220V, 50Hz temperature controller
Value of R
S
The required trigger pulse width can be found from Fig. 4
as a function of the load power, latch current and supply
voltage (2000W, 60mA, and 220V, 50Hz, respectively):
t
p(min)
=64µs. A value of R
S
=135kΩ provides a trigger pulse
of the required duration. The next preferred value above
this is 150kΩ, providing a t
p(min)
of approximately 70µs.
Value of R
G
The maximum value of R
G
that may be used is determined
by the minimum conditions to reliably trigger all samples of
the triac. In Fig. 6 it can be seen that the operation point of
1.6V and 98 mA lies on the load line for 82Ωand this is the
value chosen.
Value of C
T
For a load of 2kW, the repetition period must be at least
24s (from Table 1). From Table 4 the minimum preferred
value of C
T
to provide this period is 68µF. However, due to
the different performance under AC and DC conditions,
then from Table 3, the actual capacitor used should be
47µF, 25V.
Value of R
1
and R
P
For control over the range 5˚C to 35˚C and a thermistor
characteristics with R
25
=22kΩ, a suitable value of R
1
is
18.7kΩ t1%. A suitable value for R
P
is 22kΩ.
Value of R
D
First, the maximum average output current must be found.
From Fig. 10 the maximum gate current I
G
is given as a
function of the values of resistors R
S
and R
G
. For this circuit
I
G(AV)max
=5 mA. Once the maximum average output current
is known, the minimum required supply current can be
found from Fig. 11. With minimum hysteresis and
proportional band, the average value of the supply current
is 12.5 mA. Using this value of input current the required
value of R
D
can be found from Fig. 12 giving R
D
= 5.6 kΩ.
The power dissipation in the resistor when diode D
1
is
present in the circuit is then 5 W.
3. Time proportional power control
The TDA1023 may be used to provide proportional control
of devices such as electric cooker elements. The
temperature-sensingbridgeis replacedby apotentiometer,
the power in the load being proportional to the
potentiometer setting. Proportional power control is thus
obtained while the potentiometer voltage lies between the
upper and lower limits of the triangular waveform
comparator input.
As the timing capacitor is charged and discharged by
current sources, the voltage across it will never reach zero,
so that load power will be zero before the potentiometer
reaches its minimum setting. Similarly, maximum load
power is reached before the maximum setting of the
potentiometer. This effect can be reduced by the addition
of resistors R
1
and R
2
. To ensure that 0% and 100% load
545
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
power can be selected by the potentiometer setting, the
values of R
1
and R
2
should each be limited to 10% of the
value of R
P
.
All the circuit components are calculated in the same way
as for the temperature controller, including the timing
capacitor C
T
. An example circuit, with components suitable
for the control of loads from 1kW to 2kW from 220V, 50Hz
supplies, is shown in Fig. 14 and Table 8.
Component Value
T
1
BT139-500
VDR ZnO, 350V, 1mA
D
1
BYX10G
R
1
4.7kΩ
R
2
4.7kΩ
R
P
47kΩ
R
D
5.6kΩ
R
G
82Ω
R
S
220kΩ
C
S
220µF, 16V
C
T
47µF, 25V
Table 8. Time proportional power controller
Fig. 14 Time-proportional power regulation circuit
4. Phase control circuit using the TDA1023
Figure 15 shows an adjustable phase control trigger circuit
suitable for thyristor or triac controller applications. The
circuit uses the TDA1023 control chip and an NE555 timer
device to give output phase control proportional to the input
voltage command.
AC line
AC line
D1
RD
RS
CS
Load
RG
CT
Varistor
TDA1023
R1
R2
RP
11 14 10 16
3
13 12
7
6
9
T1
Fig. 15 Adjustable phase SCR/Triac trigger circuit
546
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Hi-Com Triacs
547
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
6.3.1 Understanding Hi-Com Triacs
Hi-Comtriacs fromPhilips Semiconductors are specifically
designed to give superior triac commutation performance
in the control of motors for domestic equipment and tools.
These devices are suitable for use with a wide variety of
motor and inductive loads without the need for a protective
snubber. The use of a Hi-Comtriac greatly simplifies circuit
design and gives significant cost savings to the designer.
This product information sheet explains how the superior
characteristics and performance of Hi-Com triacs removes
design limitations of standard devices.
Triac commutation explained
A triac is an AC conduction device, and may be thought of
as two antiparallel thyristors monolithically integrated onto
the same silicon chip.
In phase control circuits the triac often has to be triggered
into conduction part way into each half cycle. This means
that at the end of each half cycle the on-state current in one
direction must drop to zero and not resume in the other
direction until the device is triggered again. This
"commutation" turn-off capability is at the heart of triac
power control applications.
If the triac were truly two separate thyristors this
requirement would not present any problems. However, as
the two are on the same piece of silicon there is the
possibility that the "reverse recovery current" (due to
unrecombined charge carriers) of one thyristor as it turns
off, may act as gate current to trigger the other thyristor as
the voltage rises in the opposite direction. This is described
asa"commutationfailure" andresults inthe triac continuing
to conduct in the opposite direction instead of blocking.
The probability of any device failing commutation is
dependent on the rate of rise of reverse voltage (dV/dt) and
the rate of decrease of conduction current (dI/dt). The
higher the dI/dt the more unrecombined charge carriers are
left at the instant of turn-off. The higher the dV/dt the more
probable it is that some of these carriers will act as gate
current. Thus the commutation capability of any device is
usually specified in terms of the turn-off dI/dt and the
re-applied dV/dt it can withstand, at any particular junction
temperature.
If a triac has to be operated in an inductive load circuit with
a combination of dI/dt and dV/dt that exceeds its
specification, it is necessary to use an RC-snubber network
in parallel with the device to limit the dV/dt. This is at a
penalty of extra circuit complexity and dissipation in the
snubber. The "High Commutation" triacs (Hi-Com triacs)
are designed to have superior commutation capability, so
that even at a high rate of turn-off (dI/dt) and a high rate of
re-applied dV/dt they can be used without the aid of a
snubber network, thus greatly simplifying the circuit. The
design features of Hi-Com devices that have made this
possible are:
Geometric separation of the two
antiparallel thyristors
Commutation failure can be avoided by physically
separating the two ’thyristor halves’ of a triac. However,
separating them into two discrete chips would remove the
advantage of a triac being triggerable in both directions by
the same gate connection. Within the integrated structure
of a Hi-Com triac the two halves of the device are kept
further apart by modifying the layout of the chip in order to
lessen the chance of conduction in one half affecting the
other half.
Emitter shorting
"Emitter shorts" refer tothe on-chip resistive paths between
emitter and base of a transistor. A higher degree of emitter
shortingmeans the presence of more such paths and lower
resistance values in them. The use of emitter shorts in a
triac has two effects on commutation.
Fig. 1 Standard triac triggering quadrants
Firstly it reduces the gain of the internal transistors that
make up the triac. This means there will be fewer carriers
left to recombine when the conduction current falls to zero,
and therefore a smaller probability that a sufficient number
will be available to re-trigger the triac. The second way in
which emitter shorts help commutation is that any
unrecombined carriers inthe conducting thyristor at turn-off
Quadrant 1
Quadrant 2
Quadrant 4
Quadrant 3
G+ G-
MT2+
MT2-
I
G
I
G
I
G
I
G
+ +
-
-
+
-
+
-
549
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
will have more chance of flowing out through the emitter
shorts (of the opposite thyristor) rather than acting as gate
current to trigger that thyristor on.
The Hi-Comtriacs have a higher degree of emitter shorting
both around the periphery of the device and in the central
part of the active area. This both reduces the number of
carriers available, and lessens the danger of any available
carriers acting as gate current for undesirable triggering.
Modified gate structure
The gate of a triac allows conduction in both directions to
be initiated by either a positive or a negative current pulse
between gate (G) and main terminal (MT1). The four
different modes of triggering are often called 1+, 1-, 3- and
3+ (or sometimes quadrants 1, 2, 3 and 4) and are shown
in Fig. 1.
This triggering versatility arises from the fact that the gate
consists of some elements which conduct temporarily
during the turn-on phase. In particular, one of the triggering
modes, 3+ (or quadrant 4), relies on the main terminal 1
supplying electrons to trigger a thyristor element in the
gate-MT1 boundary. Conduction then spreads to the main
thyristor element from this boundary.
Unfortunately the carrier distribution in this triggering mode
of operation is very similar to that existing when the triac is
commutating in the 1-to-3 direction (i.e changing from
conductionwithMT2positivetoblockingwith MT1positive).
The presence of the element in the gate to allow 3+
triggering will therefore always also undermine
commutation capability in the 1-to-3 direction. For this
reason the Hi-Com triacs have a modified gate design to
remove this structure. This incurs the penalty that the 3+
trigger mode cannot be used, but it greatly improves the
commutation performance of the device.
Conclusions
By modifications to the internal design and layout of the
triac it is possible to achieve a high commutation capability
triac for use in inductive andmotor load applications. These
modifications have been implemented in the Hi-Comrange
of devices from Philips Semiconductors. The devices can
be used in all typical motor control applications without the
need for a snubber circuit. The commutation capability of
the devices is well in excess of the operating conditions in
typical applications.
As the loss of the fourth trigger quadrant can usually be
tolerated in most designs, Hi-Com triacs can be used in
existing motor control applications without the snubber
network required for a standard device. This gives the
designer significant savings in design simplicity, board
space and system cost.
550
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
6.3.2 Using Hi-Com Triacs
Hi-Comtriacs fromPhilips Semiconductors are specifically
designed to give superior triac commutation performance
in the control of motors for domestic equipment and tools.
These devices are suitable for use with a wide variety of
motor and inductive loads without the need for a snubber.
The use of a Hi-Com triac greatly simplifies circuit design
and gives significant cost savings to the designer.
This product information sheet explains how the need for
a triac snubber arises and how the superior performance
of Hi-Com triacs removes design limitations of standard
devices. The Hi-Com range is summarised in Table 1.
Triac commutation
For resistive loads the device current is in phase with the
line voltage. Under such conditions triac turn-off
(commutation) occursat thevoltage"zero-crossover" point.
This is not a very severe condition for triac commutation:
the slow rising dV/dt gives time for the triac to turn off
(commutate) easily.
The situation is quite different with inductiveor motor loads.
For these circuits conduction current lags behind the line
voltage as shown in Fig. 1. When triac commutation occurs
the rate of rise of voltage in the opposite direction can be
very rapid and is governed by the circuit and device
characteristics. This high dV/dt means there is a much
higher probability of charge carriers in the device
re-triggering the triac and causing a commutation failure.
Hi-Com triacs
Hi-Com triacs are specifically designed for use with ac
inductive loads such as motors. As commutation capability
is not an issue for resistive load applications then standard
triacs are still the most appropriate devices for these
applications. The significant advantage of a Hi-Com triac
is that it has no limitation on the rate of rise of reapplied
voltage at commutation. This removes the requirement for
a snubber circuit in inductive load circuits. An additional
advantage of the Hi-Comdesign is that the off-state (static)
dv/dt capability of the device is also significantly improved.
Whenusing Hi-Comtriacs in inductiveload applications the
trigger circuit cannot trigger the device in the fourth (3+)
quadrant (Fig. 2). Fortunately the vast majority of circuit
designs do not require this mode of operation and so are
suitablefor usewithHi-Comtriacswithout modification. The
circuit of Fig. 3 is a typical example of the simplest type of
trigger circuit. Hi-Com triacs are equally suitable for use
with microcontroller trigger circuits.
Parameter BTA212-600B BTA212-800B BTA216-600B BTA216-800B
Repetitive peak voltage V
DRM
(V) 600 800 600 800
RMS on-state current I
T(RMS)
(A) 12 12 16 16
Gate trigger current I
GT
(mA) 2 - 50 2 - 50 2 - 50 2 - 50
Off state dv/dt dV
D
/dt (V/µs) 1000 1000 1000 1000
Commutating di/dt dI
com
/dt (A/ms) 24 24 28 28
Turn-on di/dt dI
T
/dt (A/µs) 50 50 50 50
Package TO220 TO220 TO220 TO220
Table 1. Philips Semiconductors Hi-Com Triac range
551
Thyristors and Triacs Power Semiconductor Applications
Philips Semiconductors
Fig. 1 Triac commutation waveforms (inductive load)
Fig. 2 Hi-Com triac triggering quadrants
Device limiting values
i) Trigger current, I
GT
Trigger current for the Hi-Comtriacs is in the range 2mA to
50mA. This means that gate currents due to noise that are
below 2mA in amplitude can be guaranteed not to trigger
thedevices. This gives thedevicesanoiseimmunity feature
that is important in many applications. The trigger current
delivered by the trigger circuit must be greater than 50mA
under all conditions in order to guarantee triggering of the
device when required. As discussed above, triggering is
only possible in the 1+, 1- and 3- quadrants.
Fig. 3 Phase control circuit using Hi-Com triac
ii) Rate of change of current, dI
com
/dt
Hi-Com triacs do not require a snubber network providing
that the rate of change of current prior to commutation is
less than the rating specified in the device data sheet. This
dI
com
/dt limit is well in excess of the currents that occur in
the device under normal operating conditions, during
transients such as start-up and faults such as the stalled
motor condition.
For the 12A Hi-Com triacs the limit commutating current is
typically 24A/ms at 125˚C. This corresponds to an RMS
current of 54A at 50Hz. For the 16A Hi-Com triacs the limit
commutating current is typically 28A/ms at 125˚C. This
corresponds to an RMS current of 63A at 50Hz. Typical
stall currents for an 800W domestic appliance motor are in
the range 15A to 20A and so the commutation capability of
the Hi-Comtriacs is well above the requirement for this type
of application.
Conclusions
The Hi-Comrange of devices fromPhilips Semiconductors
can be used in all typical motor control applications without
the need for a snubber circuit. The commutation capability
of the devices is well in excess of the operating conditions
in typical applications.
As the loss of the fourth trigger quadrant can usually be
tolerated in most designs, Hi-Com triacs can be used in
existing motor control applications. By removing the
snubber the use of a Hi-Com triac gives the designer
significant savings in design simplicity, board space and
system cost.
V
DWM
-dI/dt
dV
com
/dt
Time
Time
Time
Supply
voltage
Load
current
Voltage
across
triac
Trigger
pulses
Current
Hi-Com triac
M
Quadrant 1
Quadrant 2
Quadrant 4
Quadrant 3
G+ G-
MT2+
MT2-
I
G
I
G
I
G
+ +
-
-
+
-
No triggering
possible
552
Preface Power Semiconductor Applications
Philips Semiconductors
Acknowledgments
We are grateful for all the contributions fromour colleagues within Philips and to the Application Laboratories in Eindhoven
and Hamburg.
We would also like to thank Dr.P.H.Mellor of the University of Sheffield for contributing the application note of section 3.1.5.
The authors thank Mrs.R.Hayes for her considerable help in the preparation of this book.
The authors also thank Mr.D.F.Haslam for his assistance in the formatting and printing of the manuscripts.
Contributing Authors
N.Bennett
M.Bennion
D.Brown
C.Buethker
L.Burley
G.M.Fry
R.P.Gant
J.Gilliam
D.Grant
N.J.Ham
C.J.Hammerton
D.J.Harper
W.Hettersheid
J.v.d.Hooff
J.Houldsworth
M.J.Humphreys
P.H.Mellor
R.Miller
H.Misdom
P.Moody
S.A.Mulder
E.B.G. Nijhof
J.Oosterling
N.Pichowicz
W.B.Rosink
D.C. de Ruiter
D.Sharples
H.Simons
T.Stork
D.Tebb
H.Verhees
F.A.Woodworth
T.van de Wouw
This book was originally prepared by the Power Semiconductor Applications Laboratory, of the Philips Semiconductors
product division, Hazel Grove:
M.J.Humphreys
C.J.Hammerton
D.Brown
R.Miller
L.Burley
It was revised and updated, in 1994, by:
N.J.Ham C.J.Hammerton D.Sharples
Preface Power Semiconductor Applications
Philips Semiconductors
Preface
This book was prepared by the Power Semiconductor Applications Laboratory of the Philips Semiconductors product
division, Hazel Grove. The book is intended as a guide to using power semiconductors both efficiently and reliably in power
conversion applications. It is made up of eight main chapters each of which contains a number of application notes aimed
at making it easier to select and use power semiconductors.
CHAPTER 1 forms an introduction to power semiconductors concentrating particularly on the two major power transistor
technologies, Power MOSFETs and High Voltage Bipolar Transistors.
CHAPTER 2 is devoted to Switched Mode Power Supplies. It begins with a basic description of the most commonly used
topologies and discusses the major issues surrounding the use of power semiconductors including rectifiers. Specific
design examples are given as well as a look at designing the magnetic components. The end of this chapter describes
resonant power supply technology.
CHAPTER 3 describes motion control in terms of ac, dc and stepper motor operation and control. This chapter looks only
at transistor controls, phase control using thyristors and triacs is discussed separately in chapter 6.
CHAPTER 4 looks at television and monitor applications. A description of the operation of horizontal deflection circuits is
givenfollowed by transistor selection guides for both deflectionand power supply applications. Deflection and power supply
circuit examples arealsogivenbasedoncircuitsdesignedbytheProduct Concept andApplicationLaboratories(Eindhoven).
CHAPTER5 concentrates on automotive electronics looking in detail at the requirements for the electronic switches taking
into consideration the harsh environment in which they must operate.
CHAPTER6 reviews thyristor andtriac applications fromthe basics of device technology and operation to the simple design
rules which should be followed to achieve maximum reliability. Specific examples are given in this chapter for a number
of the common applications.
CHAPTER 7 looks at the thermal considerations for power semiconductors in terms of power dissipation and junction
temperaturelimits. Part of this chapter is devotedto worked examples showing howjunctiontemperatures can becalculated
to ensure the limits are not exceeded. Heatsink requirements and designs are also discussed in the second half of this
chapter.
CHAPTER 8 is an introduction to the use of high voltage bipolar transistors in electronic lighting ballasts. Many of the
possible topologies are described.
Contents Power Semiconductor Applications
Philips Semiconductors
Table of Contents
CHAPTER 1 Introduction to Power Semiconductors 1
General 3
1.1.1 An Introduction To Power Devices ............................................................ 5
Power MOSFET 17
1.2.1 PowerMOS Introduction ............................................................................. 19
1.2.2 Understanding Power MOSFET Switching Behaviour ............................... 29
1.2.3 Power MOSFET Drive Circuits .................................................................. 39
1.2.4 Parallel Operation of Power MOSFETs ..................................................... 49
1.2.5 Series Operation of Power MOSFETs ....................................................... 53
1.2.6 Logic Level FETS ...................................................................................... 57
1.2.7 Avalanche Ruggedness ............................................................................. 61
1.2.8 Electrostatic Discharge (ESD) Considerations .......................................... 67
1.2.9 Understanding the Data Sheet: PowerMOS .............................................. 69
High Voltage Bipolar Transistor 77
1.3.1 Introduction To High Voltage Bipolar Transistors ...................................... 79
1.3.2 Effects of Base Drive on Switching Times ................................................. 83
1.3.3 Using High Voltage Bipolar Transistors ..................................................... 91
1.3.4 Understanding The Data Sheet: High Voltage Transistors ....................... 97
CHAPTER 2 Switched Mode Power Supplies 103
Using Power Semiconductors in Switched Mode Topologies 105
2.1.1 An Introduction to Switched Mode Power Supply Topologies ................... 107
2.1.2 The Power Supply Designer’s Guide to High Voltage Transistors ............ 129
2.1.3 Base Circuit Design for High Voltage Bipolar Transistors in Power
Converters ........................................................................................................... 141
2.1.4 Isolated Power Semiconductors for High Frequency Power Supply
Applications ......................................................................................................... 153
Output Rectification 159
2.2.1 Fast Recovery Epitaxial Diodes for use in High Frequency Rectification 161
2.2.2 Schottky Diodes from Philips Semiconductors .......................................... 173
2.2.3 An Introduction to Synchronous Rectifier Circuits using PowerMOS
Transistors ........................................................................................................... 179
i
Contents Power Semiconductor Applications
Philips Semiconductors
Design Examples 185
2.3.1 Mains Input 100 W Forward Converter SMPS: MOSFET and Bipolar
Transistor Solutions featuring ETD Cores ........................................................... 187
2.3.2 Flexible, Low Cost, Self-Oscillating Power Supply using an ETD34
Two-Part Coil Former and 3C85 Ferrite .............................................................. 199
Magnetics Design 205
2.4.1 Improved Ferrite Materials and Core Outlines for High Frequency Power
Supplies ............................................................................................................... 207
Resonant Power Supplies 217
2.5.1. An Introduction To Resonant Power Supplies .......................................... 219
2.5.2. Resonant Power Supply Converters - The Solution For Mains Pollution
Problems .............................................................................................................. 225
CHAPTER 3 Motor Control 241
AC Motor Control 243
3.1.1 Noiseless A.C. Motor Control: Introduction to a 20 kHz System ............... 245
3.1.2 The Effect of a MOSFET’s Peak to Average Current Rating on Invertor
Efficiency ............................................................................................................. 251
3.1.3 MOSFETs and FREDFETs for Motor Drive Equipment ............................. 253
3.1.4 A Designers Guide to PowerMOS Devices for Motor Control ................... 259
3.1.5 A 300V, 40A High Frequency Inverter Pole Using Paralleled FREDFET
Modules ............................................................................................................... 273
DC Motor Control 283
3.2.1 Chopper circuits for DC motor control ....................................................... 285
3.2.2 A switched-mode controller for DC motors ................................................ 293
3.2.3 Brushless DC Motor Systems .................................................................... 301
Stepper Motor Control 307
3.3.1 Stepper Motor Control ............................................................................... 309
CHAPTER 4 Televisions and Monitors 317
Power Devices in TV & Monitor Applications (including selection
guides) 319
4.1.1 An Introduction to Horizontal Deflection .................................................... 321
4.1.2 The BU25XXA/D Range of Deflection Transistors .................................... 331
ii
Contents Power Semiconductor Applications
Philips Semiconductors
4.1.3 Philips HVT’s for TV & Monitor Applications .............................................. 339
4.1.4 TV and Monitor Damper Diodes ................................................................ 345
TV Deflection Circuit Examples 349
4.2.1 Application Information for the 16 kHz Black Line Picture Tubes .............. 351
4.2.2 32 kHz / 100 Hz Deflection Circuits for the 66FS Black Line Picture Tube 361
SMPS Circuit Examples 377
4.3.1 A 70W Full Performance TV SMPS Using The TDA8380 ......................... 379
4.3.2 A Synchronous 200W SMPS for 16 and 32 kHz TV .................................. 389
Monitor Deflection Circuit Example 397
4.4.1 A Versatile 30 - 64 kHz Autosync Monitor ................................................. 399
CHAPTER 5 Automotive Power Electronics 421
Automotive Motor Control (including selection guides) 423
5.1.1 Automotive Motor Control With Philips MOSFETS .................................... 425
Automotive Lamp Control (including selection guides) 433
5.2.1 Automotive Lamp Control With Philips MOSFETS .................................... 435
The TOPFET 443
5.3.1 An Introduction to the 3 pin TOPFET ......................................................... 445
5.3.2 An Introduction to the 5 pin TOPFET ......................................................... 447
5.3.3 BUK101-50DL - a Microcontroller compatible TOPFET ............................ 449
5.3.4 Protection with 5 pin TOPFETs ................................................................. 451
5.3.5 Driving TOPFETs ....................................................................................... 453
5.3.6 High Side PWM Lamp Dimmer using TOPFET ......................................... 455
5.3.7 Linear Control with TOPFET ...................................................................... 457
5.3.8 PWM Control with TOPFET ....................................................................... 459
5.3.9 Isolated Drive for TOPFET ........................................................................ 461
5.3.10 3 pin and 5 pin TOPFET Leadforms ........................................................ 463
5.3.11 TOPFET Input Voltage ............................................................................ 465
5.3.12 Negative Input and TOPFET ................................................................... 467
5.3.13 Switching Inductive Loads with TOPFET ................................................. 469
5.3.14 Driving DC Motors with TOPFET ............................................................. 471
5.3.15 An Introduction to the High Side TOPFET ............................................... 473
5.3.16 High Side Linear Drive with TOPFET ...................................................... 475
iii
Contents Power Semiconductor Applications
Philips Semiconductors
Automotive Ignition 477
5.4.1 An Introduction to Electronic Automotive Ignition ...................................... 479
5.4.2 IGBTs for Automotive Ignition .................................................................... 481
5.4.3 Electronic Switches for Automotive Ignition ............................................... 483
CHAPTER 6 Power Control with Thyristors and Triacs 485
Using Thyristors and Triacs 487
6.1.1 Introduction to Thyristors and Triacs ......................................................... 489
6.1.2 Using Thyristors and Triacs ....................................................................... 497
6.1.3 The Peak Current Handling Capability of Thyristors .................................. 505
6.1.4 Understanding Thyristor and Triac Data .................................................... 509
Thyristor and Triac Applications 521
6.2.1 Triac Control of DC Inductive Loads .......................................................... 523
6.2.2 Domestic Power Control with Triacs and Thyristors .................................. 527
6.2.3 Design of a Time Proportional Temperature Controller ............................. 537
Hi-Com Triacs 547
6.3.1 Understanding Hi-Com Triacs ................................................................... 549
6.3.2 Using Hi-Com Triacs .................................................................................. 551
CHAPTER 7 Thermal Management 553
Thermal Considerations 555
7.1.1 Thermal Considerations for Power Semiconductors ................................. 557
7.1.2 Heat Dissipation ......................................................................................... 567
CHAPTER 8 Lighting 575
Fluorescent Lamp Control 577
8.1.1 Efficient Fluorescent Lighting using Electronic Ballasts ............................. 579
8.1.2 Electronic Ballasts - Philips Transistor Selection Guide ............................ 587
8.1.3 An Electronic Ballast - Base Drive Optimisation ........................................ 589
iv
Index Power Semiconductor Applications
Philips Semiconductors
Index
Airgap, transformer core, 111, 113
Anti saturation diode, 590
Asynchronous, 497
Automotive
fans
see motor control
IGBT, 481, 483
ignition, 479, 481, 483
lamps, 435, 455
motor control, 425, 457, 459, 471, 475
resistive loads, 442
reverse battery, 452, 473, 479
screen heater, 442
seat heater, 442
solenoids, 469
TOPFET, 473
Avalanche, 61
Avalanche breakdown
thyristor, 490
Avalanche multiplication, 134
Baker clamp, 138, 187, 190
Ballast
electronic, 580
fluorescent lamp, 579
switchstart, 579
Base drive, 136
base inductor, 147
base inductor, diode assisted, 148
base resistor, 146
drive transformer, 145
drive transformer leakage inductance, 149
electronic ballast, 589
forward converter, 187
power converters, 141
speed-up capacitor, 143
Base inductor, 144, 147
Base inductor, diode assisted, 148
Boost converter, 109
continuous mode, 109
discontinuous mode, 109
output ripple, 109
Bootstrap, 303
Breakback voltage
diac, 492
Breakdown voltage, 70
Breakover current
diac, 492
Breakover voltage
diac, 492, 592
thyristor, 490
Bridge circuits
see Motor Control - AC
Brushless motor, 301, 303
Buck-boost converter, 110
Buck converter, 108 - 109
Burst firing, 537
Burst pulses, 564
Capacitance
junction, 29
Capacitor
mains dropper, 544
CENELEC, 537
Charge carriers, 133
triac commutation, 549
Choke
fluorescent lamp, 580
Choppers, 285
Clamp diode, 117
Clamp winding, 113
Commutation
diode, 164
Hi-Com triac, 551
thyristor, 492
triac, 494, 523, 529
Compact fluorescent lamp, 585
Continuous mode
see Switched Mode Power Supplies
Continuous operation, 557
Converter (dc-dc)
switched mode power supply, 107
Cookers, 537
Cooling
forced, 572
natural, 570
Crest factor, 529
Critical electric field, 134
Cross regulation, 114, 117
Current fed resonant inverter, 589
Current Mode Control, 120
Current tail, 138, 143
Damper Diodes, 345, 367
forward recovery, 328, 348
losses, 347
outlines, 345
picture distortion, 328, 348
selection guide, 345
Darlington, 13
Data Sheets
High Voltage Bipolar Transistor, 92,97,331
MOSFET, 69
i
Index Power Semiconductor Applications
Philips Semiconductors
dc-dc converter, 119
Depletion region, 133
Desaturation networks, 86
Baker clamp, 91, 138
dI/dt
triac, 531
Diac, 492, 500, 527, 530, 591
Diode, 6
double diffused, 162
epitaxial, 161
schottky, 173
structure, 161
Diode Modulator, 327, 367
Disc drives, 302
Discontinuous mode
see Switched Mode Power Supplies
Domestic Appliances, 527
Dropper
capacitive, 544
resistive, 544, 545
Duty cycle, 561
EFD core
see magnetics
Efficiency Diodes
see Damper Diodes
Electric drill, 531
Electronic ballast, 580
base drive optimisation, 589
current fed half bridge, 584, 587, 589
current fed push pull, 583, 587
flyback, 582
transistor selection guide, 587
voltage fed half bridge, 584, 588
voltage fed push pull, 583, 587
EMC, 260, 455
see RFI, ESD
TOPFET, 473
Emitter shorting
triac, 549
Epitaxial diode, 161
characteristics, 163
dI/dt, 164
forward recovery, 168
lifetime control, 162
operating frequency, 165
passivation, 162
reverse leakage, 169
reverse recovery, 162, 164
reverse recovery softness, 167
selection guide, 171
snap-off, 167
softness factor, 167
stored charge, 162
technology, 162
ESD, 67
see Protection, ESD
precautions, 67
ETD core
see magnetics
F-pack
see isolated package
Fall time, 143, 144
Fast Recovery Epitaxial Diode (FRED)
see epitaxial diode
FBSOA, 134
Ferrites
see magnetics
Flicker
fluorescent lamp, 580
Fluorescent lamp, 579
colour rendering, 579
colour temperature, 579
efficacy, 579, 580
triphosphor, 579
Flyback converter, 110, 111, 113
advantages, 114
clamp winding, 113
continuous mode, 114
coupled inductor, 113
cross regulation, 114
diodes, 115
disadvantages, 114
discontinuous mode, 114
electronic ballast, 582
leakage inductance, 113
magnetics, 213
operation, 113
rectifier circuit, 180
self oscillating power supply, 199
synchronous rectifier, 156, 181
transformer core airgap, 111, 113
transistors, 115
Flyback converter (two transistor), 111, 114
Food mixer, 531
Forward converter, 111, 116
advantages, 116
clamp diode, 117
conduction loss, 197
continuous mode, 116
core loss, 116
core saturation, 117
cross regulation, 117
diodes, 118
disadvantages, 117
duty ratio, 117
ferrite cores, 116
magnetics, 213
magnetisation energy, 116, 117
ii
Index Power Semiconductor Applications
Philips Semiconductors
operation, 116
output diodes, 117
output ripple, 116
rectifier circuit, 180
reset winding, 117
switched mode power supply, 187
switching frequency, 195
switching losses, 196
synchronous rectifier, 157, 181
transistors, 118
Forward converter (two transistor), 111, 117
Forward recovery, 168
FREDFET, 250, 253, 305
bridge circuit, 255
charge, 254
diode, 254
drive, 262
loss, 256
reverse recovery, 254
FREDFETs
motor control, 259
Full bridge converter, 111, 125
advantages, 125
diodes, 126
disadvantages, 125
operation, 125
transistors, 126
Gate
triac, 538
Gate drive
forward converter, 195
Gold doping, 162, 169
GTO, 11
Guard ring
schottky diode, 174
Half bridge, 253
Half bridge circuits
see also Motor Control - AC
Half bridge converter, 111, 122
advantages, 122
clamp diodes, 122
cross conduction, 122
diodes, 124
disadvantages, 122
electronic ballast, 584, 587, 589
flux symmetry, 122
magnetics, 214
operation, 122
synchronous rectifier, 157
transistor voltage, 122
transistors, 124
voltage doubling, 122
Heat dissipation, 567
Heat sink compound, 567
Heater controller, 544
Heaters, 537
Heatsink, 569
Heatsink compound, 514
Hi-Com triac, 519, 549, 551
commutation, 551
dIcom/dt, 552
gate trigger current, 552
inductive load control, 551
High side switch
MOSFET, 44, 436
TOPFET, 430, 473
High Voltage Bipolar Transistor, 8, 79, 91,
141, 341
‘bathtub’ curves, 333
avalanche breakdown, 131
avalanche multiplication, 134
Baker clamp, 91, 138
base-emitter breakdown, 144
base drive, 83, 92, 96, 136, 336, 385
base drive circuit, 145
base inductor, 138, 144, 147
base inductor, diode assisted, 148
base resistor, 146
breakdown voltage, 79, 86, 92
carrier concentration, 151
carrier injection, 150
conductivity modulation, 135, 150
critical electric field, 134
current crowding, 135, 136
current limiting values, 132
current tail, 138, 143
current tails, 86, 91
d-type, 346
data sheet, 92, 97, 331
depletion region, 133
desaturation, 86, 88, 91
device construction, 79
dI/dt, 139
drive transformer, 145
drive transformer leakage inductance, 149
dV/dt, 139
electric field, 133
electronic ballast, 581, 585, 587, 589
Fact Sheets, 334
fall time, 86, 99, 143, 144
FBSOA, 92, 99, 134
hard turn-off, 86
horizontal deflection, 321, 331, 341
leakage current, 98
limiting values, 97
losses, 92, 333, 342
Miller capacitance, 139
operation, 150
iii
Index Power Semiconductor Applications
Philips Semiconductors
optimum drive, 88
outlines, 332, 346
over current, 92, 98
over voltage, 92, 97
overdrive, 85, 88, 137, 138
passivation, 131
power limiting value, 132
process technology, 80
ratings, 97
RBSOA, 93, 99, 135, 138, 139
RC network, 148
reverse recovery, 143, 151
safe operating area, 99, 134
saturation, 150
saturation current, 79, 98, 341
secondary breakdown, 92, 133
smooth turn-off, 86
SMPS, 94, 339, 383
snubber, 139
space charge, 133
speed-up capacitor, 143
storage time, 86, 91, 92, 99, 138, 144, 342
sub emitter resistance, 135
switching, 80, 83, 86, 91, 98, 342
technology, 129, 149
thermal breakdown, 134
thermal runaway, 152
turn-off, 91, 92, 138, 142, 146, 151
turn-on, 91, 136, 141, 149, 150
underdrive, 85, 88
voltage limiting values, 130
Horizontal Deflection, 321, 367
base drive, 336
control ic, 401
d-type transistors, 346
damper diodes, 345, 367
diode modulator, 327, 347, 352, 367
drive circuit, 352, 365, 406
east-west correction, 325, 352, 367
line output transformer, 354
linearity correction, 323
operating cycle, 321, 332, 347
s-correction, 323, 352, 404
TDA2595, 364, 368
TDA4851, 400
TDA8433, 363, 369
test circuit, 321
transistors, 331, 341, 408
waveforms, 322
IGBT, 11, 305
automotive, 481, 483
clamped, 482, 484
ignition, 481, 483
Ignition
automotive, 479, 481, 483
darlington, 483
Induction heating, 53
Induction motor
see Motor Control - AC
Inductive load
see Solenoid
Inrush current, 528, 530
Intrinsic silicon, 133
Inverter, 260, 273
see motor control ac
current fed, 52, 53
switched mode power supply, 107
Irons, electric, 537
Isolated package, 154
stray capacitance, 154, 155
thermal resistance, 154
Isolation, 153
J-FET, 9
Junction temperature, 470, 557, 561
burst pulses, 564
non-rectangular pulse, 565
rectangular pulse, composite, 562
rectangular pulse, periodic, 561
rectangular pulse, single shot, 561
Lamp dimmer, 530
Lamps, 435
dI/dt, 438
inrush current, 438
MOSFET, 435
PWM control, 455
switch rate, 438
TOPFET, 455
Latching current
thyristor, 490
Leakage inductance, 113, 200, 523
Lifetime control, 162
Lighting
fluorescent, 579
phase control, 530
Logic Level FET
motor control, 432
Logic level MOSFET, 436
Magnetics, 207
100W 100kHz forward converter, 197
100W 50kHz forward converter, 191
50W flyback converter, 199
core losses, 208
core materials, 207
EFD core, 210
ETD core, 199, 207
iv
Thermal Management Power Semiconductor Applications
Philips Semiconductors
CHAPTER 7
Thermal Management
7.1 Thermal Considerations
553
Thermal Management Power Semiconductor Applications
Philips Semiconductors
Thermal Considerations
555
Thermal Management Power Semiconductor Applications
Philips Semiconductors
7.1.1 Thermal Considerations for Power Semiconductors
The perfect power switch is not yet available. All power
semiconductors dissipate power internally both during the
on-state and during the transition between the on and off
states. The amount of power dissipated internally generally
speaking increases in line with the power being switched
by the semiconductor. The capability of a switch to operate
in a particular circuit will therefore depend upon the amount
of power dissipated internally and the rise in the operating
temperature of the silicon junction that this power
dissipation causes. It is therefore important that circuit
designers are familiar with the thermal characteristics of
power semiconductors and are able to calculate power
dissipation limits and junction operating temperatures.
This chapter is divided into two parts. Part One describes
the essential thermal properties of semiconductors and
explains the concept of a limit in terms of continuous mode
and pulse mode operation. Part Two gives worked
examples showing junction temperature calculations for a
variety of applied power pulse waveforms.
PART ONE
The power dissipation limit
The maximum allowable power dissipation forms a limit to
the safe operating area of power transistors. Power
dissipation causes a rise in junction temperature which will,
in turn, start chemical and metallurgical changes. The rate
at which these changes proceed is exponentially related to
temperature, and thus prolonged operation of a power
transistor above its junction temperature rating is liable to
result in reduced life. Operation of a device at, or below, its
power dissipation rating (together with careful
consideration of thermal resistances associated with the
device) ensures that the junction temperature rating is not
exceeded.
All power semiconductors have a power dissipation
limitation. For rectifier products such as diodes, thyristors
and triacs, the power dissipation rating can be easily
translated in terms of current ratings; in the on-state the
voltage drop is well defined. Transistors are, however,
somewhat more complicated. A transistor, be it a power
MOSFET or a bipolar, can operate in its on-state at any
voltage up to its maximum rating depending on the circuit
conditions. It is therefore necessary to specify a Safe
Operating Area (SOA) for transistors which specifies the
power dissipation limit in terms of a series of boundaries in
the current and voltage plane. These operating areas are
usuallypresentedfor mountingbasetemperaturesof 25 ˚C.
At higher temperatures, operating conditions must be
checked to ensure that junction temperatures are not
exceeding the desired operating level.
Continuous power dissipation
The total power dissipation in a semiconductor may be
calculated from the product of the on-state voltage and the
forward conduction current. The heat dissipated in the
junction of the device flows through the thermal resistance
between the junction and the mounting base, R
thj-mb
. The
thermal equivalent circuit of Fig. 1 illustrates this heat flow;
P
tot
can be regarded as a thermal current, and the
temperature difference between the junction and mounting
base ∆T
j-mb
as a thermal voltage. By analogy with Ohm’s
law, it follows that:
Fig. 1 Heat transport in a transistor with power
dissipation constant with respect to time
Fig. 2 shows the dependence of the maximum power
dissipation on the temperature of the mounting base. P
totmax
is limited either by a maximum temperature difference:
or by the maximum junction temperature T
jmax
(T
mb K
is
usually 25˚C and is the value of T
mb
above which the
maximum power dissipation must be reduced to maintain
the operating point within the safe operating area).
In the first case, T
mb
≤ T
mb K
:
that is, the power dissipation has a fixed limit value (P
tot max K
is the maximum d.c. power dissipation below T
mb K
). If the
transistor is subjected to a mounting-base temperature
T
mb 1
, its junction temperature will be less than T
jmax
by an
amount (T
mbK
- T
mb 1
), as shown by the broken line in Fig. 2.
P
tot
=
T
j
− T
mb
R
thj − mb
1
∆T
j − mb max
= T
jmax
− T
mb K
2
P
tot maxK
=
∆T
j − mb max
R
thj − mb
; 3
557
Thermal Management Power Semiconductor Applications
Philips Semiconductors
Fig. 2 Maximum d.c. power dissipation in a transistor
as a function of the mounting-base temperature
In the second case, T
mb
> T
mb K
:
that is, the power dissipation must be reduced as the
mounting base temperature increases along the sloping
straight line in Fig. 2. Equation 4 shows that the lower the
thermal resistance R
thj-mb
, the steeper is the slope of the
line. In this case, T
mb
is the maximum mounting-base
temperature that can occur in operation.
Example
The following data is provided for a particular transistor.
P
tot maxK
= 75 W
T
jmax
= 175 ˚C
R
thj-mb
≤ 2 K/W
Themaximumpermissiblepower dissipationfor continuous
operation at a maximum mounting-base temperature of
T
mb
= 80 ˚C is required.
Note that the maximum value of T
mb
is chosen to be
significantlyhigher thanthe maximumambient temperature
to prevent an excessively large heatsink being required.
From Eq. 4 we obtain:
Provided that the transistor is operated within SOA limits,
this value is permissible since it is belowP
tot max K
. The same
result can be obtained graphically from the P
tot max
diagram
(Fig. 3) for the relevant transistor.
Fig. 3 Example of the determination of maximum power
dissipation
Pulse power operation
When a power transistor is subjected to a pulsed load,
higher peak power dissipation is permitted. The materials
in a power transistor have a definite thermal capacity, and
thus the critical junction temperature will not be reached
instantaneously, even when excessive power is being
dissipated in the device. The power dissipation limit may
be extended for intermittent operation. The size of the
extension will depend on the duration of the operation
period(that is, pulseduration) and the frequency with which
operation occurs (that is, duty factor).
Fig. 4 Heating of a transistor chip
If power isappliedtoatransistor, thedevice will immediately
start to warmup (Fig. 4). If the power dissipation continues,
a balance will be struck between heat generation and
removal resulting in the stabilisation of T
j
and ∆T
j-mb
. Some
heat energy will be stored by the thermal capacity of the
device, and the stable conditions will be determined by the
thermal resistances associated with the transistor and its
thermal environment. When the power dissipation ceases,
the device will cool (the heating and cooling laws will be
identical, see Fig. 5). However, if the power dissipation
ceases before the temperature of the transistor stabilises,
the peak values of T
j
and ∆T
j-mb
will be less than the values
0 20 40 60 80 100 120 140 160 180
Tmb / C
Ptotmax / W
100
90
80
70
60
50
40
30
20
10
0
47.5
P
tot max
=
T
jmax
− T
mb
R
thj − mb
; 4
P
tot max
=
175−80
2
W, =47.5W
558
Thermal Management Power Semiconductor Applications
Philips Semiconductors
reached for the same level of continuous power dissipation
(Fig. 6). If the second pulse is identical to the first, the peak
temperatureattained by the device at the end of the second
pulse will be greater than that at the end of the first pulse.
Further pulses will build up the temperature until some new
stable situation is attained (Fig. 7). The temperature of the
deviceinthis stablecondition will fluctuate aboveandbelow
the mean. If the upward excursions extend into the region
of excessive T
j
then the life expectancy of the device may
be reduced. This can happen with high-power
low-duty-factor pulses, even though the average power is
below the d.c. rating of the device.
Fig. 5 Heating and cooling follow the same law
Fig. 6 The peak temperature caused by a short power
pulse can be less than the steady-state temperature
resulting from the same power
Fig. 8 shows a typical safe operating area for d.c. operation
of apower MOSFET. The corresponding rectangular-pulse
operatingareas with afixed duty factor, δ = 0, and the pulse
time t
p
as a parameter, are also shown. These boundaries
represent the largest possible extension of the operating
area for particular pulse times. When the pulse time
becomes very short, the power dissipation does not have
alimitingactionandthe pulsecurrent andmaximumvoltage
form the only limits. This rectangle represents the largest
possible pulse operating area.
Fig. 7 A train of power pulses increases the average
temperature if the device does not have time to cool
between pulses
Fig. 8 D.C. and rectangular pulse operating areas with
fixed parameters δ=0, t
p
and T
mb
=25˚C
In general, the shorter the pulse and the lower the
frequency, the lower the temperature that the junction
reaches. By analogy with Eq. 3, it follows that:
where Z
thj-mb
is the transient thermal impedance between
the junction and mounting base of the device. It depends
on the pulse duration t
p
, and the duty factor δ, where:
1 100
VDS / V
ID / A
100
10
1
0.1
10 us
100 us
1 ms
10 ms
R
D
S
(
O
N
)

=

V
D
S
/
I
D
100 ms
DC
tp =
BUK553-100
10
B
A
P
tot M
=
T
j
−T
mb
Z
thj − mb
, 5
559
Thermal Management Power Semiconductor Applications
Philips Semiconductors
and T is the pulse period. Fig. 9 shows a typical family of
curves for thermal impedance against pulse duration, with
duty factor as a parameter.
Fig. 9 Example of the presentation of the transient
thermal impedance as a function of the pulse time with
duty factor as parameter
Again, the maximum pulse power dissipation is limited
either by the maximum temperature difference ∆T
j-mb max
(Eq. 2), or by the maximumjunction temperature T
jmax
, and
so by analogy with Eqs. 3 and 4:
when T
mb
≤ T
mb K
, and:
when T
mb
> T
mb K
. That is, below a mounting-base
temperature of T
mb K
, the maximum power dissipation has
a fixed limit value; and above T
mb K
, the power dissipation
must be reduced linearly with increasing mounting-base
temperature.
Short pulse duration (Fig. 10a)
As the pulse duration becomes very short, the fluctuations
of junction temperature become negligible, owing to the
internal thermal capacity of the transistor. Consequently,
theonly factor to beconsidered isthe heating of the junction
by the average power dissipation; that is:
The transient thermal impedance becomes:
The Z
thj-mb
curves approach this value asymptotically as t
p
decreases. Fig. 9 shows that, for duty factors in the range
0.1 to 0.5, the limit values given by Eq. 10 have virtually
been reached at t
p
= 10
-6
s.
Fig. 10 Three limit cases of rectangular pulse loads:
(a) short pulse duration
(b) long pulse duration
(c) single-shot pulse
Long pulse duration (Fig. 10b)
As the pulse duration increases, the junction temperature
approaches a stationary value towards the end of a pulse.
The transient thermal impedance tends to the thermal
resistance for continuous power dissipation; that is:
Fig. 9showsthat Z
thj-mb
approachesthis valueast
p
becomes
large. In general, transient thermal effects die out in most
power transistors within 0.1 to 1.0 seconds. This time
depends on the material and construction of the case, the
size of the chip, the way it is mounted, and other factors.
Power pulses with a duration in excess of this time have
approximately the same effect as a continuous load.
δ=
t
p
T
, 6
lim
t
p
→0
Z
thj − mb
=δR
thj − mb
10
1E-07 1E-05 1E-03 1E-01 1E+01
t / s
Zth j-mb / (K/W)
1E+01
1E+00
1E-01
1E-02
1E-03
0
0.5
0.2
0.1
0.05
0.02
=
tp
tp
T
T
P
t
D
=
BUK553-100A
P
tot maxK
=
∆T
j − mb max
Z
thj − mb
, 7
P
tot M max
=
T
jmax
−T
mb
Z
thj − mb
, 8
lim
t
p
→∞
Z
thj − mb
=R
thj − mb
11
P
tot(av)
= δP
tot M
9
560
Thermal Management Power Semiconductor Applications
Philips Semiconductors
Single-shot pulses (Fig. 10c)
As the duty factor becomes very small, the junction tends
to cool down completely between pulses so that each pulse
can be treated individually. When considering single
pulses, the Z
thj-mb
values for δ = 0 (Fig. 9) give sufficiently
accurate results.
PART TWO
Calculating junction temperatures
Most applications which include power semiconductors
usually involve some form of pulse mode operation. This
section gives several worked examples showing how
junction temperatures can be simply calculated. Examples
are given for a variety of waveforms:
(1) Periodic Waveforms
(2) Single Shot Waveforms
(3) Composite Waveforms
(4) A Pulse Burst
(5) Non Rectangular Pulses
From the point of view of reliability it is most important to
know what the peak junction temperature will be when the
power waveform is applied and also what the average
junction temperature is going to be.
Peak junction temperature will usually occur at the end of
an applied pulse and its calculation will involve transient
thermal impedance. The average junction temperature
(whereapplicable) is calculated by working out the average
power dissipation using the d.c. thermal resistance.
Fig. 11 Periodic Rectangular Pulse
When considering the junction temperature in a device, the
following formula is used:
where ∆T
j-mb
is found from a rearrangement of equation 7.
In all the following examples the mounting base
temperature (T
mb
) is assumed to be 75˚C.
Periodic rectangular pulse
Fig. 11 shows an example of a periodic rectangular pulse.
This type of pulse is commonly found in switching
applications. 100W is dissipated every 400µs for a period
of 20µs, representing a duty cycle (δ) of 0.05. The peak
junction temperature is calculated as follows:
The value for Z
th j-mb
is taken from the δ=0.05 curve shown
in Fig. 12 (This diagram repeats Fig. 9 but has been
simplified for clarity). The above calculation shows that the
peak junction temperature will be 85˚C.
Single shot rectangular pulse
Fig. 13 shows an example of a single shot rectangular
pulse. The pulse used is the same as in the previous
example, which should highlight the differences between
periodic and single shot thermal calculations. For a single
shot pulse, the time period between pulses is infinity, ie the
duty cycle δ=0. In this example 100W is dissipated for a
period of 20µs. To work out the peak junction temperature
the following steps are used:
The value for Z
th j-mb
is taken from the δ=0 curve shown in
Fig. 12. The above calculation shows that the peak junction
temperature will be 4˚C above the mounting base
temperature.
Peak T
j
: t = 2 × 10
−5
s
P = 100W
δ =
20
400
= 0.05
Z
thj −mb
= 0.12K/W
∆T
j −mb
= P × Z
thj −mb
= 100 × 0.12 = 12°C
T
j
= T
mb
+ ∆T
j −mb
= 75 + 12 = 87°C
Average T
j
: P
av
= P × δ = 100 × 0.05 = 5W
∆T
j −mb(av)
= P
av
× Z
thj −mb(δ = 1)
= 5 × 2 = 10°C
T
j(av)
= T
mb
+ ∆T
j −mb(av)
= 75 + 10 = 85°C
0
10
20
30
40
50
60
70
80
90
100
110
0 20 40 60 80 100 400 420 440 460 480
POWER(W)
TIME(uS)
0 20 40 60 80 100 400 420 440 460 480
TIME(uS)
Tmb
Tjpeak
Tj t = 2 × 10
−5
s
P = 100W
δ = 0
Z
thj −mb
= 0.04K/W
∆T
j −mb
= P × Z
thj −mb
= 100 × 0.04 = 4°C
T
j
= T
mb
+ ∆T
j −mb
14
561
Thermal Management Power Semiconductor Applications
Philips Semiconductors
Fig. 12 Thermal impedance curves for δ=0.05 and δ=0
Fig. 13 Single Shot Pulse
For a single shot pulse, the average power dissipated and
average junction temperature are not relevant.
Composite rectangular pulse
In practice, a power device frequently has to handle
composite waveforms, rather than the simple rectangular
pulses shown so far. This type of signal can be simulated
by superimposing several rectangular pulses which have a
commonperiod, but both positive and negative amplitudes,
in addition to suitable values of t
p
and δ.
By way of an example, consider the composite waveform
shown in Fig. 14. To show the way in which the method
used for periodic rectangular pulses is extended to cover
composite waveforms, the waveform shown has been
chosen to be an extension of the periodic rectangular pulse
example. The period is 400µs, and the waveform consists
of three rectangular pulses, namely 40Wfor 10µs, 20Wfor
150µs and 100Wfor 20µs. The peak junction temperature
may be calculated at any point in the cycle. To be able to
add the various effects of the pulses at this time, all the
pulses, both positive and negative, must end at time t
x
in
the first calculation andt
y
inthe secondcalculation. Positive
pulses increase the junction temperature, while negative
pulses decrease it.
Calculation for time t
x
In equation 15, the values for P
1
, P
2
and P
3
are known:
P
1
=40W, P
2
=20Wand P
3
=100W. The Z
th
values are taken
from Fig. 9. For each term in the equation, the equivalent
duty cycle must be worked out. For instance the first
superimposed pulse in Fig. 14 lasts for a time t1 = 180µs,
representing a duty cycle of 180/400 = 0.45 = δ. These
values can then be used in conjunction with Fig. 9 to find a
value for Z
th
, which in this case is 0.9K/W. Table 1a gives
the values calculated for this example.
t1 t2 t3 t4
180µs 170µs 150µs 20µs
Repetitive δ 0.450 0.425 0.375 0.050
T=400µs Z
th
0.900 0.850 0.800 0.130
Single Shot δ 0.000 0.000 0.000 0.000
T=∞ Z
th
0.130 0.125 0.120 0.040
Table 1a. Composite pulse parameters for time t
x
1E-07 1E-05 1E-03 1E-01 1E+01
t / s
Zth j-mb / (K/W)
1E+01
1E+00
1E-01
1E-02
1E-03
0
0.05
=
tp
tp
T
T
P
t
D
=
0.04
0.12
0
10
20
30
40
50
60
70
80
90
100
110
0 20 40 60 80 100
POWER(W)
0 20 40 60 80 100
Tmb
Tjpeak
Tj
∆T
j −mb@x
= P
1
.Z
thj −mb(t1)
+ P
2
.Z
thj −mb(t3)
+ P
3
.Z
thj −mb(t4)
− P
1
.Z
thj −mb(t2)
− P
2
.Z
thj −mb(t4)
15
562
Thermal Management Power Semiconductor Applications
Philips Semiconductors
Fig. 14 Periodic Composite Waveform
0 20 40 60 80 100
0
20
40
60
80
100
200
180
160
140
120
100
80
60
40
20
0
20
40
60
120 140 160 180 200 220
P3
P2
P1
P1
P2
t1
t2
t3
t4
360 380 400 420 440 460 480 500
tx ty POWER (W)
Time (uS)
160
140
120
100
80
60
40
20
0
20
40
60
80
100
t5
t6
t7
t8
120
P2
P3
P1
P2
P3
POWER (W)
POWER (W)
Time (uS)
Time (uS)
Substituting these values into equation 15 for T
j-mb@x
gives
Hence the peak values of T
j
are 104.4˚C for the repetitive
case, and 80.9˚C for the single shot case.
Single Shot: ∆T
j −mb@x
= 40 × 0.13 + 20 × 0.125
+ 100 × 0.04 − 40 × 0.125
− 20 × 0.04
= 5.9°C
T
j
= T
mb
+ ∆T
j −mb
= 75 + 5.9 = 80.9°C
Repetitive: ∆T
j −mb@x
= 40 × 0.9 + 20 × 0.85
+ 100 × 0.13 − 40 × 0.85
− 20 × 0.13
= 29.4°C
T
j
= T
mb
+ ∆T
j −mb
= 75 + 29.4 = 104.4°C
563
Thermal Management Power Semiconductor Applications
Philips Semiconductors
Calculation for time t
y
whereZ
thj-mb(t)
is the transient thermal impedance for a pulse
time t.
t5 t6 t7 t8
380µs 250µs 230µs 10µs
Repetitive δ 0.950 0.625 0.575 0.025
T=400µs Z
th
1.950 1.300 1.250 0.080
Single Shot δ 0.000 0.000 0.000 0.000
T=∞ Z
th
0.200 0.160 0.150 0.030
Table 1b. Composite pulse parameters for time t
y
Substituting these values into equation 16 for T
j-mb@y
gives
Hence the peak values of T
j
are 96.2˚C for the repetitive
case, and 78˚C for the single shot case.
The average power dissipation and the average junction
temperature can be calculated as follows:
Clearly, the junction temperature at time t
x
should be higher
than that at time t
y
, and this is proven in the above
calculations.
Burst pulses
Power devices are frequently subjectedto aburst of pulses.
This type of signal can be treated as a composite waveform
andasintheprevious examplesimulatedbysuperimposing
several rectangular pulses which have a common period,
but both positive and negative amplitudes, in addition to
suitable values of t
p
and δ.
Fig. 15 Burst Mode Waveform
Consider the waveform shown in Fig. 15. The period is
240µs, and the burst consists of three rectangular pulses
of 100Wpower and 20µs duration, separated by 30µs. The
peakjunctiontemperaturewill occur at the endof eachburst
at time t = t
x
= 140µs. To be able to add the various effects
of the pulses at this time, all the pulses, both positive and
negative, must end at time t
x
. Positive pulses increase the
junction temperature, while negative pulses decrease it.
whereZ
thj-mb(t)
is the transient thermal impedance for a pulse
time t.
The Z
th
values are taken from Fig. 9. For each term in the
equation, the equivalent duty cycle must be worked out.
These values can then be used in conjunction with Fig. 9
to find a value for Z
th
. Table 2 gives the values calculated
for this example.
t1 t2 t3 t4 t5
120µs 100µs 70µs 50µs 20µs
Repetitive δ 0.500 0.420 0.290 0.210 0.083
T=240µs Z
th
1.100 0.800 0.600 0.430 0.210
Single Shot δ 0.000 0.000 0.000 0.000 0.000
T=∞ Z
th
0.100 0.090 0.075 0.060 0.040
Table 2. Burst Mode pulse parameters
Substituting these values into equation 17 gives
∆T
j −mb@y
= P
2
.Z
thj −mb(t5)
+ P
3
.Z
thj −mb(t6)
+ P
1
.Z
thj −mb(t8)
− P
2
.Z
thj −mb(t6)
− P
3
.Z
thj −mb(t7)
16
0
50
100
150
350
300
250
200
150
100
50
0
50
100
150
200
250
300
350
0 20 40 60 80 100 120 140 160 240 260 280 300
T=240us
POWER (W)
Time (us)
Time (us)
t4
t2
t1
t3
t5
Repetitive: ∆T
j −mb@y
= 20 × 1.95 + 100 × 1.3
+ 40 × 0.08 − 20 × 1.3
− 100 × 1.25
= 21.2°C
T
j
= T
mb
+ ∆T
j −mb
= 75 + 21.2 = 96.2°C
Single Shot: ∆T
j −mb@y
= 20 × 0.2 + 100 × 0.16
+ 40 × 0.03 − 20 × 0.16
− 100 × 0.15
= 3°C
T
j
= T
mb
+ ∆T
j −mb
= 75 + 3 = 78°C
∆T
j −mb@x
= P.Z
thj −mb(t1)
+ P.Z
thj −mb(t3)
+ P.Z
thj −mb(t5)
− P.Z
thj −mb(t2)
− P.Z
thj −mb(t4)
17
P
av
=
25 × 10 + 5 × 130 + 20 × 100
400
= 7.25W
∆T
j −mb(av)
= P
av
× Z
thj −mb(δ = 1)
= 7.25 × 2 = 14.5°C
T
j(av)
= T
mb
+ ∆T
j −mb(av)
= 75 + 14.5 = 89.5°C
564
Thermal Management Power Semiconductor Applications
Philips Semiconductors
Hence the peak value of T
j
is 143˚C for the repetitive case
and 81.5˚C for the single shot case. To calculate the
average junction temperature T
j(av)
:
The above example for the repetitive waveform highlights
a case where the average junction temperature (125˚C) is
well within limits but the composite pulse calculation shows
the peak junction temperatureto be significantly higher. For
reasons of improved long term reliability it is usual to
operate devices with a peak junction temperature below
125˚C.
Non-rectangular pulses
Sofar, the worked examples have only coveredrectangular
waveforms. However, triangular, trapeziodal and
sinusoidal waveforms are also common. In order to apply
the above thermal calculations to non rectangular
waveforms, the waveform is approximated by a series of
rectangles. Each rectangle represents part of the
waveform. The equivalent rectangle must be equal in area
to the section of the waveform it represents (ie the same
energy) andalsobeof thesamepeakpower. Withreference
to Fig. 16, a triangular waveform has been approximated
to one rectangle in the first example, and two rectangles in
the second. Obviously, increasing the number of sections
the waveform is split into will improve the accuracy of the
thermal calculations.
In the first example, there is only one rectanglular pulse ,
of duration 50µs, dissipating 50W. So again using equation
14 and a rearrangement of equation 7:
Fig. 16 Non Rectangular Waveform
When the waveform is split into two rectangular pulses:
For this example P
1
= 25W, P
2
= 25W, P
3
= 50W. Table 3
below shows the rest of the parameters:
t1 t2 t3
75µs 50µs 37.5µs
Single Shot D 0.000 0.000 0.000
T=∞ Z
th
0.085 0.065 0.055
10% Duty Cycle D 0.075 0.050 0.037
T=1000µs Z
th
0.210 0.140 0.120
50% Duty Cycle D 0.375 0.250 0.188
T=200µs Z
th
0.700 0.500 0.420
Table 3. Non Rectangular Pulse Calculations
Repetitive: ∆T
j −mb@x
= 100 × 1.10 + 100 × 0.60
+ 100 × 0.21 − 100 × 0.80
− 100 × 0.43
= 68°C
T
j
= 75 + 68 = 143°C
0 20 40 60 80 100
0
10
20
30
40
50
100
90
80
70
60
50
40
30
20
10
0
10
20
30
0 20 40 60 80 100
0
10
20
30
40
50
100
90
80
70
60
50
40
30
20
10
0
10
20
30
P3
P2
P1
t3
t2
t1
Single Shot: ∆T
j −mb@x
= 100 × 0.10 + 100 × 0.075
+ 100 × 0.04 − 100 × 0.09
− 100 × 0.06
= 6.5°C
T
j
= 75 + 6.5 = 81.5°C
P
av
=
3 × 100 × 20
240
= 25W
∆T
j −mb(av)
= P
av
× Z
thj −mb(δ = 1)
= 25 × 2 = 50°C
T
j(av)
= 75 + 50 = 125°C
∆T
j −mb
= P
tot M
× Z
thj −mb
Single Shot ∆T
j −mb
= 50 × 0.065 = 3.25°C
T
jpeak
= 75 + 3.25 = 78.5°C
10% Duty cycle ∆T
j −mb
= 50 × 0.230 = 11.5°C
T
jpeak
= 75 + 11.5 = 86.5°C
50% Duty cycle ∆T
j −mb
= 50 × 1.000 = 50°C
T
jpeak
= 75 + 50 = 125°C
∆T
j −mb
= P
3
.Z
thj −mb(t3)
+ P
1
.Z
thj −mb(t1)
− P
2
.Z
thj −mb(t2)
18
565
Thermal Management Power Semiconductor Applications
Philips Semiconductors
Sustituting these values into equation 18 gives:
To calculate the average junction temperature:
Conclusion
A method has been presented to allow the calculation of
average and peak junction temperatures for a variety of
pulse types. Several worked examples have shown
calculations for various common waveforms. The method
for non rectangular pulses can be applied to any wave
shape, allowing temperature calculations for waveforms
such as exponential and sinusoidal power pulses. For
pulses such as these, care must be taken to ensure that
the calculation gives the peak junction temperature, as it
may not occur at the end of the pulse. In this instance
several calculations must be performed with different
endpoints to find the maximum junction temperature.
Single shot ∆T
j −mb
= 50 × 0.055 + 25 × 0.085 − 25 × 0.065
= 3.25°C
T
jpeak
= 75 + 3.25 = 78.5°C
10% Duty cycle ∆T
j −mb
= 50 × 0.12 + 25 × 0.21 − 25 × 0.14
= 7.75°C
T
jpeak
= 75 + 7.75 = 82.5°C
50% Duty cycle ∆T
j −mb
= 50 × 0.42 + 25 × 0.7 − 25 × 0.5
= 26°C
T
jpeak
= 75 + 26 = 101°C
50% Duty Cycle P
av
=
50 × 50
200
= 12.5W
∆T
j −mb(av)
= P
av
× Z
thj −mb(δ = 1)
= 12.5 × 2 = 25°C
T
j(av)
= 75 + 25 = 100°C
10% Duty Cycle P
av
=
50 × 50
1000
= 2.5W
∆T
j −mb(av)
= P
av
× Z
thj −mb(δ = 1)
= 2.5 × 2 = 5°C
T
j(av)
= 75 + 5 = 80°C
566
Thermal Management Power Semiconductor Applications
Philips Semiconductors
7.1.2 Heat Dissipation
All semiconductor failure mechanisms are temperature
dependent and so the lower the junction temperature, the
higher the reliability of the circuit. Thus our data specifies
a maximum junction temperature which should not be
exceeded under the worst probable conditions. However,
derating the operating temperature from T
jmax
is always
desirable to improve the reliability still further. The junction
temperature depends on both the power dissipated in the
device and the thermal resistances (or impedances)
associated with the device. Thus careful consideration of
these thermal resistances (or impedances) allows the user
to calculate the maximum power dissipation that will keep
the junction temperature below a chosen value.
The formulae and diagrams given in this section can only
be considered as a guide for determining the nature of a
heatsink. This is because the thermal resistance of a
heatsink depends on numerous parameters which cannot
be predetermined. They include the position of the
transistor on the heatsink, the extent to which air can flow
unhindered, the ratio of the lengths of the sides of the
heatsink, the screening effect of nearby components, and
heating from these components. It is always advisable to
check important temperatures in the finished equipment
under the worst probable operating conditions. The more
complextheheat dissipationconditions, the moreimportant
it becomes to carry out such checks.
Heat flow path
The heat generated in a semiconductor chip flows by
various paths to the surroundings. Small signal devices do
not usually require heatsinking; the heat flows from the
junction to the mounting base which is in close contact with
the case. Heat is then lost by the case to the surroundings
by convection and radiation (Fig. 1a). Power transistors,
however, are usually mounted on heatsinks because of the
higher power dissipation they experience. Heat flows from
the transistor case to the heatsink by way of contact
pressure, and the heatsink loses heat to the surroundings
by convection and radiation, or by conduction to cooling
water (Fig. 1b). Generally air cooling is used so that the
ambient referred to in Fig.1 is usually the surrounding air.
Note that if this is the air inside an equipment case, the
additional thermal resistance between the inside and
outsideof theequipment caseshouldbetakenintoaccount.
Contact thermal resistance R
th mb-h
The thermal resistance between the transistor mounting
base and the heatsink depends on the quality and size of
the contact areas, the type of any intermediate plates used,
and the contact pressure. Care should be taken when
drilling holes in heatsinks to avoid burring and distorting the
metal, and both mating surfaces should be clean. Paint
finishes of normal thickness, up to 50 um (as a protection
against electrolytic voltage corrosion), barely affect the
thermal resistance. Transistor case and heatsink surfaces
can never be perfectly flat, and so contact will take place
on several points only, with a small air-gap over the rest of
the area. The use of a soft substance to fill this gap lowers
the contact thermal resistance. Normally, the gap is filled
with a heatsinking compound which remains fairly viscous
at normal transistor operating temperatures and has a high
thermal conductivity. The use of such a compound also
prevents moisture from penetrating between the contact
surfaces. Proprietary heatsinkingcompounds are available
which consist of a silicone grease loaded with some
electrically insulating good thermally conducting powder
such as alumina. The contact thermal resistance R
th mb-h
is
usually small with respect to (R
th j-mb
+ R
th h-amb
) when cooling
is by natural convection. However, the heatsink thermal
resistance R
th h-amb
can be very small when either forced
ventilation or water cooling are used, and thus a close
thermal contact between the transistor case and heatsink
becomes particularly important.
Fig. 1 Thermal resistances in the heat flow process:
(a) Without a heatsink
(b) With a heatsink
Thermal resistance calculations
Fig. 1a shows that, when a heatsink is not used, the total
thermal resistance between junction and ambient is given
by:
However, power transistors are generally mounted on a
heatsink since R
th j-amb
is not usually small enough to
maintain temperatures within the chip belowdesired levels.
Fig. 1b shows that, when a heatsink is used, the total
thermal resistance is given by:
R
thj − amb
= R
thj − mb
+ R
thmb − amb
1
R
thj − amb
= R
thj − mb
+ R
thmb − h
+ R
thh − amb
2
567
Thermal Management Power Semiconductor Applications
Philips Semiconductors
Note that the direct heat loss fromthe transistor case to the
surroundings through R
th mb-amb
is negligibly small.
The first stage in determining the size and nature of the
required heatsink is to calculate the maximum heatsink
thermal resistance R
th h-amb
that will maintain the junction
temperature below the desired value
Continuous operation
Under dc conditions, the maximum heatsink thermal
resistance can be calculated directly from the maximum
desired junction temperature.
Combining equations 2 and 3 gives:
and substituting Eq 4 into Eq 5 gives:
The values of R
th j-mb
and R
th mb-h
are given in the published
data. Thus, either Eq. 5 or Eq.6 can be used to find the
maximum heatsink thermal resistance.
Intermittent operation
The thermal equivalent circuits of Fig. 1 are inappropriate
for intermittent operation, and the thermal impedance Z
th j-mb
should be considered.
The mounting-base temperature has always been
assumed to remain constant under intermittent operation.
This assumption is known to be valid in practice provided
that the pulse time is less than about one second. The
mounting-base temperature does not change significantly
under these conditions as indicated in Fig. 2. This is
because heatsinks have a high thermal capacity and thus
a high thermal time-constant.
Thus Eq.6 is valid for intermittent operation, provided that
the pulse time is less than one second. The value of Tmb
can be calculated from Eq. 7, and the heatsink thermal
resistance can be obtained from Eq.6.
Fig. 2 Variation of junction and mounting base
temperature when the pulse time is small compared
with the thermal time constant of the heatsink
The thermal time constant of a transistor is defined as that
time at which the junction temperature has reached 70%
of its final value after being subjected to a constant power
dissipation at a constant mounting base temperature.
Now, if the pulse duration tp exceeds one second, the
transistor is temporarily in thermal equilibrium since such
a pulse duration is significantly greater than the thermal
time-constant of most transistors. Consequently, for pulse
times of more than one second, the temperature difference
T
j
- T
mb
reaches a stationary final value (Fig. 3) and Eq.7
should be replaced by:
In addition, it is no longer valid to assume that the mounting
base temperature is constant since the pulse time is also
no longer small with respect to the thermal time constant
of the heatsink.
Fig. 3 Variation of junction and mounting base
temperature when the pulse time is not small compared
with the thermal time constant of the heatsink
R
thj − amb
=
T
j
− T
amb
P
tot(av)
3
and R
thj − mb
=
T
j
− T
mb
P
tot(av)
4
R
thh − amb
=
T
j
− T
amb
P
tot(av)
− R
thj − mb
− R
thmb − h
5
R
thh − amb
=
T
mb
− T
amb
P
tot(av)
− R
thmb − h
6
T
mb
= T
j
− P
totM
.R
thj − mb
8
P
totM
=
T
j
− T
mb
Z
thj − mb
thus: T
mb
= T
j
− P
totM
.Z
thj − mb
7
568
Thermal Management Power Semiconductor Applications
Philips Semiconductors
Smaller heatsinks for intermittent
operation
In many instances, the thermal capacity of a heatsink can
be utilised to design a smaller heatsink for intermittent
operation than would be necessary for the same level of
continuous power dissipation. The average power
dissipation in Eq. 6 is replaced by the peak power
dissipation to obtain the value of the thermal impedance
between the heatsink and the surroundings
The value of Z
th h-amb
will be less than the comparable
thermal resistance and thus a smaller heatsink can be
designed than that obtained using the too large value
calculated from Eq.6.
Heatsinks
Three varieties of heatsink are in common use: flat plates
(including chassis), diecast finned heatsinks, and extruded
finned heatsinks. The material normally used for heatsink
construction is aluminium although copper may be used
with advantage for flat-sheet heatsinks. Small finned clips
aresometimesusedtoimprovethedissipationof low-power
transistors.
Heatsink finish
Heatsink thermal resistance is a function of surface finish.
Apainted surface will have agreater emissivity thanabright
unpainted one. The effect is most marked with flat plate
heatsinks, where about one third of the heat is dissipated
by radiation. The colour of the paint used is relatively
unimportant, and the thermal resistance of a flat plate
heatsink painted gloss white will be only about 3% higher
than that of the same heatsink painted matt black. With
finned heatsinks, painting is less effective since heat
radiated from most fins will fall on adjacent fins but it is still
worthwhile. Both anodising and etching will decrease the
thermal resistivity. Metallic type paints, such as aluminium
paint, have the lowest emissivities, although they are
approximately ten times better than a bright aluminium
metal finish.
Flat-plate heatsinks
The simplest type of heatsink is a flat metal plate to which
the transistor is attached. Such heatsinks are used both in
the form of separate plates and as the equipment chassis
itself. The thermal resistance obtained depends on the
thickness, area and orientation of the plate, as well as on
the finish and power dissipated. A plate mounted
horizontally will have about twice the thermal resistance of
a vertically mounted plate. This is particularly important
where the equipment chassis itself is used as the heatsink.
In Fig. 4, the thermal resistance of a blackened heatsink is
plotted against surface area (one side) with power
dissipation as a parameter. The graph is accurate to within
25%for nearly square plates, where the ratio of the lengths
of the sides is less than 1.25:1.
Finned heatsinks
Finned heatsinks may be made by stacking flat plates,
although it is usually more economical to use ready made
diecast or extruded heatsinks. Since most commercially
available finned heatsinks are of reasonably optimum
design, it is possible to compare them on the basis of the
overall volume whichtheyoccupy. Thiscomparisonismade
in Fig. 5 for heatsinks with their fins mounted vertically;
again, the graph is accurate to 25%.
Fig. 4 Generalised heatsink characteristics: flat vertical
black aluminium plates, 3mm thick, approximately
square
Fig. 5 Generalised heatsink characteristic: blackened
aluminium finned heatsinks
Z
thh − amb
=
T
mb
− T
amb
P
totM
− R
thmb − h
9
569
Thermal Management Power Semiconductor Applications
Philips Semiconductors
Fig. 6 Heatsink nomogram
1
10
100
10
10
10
3
4
5
10 100 1000
Bright Horizontal
Bright Vertical
Blackened Horizontal
Blackened Vertical
1W
2W
5W
10W
20W
50W
100W
1mm
2mm
3mm
30D
40D
EXTRUDED
FLAT PLATE
SOT93
TO220
SOT82
Length of Extruded Heatsink (mm)
Area
of
one
side
mm
2
Heatsink dimensions
The maximum thermal resistance through which sufficient
power can be dissipated without damaging the transistor
can be calculated as discussed previously. This section
explains how to arrive at a type and size of heatsink that
gives a sufficiently low thermal resistance.
Natural air cooling
The required size of aluminium heatsinks - whether flat or
extruded (finned) can be derived from the nomogram in
Fig. 6. Like all heatsink diagrams, the nomogramdoes not
giveexact values for R
th h-amb
as afunctionof the dimensions
sincethe practical conditionsalways deviatetosomeextent
fromthose under which the nomogramwas drawn up. The
actual values for the heatsink thermal resistance may differ
by up to 10%from the nomogram values. Consequently, it
is advisable to take temperature measurements in the
finished equipment, particularly where the thermal
conditions are critical.
The conditions to which the nomogram applies are as
follows:
• natural air cooling (unimpeded natural convection with no
build up of heat);
• ambient temperature about 25˚C, measured about 50mm
below the lower edge of the heatsink (see Fig. 7);
• atmospheric pressure about 10 N/m
2
;
• singlemounting (that is, not affectedby nearby heatsinks);
• distance between the bottomof the heatsink and the base
of a draught-free space about 100mm (see Fig. 7);
• transistor mounted roughly in the centre of the heatsink
(this is not so important for finned heatsinks because of
the good thermal conduction).
The appropriately-sized heatsink is found as follows.
1. Enter the nomogram from the right hand side of section
1 at the appropriate R
th h-amb
value (see Fig. 8). Move
horizontally to the left, until the appropriate curve for
orientation and surface finish is reached.
2. Move vertically upwards to intersect the appropriate
power dissipation curve in section 2.
570
Thermal Management Power Semiconductor Applications
Philips Semiconductors
3. Move horizontally to the left into section 3 for the desired
thickness of a flat-plate heatsink, or the type of extrusion.
4. If an extruded heatsink is required, move vertically
upwards to obtain its length (Figs. 9a and 9b give the
outlines of the extrusions).
5. If a flat-plate heatsink is to be used, move vertically
downwards to intersect the appropriate curve for
envelope type in section 4.
6. Move horizontally to the left to obtain heatsink area.
7. The heatsink dimensions should not exceed the ratio of
1.25:1.
Fig. 7 Conditions applicable to nomogram in Fig. 6
Fig. 8 Use of the heatsink nomogram
Fig. 9a Outline of Extrusion 30D
Fig. 9b Outline of Extrusion 40D
Thecurves insection2take account of thenonlinear nature
of the relationship between the temperature drop across
the heatsink and the power dissipation loss. Thus, at a
constant value of the heatsink thermal resistance, the
greater the power dissipation, the smaller is the required
size of heatsink. This is illustrated by the followingexample.
Example
An extruded heatsink mounted vertically and with a painted
surface is required to have a maximum thermal resistance
of R
th h-amb
= 2.6 ˚C/Wat the following powers:
Enter the nomogramat the appropriate value of the thermal
resistance in section 1, and via either the 50W or 5W line
in section 2, the appropriate lengths of the extruded
heatsink 30D are found to be:
Case (b) requires a shorter length since the temperature
difference is ten times greater than in case (a).
As the ambient temperature increases beyond 25˚C, so
does the temperature of the heatsink and thus the thermal
resistance (at constant power) decreases owing to the
increasing role of radiation in the heat removal process.
Consequently, a heatsink with dimensions derived from
Fig. 6 at T
amb
> 25˚C will be more than adequate. If the
maximumambient temperature is less than 25˚C, then the
thermal resistance will increase slightly. However, any
Approx
100mm
Approx
50mm
Tamb Tamb
Th
(a)P
tot (av)
= 5W (b)P
tot (av)
= 50W
(a) length = 110mm and (b) length = 44mm.
571
Thermal Management Power Semiconductor Applications
Philips Semiconductors
increase will lie within the limits of accuracy of the
nomogram and within the limits set by other uncertainties
associated with heatsink calculations.
For heatsinks with relatively small areas, a considerable
part of the heat is dissipated from the transistor case. This
is why the curves in section 4 tend to flatten out with
decreasing heatsink area. The area of extruded heatsinks
is always large with respect to the surface of the transistor
case, even when the length is small.
Fig. 10 Arrangement of two equally loaded transistors
mounted on a common heatsink
If several transistors are mounted on a common heatsink,
each transistor should be associated with a particular
section of the heatsink (either an area or length according
to type) whose maximum thermal resistance is calculated
from equations 5 or 6; that is, without taking the heat
produced by nearby transistors into account. Fromthe sum
of these areas or lengths, the size of the common heatsink
can then be obtained. If a flat heatsink is used, the
transistors are best arranged as shown in Fig. 10. The
maximum mounting base temperatures of transistors in
such a grouping should always be checked once the
equipment has been constructed.
Forced air cooling
If the thermal resistance needs to bemuchless than 1˚C/W,
or the heatsink not too large, forced air cooling by means
of fans can be provided. Apart fromthe size of the heatsink,
the thermal resistance now only depends on the speed of
the cooling air. Provided that the cooling air flows parallel
to the fins and with sufficient speed (>0.5m/s), the thermal
resistancehardly depends on the power dissipationandthe
orientation of the heatsink. Note that turbulence in the air
current can result in practical values deviating from
theoretical values.
Fig. 11 shows the formin which the thermal resistances for
forced air cooling are given in the case of extruded
heatsinks. It also shows the reduction in thermal resistance
or length of heatsink which may be obtained with forced air
cooling.
The effect of forced air cooling in the case of flat heatsinks
is seen from Fig. 12. Here, too, the dissipated power and
the orientation of the heatsink have only a slight effect on
the thermal resistance, provided that the air flow is
sufficiently fast.
Fig. 11 Thermal Resistance of a finned heatsink (type 40D) as a function of the length with natural and forced air
cooling
0 50 100 150 200 250 300 350
0
0.5
1
1.5
2
2.5
Length (mm)
Rth h-amb (K/W)
P=3W
P=10W
P=30W
P=100W
1m/s
2m/s
5m/s
Natural
Convection
Forced
Cooling
(Vertical)
Blackened
572
Thermal Management Power Semiconductor Applications
Philips Semiconductors
Fig. 12 Thermal Resistances of heatsinks (2mm thick copper or 3mm thick aluminium) under natural convection and
forced cooling conditions, with a SOT93 envelope.
(a) blackened
(b) bright
Summary
The majority of power transistors require heatsinking, and
once the maximumthermal resistance that will maintain the
device’s junction temperature below its rating has been
calculated, a heatsink of appropriate type and size can be
chosen. The practical conditions under which a transistor
will be operated are likely to differ from the theoretical
considerations used to determine the required heatsink,
and thus temperatures should always be checked in the
finished equipment. Finally, some applications require a
small heatsink, or one with a very low thermal resistance,
in which case forced air cooling by means of fans should
be provided.
573
Preface Power Semiconductor Applications
Philips Semiconductors
Acknowledgments
We are grateful for all the contributions fromour colleagues within Philips and to the Application Laboratories in Eindhoven
and Hamburg.
We would also like to thank Dr.P.H.Mellor of the University of Sheffield for contributing the application note of section 3.1.5.
The authors thank Mrs.R.Hayes for her considerable help in the preparation of this book.
The authors also thank Mr.D.F.Haslam for his assistance in the formatting and printing of the manuscripts.
Contributing Authors
N.Bennett
M.Bennion
D.Brown
C.Buethker
L.Burley
G.M.Fry
R.P.Gant
J.Gilliam
D.Grant
N.J.Ham
C.J.Hammerton
D.J.Harper
W.Hettersheid
J.v.d.Hooff
J.Houldsworth
M.J.Humphreys
P.H.Mellor
R.Miller
H.Misdom
P.Moody
S.A.Mulder
E.B.G. Nijhof
J.Oosterling
N.Pichowicz
W.B.Rosink
D.C. de Ruiter
D.Sharples
H.Simons
T.Stork
D.Tebb
H.Verhees
F.A.Woodworth
T.van de Wouw
This book was originally prepared by the Power Semiconductor Applications Laboratory, of the Philips Semiconductors
product division, Hazel Grove:
M.J.Humphreys
C.J.Hammerton
D.Brown
R.Miller
L.Burley
It was revised and updated, in 1994, by:
N.J.Ham C.J.Hammerton D.Sharples
Preface Power Semiconductor Applications
Philips Semiconductors
Preface
This book was prepared by the Power Semiconductor Applications Laboratory of the Philips Semiconductors product
division, Hazel Grove. The book is intended as a guide to using power semiconductors both efficiently and reliably in power
conversion applications. It is made up of eight main chapters each of which contains a number of application notes aimed
at making it easier to select and use power semiconductors.
CHAPTER 1 forms an introduction to power semiconductors concentrating particularly on the two major power transistor
technologies, Power MOSFETs and High Voltage Bipolar Transistors.
CHAPTER 2 is devoted to Switched Mode Power Supplies. It begins with a basic description of the most commonly used
topologies and discusses the major issues surrounding the use of power semiconductors including rectifiers. Specific
design examples are given as well as a look at designing the magnetic components. The end of this chapter describes
resonant power supply technology.
CHAPTER 3 describes motion control in terms of ac, dc and stepper motor operation and control. This chapter looks only
at transistor controls, phase control using thyristors and triacs is discussed separately in chapter 6.
CHAPTER 4 looks at television and monitor applications. A description of the operation of horizontal deflection circuits is
givenfollowed by transistor selection guides for both deflectionand power supply applications. Deflection and power supply
circuit examples arealsogivenbasedoncircuitsdesignedbytheProduct Concept andApplicationLaboratories(Eindhoven).
CHAPTER5 concentrates on automotive electronics looking in detail at the requirements for the electronic switches taking
into consideration the harsh environment in which they must operate.
CHAPTER6 reviews thyristor andtriac applications fromthe basics of device technology and operation to the simple design
rules which should be followed to achieve maximum reliability. Specific examples are given in this chapter for a number
of the common applications.
CHAPTER 7 looks at the thermal considerations for power semiconductors in terms of power dissipation and junction
temperaturelimits. Part of this chapter is devotedto worked examples showing howjunctiontemperatures can becalculated
to ensure the limits are not exceeded. Heatsink requirements and designs are also discussed in the second half of this
chapter.
CHAPTER 8 is an introduction to the use of high voltage bipolar transistors in electronic lighting ballasts. Many of the
possible topologies are described.
Contents Power Semiconductor Applications
Philips Semiconductors
Table of Contents
CHAPTER 1 Introduction to Power Semiconductors 1
General 3
1.1.1 An Introduction To Power Devices ............................................................ 5
Power MOSFET 17
1.2.1 PowerMOS Introduction ............................................................................. 19
1.2.2 Understanding Power MOSFET Switching Behaviour ............................... 29
1.2.3 Power MOSFET Drive Circuits .................................................................. 39
1.2.4 Parallel Operation of Power MOSFETs ..................................................... 49
1.2.5 Series Operation of Power MOSFETs ....................................................... 53
1.2.6 Logic Level FETS ...................................................................................... 57
1.2.7 Avalanche Ruggedness ............................................................................. 61
1.2.8 Electrostatic Discharge (ESD) Considerations .......................................... 67
1.2.9 Understanding the Data Sheet: PowerMOS .............................................. 69
High Voltage Bipolar Transistor 77
1.3.1 Introduction To High Voltage Bipolar Transistors ...................................... 79
1.3.2 Effects of Base Drive on Switching Times ................................................. 83
1.3.3 Using High Voltage Bipolar Transistors ..................................................... 91
1.3.4 Understanding The Data Sheet: High Voltage Transistors ....................... 97
CHAPTER 2 Switched Mode Power Supplies 103
Using Power Semiconductors in Switched Mode Topologies 105
2.1.1 An Introduction to Switched Mode Power Supply Topologies ................... 107
2.1.2 The Power Supply Designer’s Guide to High Voltage Transistors ............ 129
2.1.3 Base Circuit Design for High Voltage Bipolar Transistors in Power
Converters ........................................................................................................... 141
2.1.4 Isolated Power Semiconductors for High Frequency Power Supply
Applications ......................................................................................................... 153
Output Rectification 159
2.2.1 Fast Recovery Epitaxial Diodes for use in High Frequency Rectification 161
2.2.2 Schottky Diodes from Philips Semiconductors .......................................... 173
2.2.3 An Introduction to Synchronous Rectifier Circuits using PowerMOS
Transistors ........................................................................................................... 179
i
Contents Power Semiconductor Applications
Philips Semiconductors
Design Examples 185
2.3.1 Mains Input 100 W Forward Converter SMPS: MOSFET and Bipolar
Transistor Solutions featuring ETD Cores ........................................................... 187
2.3.2 Flexible, Low Cost, Self-Oscillating Power Supply using an ETD34
Two-Part Coil Former and 3C85 Ferrite .............................................................. 199
Magnetics Design 205
2.4.1 Improved Ferrite Materials and Core Outlines for High Frequency Power
Supplies ............................................................................................................... 207
Resonant Power Supplies 217
2.5.1. An Introduction To Resonant Power Supplies .......................................... 219
2.5.2. Resonant Power Supply Converters - The Solution For Mains Pollution
Problems .............................................................................................................. 225
CHAPTER 3 Motor Control 241
AC Motor Control 243
3.1.1 Noiseless A.C. Motor Control: Introduction to a 20 kHz System ............... 245
3.1.2 The Effect of a MOSFET’s Peak to Average Current Rating on Invertor
Efficiency ............................................................................................................. 251
3.1.3 MOSFETs and FREDFETs for Motor Drive Equipment ............................. 253
3.1.4 A Designers Guide to PowerMOS Devices for Motor Control ................... 259
3.1.5 A 300V, 40A High Frequency Inverter Pole Using Paralleled FREDFET
Modules ............................................................................................................... 273
DC Motor Control 283
3.2.1 Chopper circuits for DC motor control ....................................................... 285
3.2.2 A switched-mode controller for DC motors ................................................ 293
3.2.3 Brushless DC Motor Systems .................................................................... 301
Stepper Motor Control 307
3.3.1 Stepper Motor Control ............................................................................... 309
CHAPTER 4 Televisions and Monitors 317
Power Devices in TV & Monitor Applications (including selection
guides) 319
4.1.1 An Introduction to Horizontal Deflection .................................................... 321
4.1.2 The BU25XXA/D Range of Deflection Transistors .................................... 331
ii
Contents Power Semiconductor Applications
Philips Semiconductors
4.1.3 Philips HVT’s for TV & Monitor Applications .............................................. 339
4.1.4 TV and Monitor Damper Diodes ................................................................ 345
TV Deflection Circuit Examples 349
4.2.1 Application Information for the 16 kHz Black Line Picture Tubes .............. 351
4.2.2 32 kHz / 100 Hz Deflection Circuits for the 66FS Black Line Picture Tube 361
SMPS Circuit Examples 377
4.3.1 A 70W Full Performance TV SMPS Using The TDA8380 ......................... 379
4.3.2 A Synchronous 200W SMPS for 16 and 32 kHz TV .................................. 389
Monitor Deflection Circuit Example 397
4.4.1 A Versatile 30 - 64 kHz Autosync Monitor ................................................. 399
CHAPTER 5 Automotive Power Electronics 421
Automotive Motor Control (including selection guides) 423
5.1.1 Automotive Motor Control With Philips MOSFETS .................................... 425
Automotive Lamp Control (including selection guides) 433
5.2.1 Automotive Lamp Control With Philips MOSFETS .................................... 435
The TOPFET 443
5.3.1 An Introduction to the 3 pin TOPFET ......................................................... 445
5.3.2 An Introduction to the 5 pin TOPFET ......................................................... 447
5.3.3 BUK101-50DL - a Microcontroller compatible TOPFET ............................ 449
5.3.4 Protection with 5 pin TOPFETs ................................................................. 451
5.3.5 Driving TOPFETs ....................................................................................... 453
5.3.6 High Side PWM Lamp Dimmer using TOPFET ......................................... 455
5.3.7 Linear Control with TOPFET ...................................................................... 457
5.3.8 PWM Control with TOPFET ....................................................................... 459
5.3.9 Isolated Drive for TOPFET ........................................................................ 461
5.3.10 3 pin and 5 pin TOPFET Leadforms ........................................................ 463
5.3.11 TOPFET Input Voltage ............................................................................ 465
5.3.12 Negative Input and TOPFET ................................................................... 467
5.3.13 Switching Inductive Loads with TOPFET ................................................. 469
5.3.14 Driving DC Motors with TOPFET ............................................................. 471
5.3.15 An Introduction to the High Side TOPFET ............................................... 473
5.3.16 High Side Linear Drive with TOPFET ...................................................... 475
iii
Contents Power Semiconductor Applications
Philips Semiconductors
Automotive Ignition 477
5.4.1 An Introduction to Electronic Automotive Ignition ...................................... 479
5.4.2 IGBTs for Automotive Ignition .................................................................... 481
5.4.3 Electronic Switches for Automotive Ignition ............................................... 483
CHAPTER 6 Power Control with Thyristors and Triacs 485
Using Thyristors and Triacs 487
6.1.1 Introduction to Thyristors and Triacs ......................................................... 489
6.1.2 Using Thyristors and Triacs ....................................................................... 497
6.1.3 The Peak Current Handling Capability of Thyristors .................................. 505
6.1.4 Understanding Thyristor and Triac Data .................................................... 509
Thyristor and Triac Applications 521
6.2.1 Triac Control of DC Inductive Loads .......................................................... 523
6.2.2 Domestic Power Control with Triacs and Thyristors .................................. 527
6.2.3 Design of a Time Proportional Temperature Controller ............................. 537
Hi-Com Triacs 547
6.3.1 Understanding Hi-Com Triacs ................................................................... 549
6.3.2 Using Hi-Com Triacs .................................................................................. 551
CHAPTER 7 Thermal Management 553
Thermal Considerations 555
7.1.1 Thermal Considerations for Power Semiconductors ................................. 557
7.1.2 Heat Dissipation ......................................................................................... 567
CHAPTER 8 Lighting 575
Fluorescent Lamp Control 577
8.1.1 Efficient Fluorescent Lighting using Electronic Ballasts ............................. 579
8.1.2 Electronic Ballasts - Philips Transistor Selection Guide ............................ 587
8.1.3 An Electronic Ballast - Base Drive Optimisation ........................................ 589
iv
Index Power Semiconductor Applications
Philips Semiconductors
Index
Airgap, transformer core, 111, 113
Anti saturation diode, 590
Asynchronous, 497
Automotive
fans
see motor control
IGBT, 481, 483
ignition, 479, 481, 483
lamps, 435, 455
motor control, 425, 457, 459, 471, 475
resistive loads, 442
reverse battery, 452, 473, 479
screen heater, 442
seat heater, 442
solenoids, 469
TOPFET, 473
Avalanche, 61
Avalanche breakdown
thyristor, 490
Avalanche multiplication, 134
Baker clamp, 138, 187, 190
Ballast
electronic, 580
fluorescent lamp, 579
switchstart, 579
Base drive, 136
base inductor, 147
base inductor, diode assisted, 148
base resistor, 146
drive transformer, 145
drive transformer leakage inductance, 149
electronic ballast, 589
forward converter, 187
power converters, 141
speed-up capacitor, 143
Base inductor, 144, 147
Base inductor, diode assisted, 148
Boost converter, 109
continuous mode, 109
discontinuous mode, 109
output ripple, 109
Bootstrap, 303
Breakback voltage
diac, 492
Breakdown voltage, 70
Breakover current
diac, 492
Breakover voltage
diac, 492, 592
thyristor, 490
Bridge circuits
see Motor Control - AC
Brushless motor, 301, 303
Buck-boost converter, 110
Buck converter, 108 - 109
Burst firing, 537
Burst pulses, 564
Capacitance
junction, 29
Capacitor
mains dropper, 544
CENELEC, 537
Charge carriers, 133
triac commutation, 549
Choke
fluorescent lamp, 580
Choppers, 285
Clamp diode, 117
Clamp winding, 113
Commutation
diode, 164
Hi-Com triac, 551
thyristor, 492
triac, 494, 523, 529
Compact fluorescent lamp, 585
Continuous mode
see Switched Mode Power Supplies
Continuous operation, 557
Converter (dc-dc)
switched mode power supply, 107
Cookers, 537
Cooling
forced, 572
natural, 570
Crest factor, 529
Critical electric field, 134
Cross regulation, 114, 117
Current fed resonant inverter, 589
Current Mode Control, 120
Current tail, 138, 143
Damper Diodes, 345, 367
forward recovery, 328, 348
losses, 347
outlines, 345
picture distortion, 328, 348
selection guide, 345
Darlington, 13
Data Sheets
High Voltage Bipolar Transistor, 92,97,331
MOSFET, 69
i
Index Power Semiconductor Applications
Philips Semiconductors
dc-dc converter, 119
Depletion region, 133
Desaturation networks, 86
Baker clamp, 91, 138
dI/dt
triac, 531
Diac, 492, 500, 527, 530, 591
Diode, 6
double diffused, 162
epitaxial, 161
schottky, 173
structure, 161
Diode Modulator, 327, 367
Disc drives, 302
Discontinuous mode
see Switched Mode Power Supplies
Domestic Appliances, 527
Dropper
capacitive, 544
resistive, 544, 545
Duty cycle, 561
EFD core
see magnetics
Efficiency Diodes
see Damper Diodes
Electric drill, 531
Electronic ballast, 580
base drive optimisation, 589
current fed half bridge, 584, 587, 589
current fed push pull, 583, 587
flyback, 582
transistor selection guide, 587
voltage fed half bridge, 584, 588
voltage fed push pull, 583, 587
EMC, 260, 455
see RFI, ESD
TOPFET, 473
Emitter shorting
triac, 549
Epitaxial diode, 161
characteristics, 163
dI/dt, 164
forward recovery, 168
lifetime control, 162
operating frequency, 165
passivation, 162
reverse leakage, 169
reverse recovery, 162, 164
reverse recovery softness, 167
selection guide, 171
snap-off, 167
softness factor, 167
stored charge, 162
technology, 162
ESD, 67
see Protection, ESD
precautions, 67
ETD core
see magnetics
F-pack
see isolated package
Fall time, 143, 144
Fast Recovery Epitaxial Diode (FRED)
see epitaxial diode
FBSOA, 134
Ferrites
see magnetics
Flicker
fluorescent lamp, 580
Fluorescent lamp, 579
colour rendering, 579
colour temperature, 579
efficacy, 579, 580
triphosphor, 579
Flyback converter, 110, 111, 113
advantages, 114
clamp winding, 113
continuous mode, 114
coupled inductor, 113
cross regulation, 114
diodes, 115
disadvantages, 114
discontinuous mode, 114
electronic ballast, 582
leakage inductance, 113
magnetics, 213
operation, 113
rectifier circuit, 180
self oscillating power supply, 199
synchronous rectifier, 156, 181
transformer core airgap, 111, 113
transistors, 115
Flyback converter (two transistor), 111, 114
Food mixer, 531
Forward converter, 111, 116
advantages, 116
clamp diode, 117
conduction loss, 197
continuous mode, 116
core loss, 116
core saturation, 117
cross regulation, 117
diodes, 118
disadvantages, 117
duty ratio, 117
ferrite cores, 116
magnetics, 213
magnetisation energy, 116, 117
ii
Index Power Semiconductor Applications
Philips Semiconductors
operation, 116
output diodes, 117
output ripple, 116
rectifier circuit, 180
reset winding, 117
switched mode power supply, 187
switching frequency, 195
switching losses, 196
synchronous rectifier, 157, 181
transistors, 118
Forward converter (two transistor), 111, 117
Forward recovery, 168
FREDFET, 250, 253, 305
bridge circuit, 255
charge, 254
diode, 254
drive, 262
loss, 256
reverse recovery, 254
FREDFETs
motor control, 259
Full bridge converter, 111, 125
advantages, 125
diodes, 126
disadvantages, 125
operation, 125
transistors, 126
Gate
triac, 538
Gate drive
forward converter, 195
Gold doping, 162, 169
GTO, 11
Guard ring
schottky diode, 174
Half bridge, 253
Half bridge circuits
see also Motor Control - AC
Half bridge converter, 111, 122
advantages, 122
clamp diodes, 122
cross conduction, 122
diodes, 124
disadvantages, 122
electronic ballast, 584, 587, 589
flux symmetry, 122
magnetics, 214
operation, 122
synchronous rectifier, 157
transistor voltage, 122
transistors, 124
voltage doubling, 122
Heat dissipation, 567
Heat sink compound, 567
Heater controller, 544
Heaters, 537
Heatsink, 569
Heatsink compound, 514
Hi-Com triac, 519, 549, 551
commutation, 551
dIcom/dt, 552
gate trigger current, 552
inductive load control, 551
High side switch
MOSFET, 44, 436
TOPFET, 430, 473
High Voltage Bipolar Transistor, 8, 79, 91,
141, 341
‘bathtub’ curves, 333
avalanche breakdown, 131
avalanche multiplication, 134
Baker clamp, 91, 138
base-emitter breakdown, 144
base drive, 83, 92, 96, 136, 336, 385
base drive circuit, 145
base inductor, 138, 144, 147
base inductor, diode assisted, 148
base resistor, 146
breakdown voltage, 79, 86, 92
carrier concentration, 151
carrier injection, 150
conductivity modulation, 135, 150
critical electric field, 134
current crowding, 135, 136
current limiting values, 132
current tail, 138, 143
current tails, 86, 91
d-type, 346
data sheet, 92, 97, 331
depletion region, 133
desaturation, 86, 88, 91
device construction, 79
dI/dt, 139
drive transformer, 145
drive transformer leakage inductance, 149
dV/dt, 139
electric field, 133
electronic ballast, 581, 585, 587, 589
Fact Sheets, 334
fall time, 86, 99, 143, 144
FBSOA, 92, 99, 134
hard turn-off, 86
horizontal deflection, 321, 331, 341
leakage current, 98
limiting values, 97
losses, 92, 333, 342
Miller capacitance, 139
operation, 150
iii
Index Power Semiconductor Applications
Philips Semiconductors
optimum drive, 88
outlines, 332, 346
over current, 92, 98
over voltage, 92, 97
overdrive, 85, 88, 137, 138
passivation, 131
power limiting value, 132
process technology, 80
ratings, 97
RBSOA, 93, 99, 135, 138, 139
RC network, 148
reverse recovery, 143, 151
safe operating area, 99, 134
saturation, 150
saturation current, 79, 98, 341
secondary breakdown, 92, 133
smooth turn-off, 86
SMPS, 94, 339, 383
snubber, 139
space charge, 133
speed-up capacitor, 143
storage time, 86, 91, 92, 99, 138, 144, 342
sub emitter resistance, 135
switching, 80, 83, 86, 91, 98, 342
technology, 129, 149
thermal breakdown, 134
thermal runaway, 152
turn-off, 91, 92, 138, 142, 146, 151
turn-on, 91, 136, 141, 149, 150
underdrive, 85, 88
voltage limiting values, 130
Horizontal Deflection, 321, 367
base drive, 336
control ic, 401
d-type transistors, 346
damper diodes, 345, 367
diode modulator, 327, 347, 352, 367
drive circuit, 352, 365, 406
east-west correction, 325, 352, 367
line output transformer, 354
linearity correction, 323
operating cycle, 321, 332, 347
s-correction, 323, 352, 404
TDA2595, 364, 368
TDA4851, 400
TDA8433, 363, 369
test circuit, 321
transistors, 331, 341, 408
waveforms, 322
IGBT, 11, 305
automotive, 481, 483
clamped, 482, 484
ignition, 481, 483
Ignition
automotive, 479, 481, 483
darlington, 483
Induction heating, 53
Induction motor
see Motor Control - AC
Inductive load
see Solenoid
Inrush current, 528, 530
Intrinsic silicon, 133
Inverter, 260, 273
see motor control ac
current fed, 52, 53
switched mode power supply, 107
Irons, electric, 537
Isolated package, 154
stray capacitance, 154, 155
thermal resistance, 154
Isolation, 153
J-FET, 9
Junction temperature, 470, 557, 561
burst pulses, 564
non-rectangular pulse, 565
rectangular pulse, composite, 562
rectangular pulse, periodic, 561
rectangular pulse, single shot, 561
Lamp dimmer, 530
Lamps, 435
dI/dt, 438
inrush current, 438
MOSFET, 435
PWM control, 455
switch rate, 438
TOPFET, 455
Latching current
thyristor, 490
Leakage inductance, 113, 200, 523
Lifetime control, 162
Lighting
fluorescent, 579
phase control, 530
Logic Level FET
motor control, 432
Logic level MOSFET, 436
Magnetics, 207
100W 100kHz forward converter, 197
100W 50kHz forward converter, 191
50W flyback converter, 199
core losses, 208
core materials, 207
EFD core, 210
ETD core, 199, 207
iv
Index Power Semiconductor Applications
Philips Semiconductors
flyback converter, 213
forward converter, 213
half bridge converter, 214
power density, 211
push-pull converter, 213
switched mode power supply, 187
switching frequency, 215
transformer construction, 215
Mains Flicker, 537
Mains pollution, 225
pre-converter, 225
Mains transient, 544
Mesa glass, 162
Metal Oxide Varistor (MOV), 503
Miller capacitance, 139
Modelling, 236, 265
MOS Controlled Thyristor, 13
MOSFET, 9, 19, 153, 253
bootstrap, 303
breakdown voltage, 22, 70
capacitance, 30, 57, 72, 155, 156
capacitances, 24
characteristics, 23, 70 - 72
charge, 32, 57
data sheet, 69
dI/dt, 36
diode, 253
drive, 262, 264
drive circuit loss, 156
driving, 39, 250
dV/dt, 36, 39, 264
ESD, 67
gate-source protection, 264
gate charge, 195
gate drive, 195
gate resistor, 156
high side, 436
high side drive, 44
inductive load, 62
lamps, 435
leakage current, 71
linear mode, parallelling, 52
logic level, 37, 57, 305
loss, 26, 34
maximum current, 69
motor control, 259, 429
modelling, 265
on-resistance, 21, 71
package inductance, 49, 73
parallel operation, 26, 47, 49, 265
parasitic oscillations, 51
peak current rating, 251
Resonant supply, 53
reverse diode, 73
ruggedness, 61, 73
safe operating area, 25, 74
series operation, 53
SMPS, 339, 384
solenoid, 62
structure, 19
switching, 24, 29, 58, 73, 194, 262
switching loss, 196
synchronous rectifier, 179
thermal impedance, 74
thermal resistance, 70
threshold voltage, 21, 70
transconductance, 57, 72
turn-off, 34, 36
turn-on, 32, 34, 35, 155, 256
Motor, universal
back EMF, 531
starting, 528
Motor Control - AC, 245, 273
anti-parallel diode, 253
antiparallel diode, 250
carrier frequency, 245
control, 248
current rating, 262
dc link, 249
diode, 261
diode recovery, 250
duty ratio, 246
efficiency, 262
EMC, 260
filter, 250
FREDFET, 250, 259, 276
gate drives, 249
half bridge, 245
inverter, 250, 260, 273
line voltage, 262
loss, 267
MOSFET, 259
Parallel MOSFETs, 276
peak current, 251
phase voltage, 262
power factor, 262
pulse width modulation, 245, 260
ripple, 246
short circuit, 251
signal isolation, 250
snubber, 276
speed control, 248
switching frequency, 246
three phase bridge, 246
underlap, 248
Motor Control - DC, 285, 293, 425
braking, 285, 299
brushless, 301
control, 290, 295, 303
current rating, 288
v
Index Power Semiconductor Applications
Philips Semiconductors
drive, 303
duty cycle, 286
efficiency, 293
FREDFET, 287
freewheel diode, 286
full bridge, 287
half bridge, 287
high side switch, 429
IGBT, 305
inrush, 430
inverter, 302
linear, 457, 475
logic level FET, 432
loss, 288
MOSFET, 287, 429
motor current, 295
overload, 430
permanent magnet, 293, 301
permanent magnet motor, 285
PWM, 286, 293, 459, 471
servo, 298
short circuit, 431
stall, 431
TOPFET, 430, 457, 459, 475
topologies, 286
torque, 285, 294
triac, 525
voltage rating, 288
Motor Control - Stepper, 309
bipolar, 310
chopper, 314
drive, 313
hybrid, 312
permanent magnet, 309
reluctance, 311
step angle, 309
unipolar, 310
Mounting, transistor, 154
Mounting base temperature, 557
Mounting torque, 514
Parasitic oscillation, 149
Passivation, 131, 162
PCB Design, 368, 419
Phase angle, 500
Phase control, 546
thyristors and triacs, 498
triac, 523
Phase voltage
see motor control - ac
Power dissipation, 557
see High Voltage Bipolar Transistor loss,
MOSFET loss
Power factor correction, 580
active, boost converted, 581
Power MOSFET
see MOSFET
Proportional control, 537
Protection
ESD, 446, 448, 482
overvoltage, 446, 448, 469
reverse battery, 452, 473, 479
short circuit, 251, 446, 448
temperature, 446, 447, 471
TOPFET, 445, 447, 451
Pulse operation, 558
Pulse Width Modulation (PWM), 108
Push-pull converter, 111, 119
advantages, 119
clamp diodes, 119
cross conduction, 119
current mode control, 120
diodes, 121
disadvantages, 119
duty ratio, 119
electronic ballast, 582, 587
flux symmetry, 119, 120
magnetics, 213
multiple outputs, 119
operation, 119
output filter, 119
output ripple, 119
rectifier circuit, 180
switching frequency, 119
transformer, 119
transistor voltage, 119
transistors, 121
Qs (stored charge), 162
RBSOA, 93, 99, 135, 138, 139
Rectification, synchronous, 179
Reset winding, 117
Resistor
mains dropper, 544, 545
Resonant power supply, 219, 225
modelling, 236
MOSFET, 52, 53
pre-converter, 225
Reverse leakage, 169
Reverse recovery, 143, 162
RFI, 154, 158, 167, 393, 396, 497, 529, 530,
537
Ruggedness
MOSFET, 62, 73
schottky diode, 173
Safe Operating Area (SOA), 25, 74, 134, 557
forward biased, 92, 99, 134
reverse biased, 93, 99, 135, 138, 139
vi
Index Power Semiconductor Applications
Philips Semiconductors
Saturable choke
triac, 523
Schottky diode, 173
bulk leakage, 174
edge leakage, 174
guard ring, 174
reverse leakage, 174
ruggedness, 173
selection guide, 176
technology, 173
SCR
see Thyristor
Secondary breakdown, 133
Selection Guides
BU25XXA, 331
BU25XXD, 331
damper diodes, 345
EPI diodes, 171
horizontal deflection, 343
MOSFETs driving heaters, 442
MOSFETs driving lamps, 441
MOSFETs driving motors, 426
Schottky diodes, 176
SMPS, 339
Self Oscillating Power Supply (SOPS)
50W microcomputer flyback converter, 199
ETD transformer, 199
Servo, 298
Single ended push-pull
see half bridge converter
Snap-off, 167
Snubber, 93, 139, 495, 502, 523, 529, 549
active, 279
Softness factor, 167
Solenoid
TOPFET, 469, 473
turn off, 469, 473
Solid state relay, 501
SOT186, 154
SOT186A, 154
SOT199, 154
Space charge, 133
Speed-up capacitor, 143
Speed control
thyristor, 531
triac, 527
Starter
fluorescent lamp, 580
Startup circuit
electronic ballast, 591
self oscillating power supply, 201
Static Induction Thyristor, 11
Stepdown converter, 109
Stepper motor, 309
Stepup converter, 109
Storage time, 144
Stored charge, 162
Suppression
mains transient, 544
Switched Mode Power Supply (SMPS)
see also self oscillating power supply
100W 100kHz MOSFET forward converter,
192
100W 500kHz half bridge converter, 153
100W 50kHz bipolar forward converter, 187
16 & 32 kHz TV, 389
asymmetrical, 111, 113
base circuit design, 149
boost converter, 109
buck-boost converter, 110
buck converter, 108
ceramic output filter, 153
continuous mode, 109, 379
control ic, 391
control loop, 108
core excitation, 113
core loss, 167
current mode control, 120
dc-dc converter, 119
diode loss, 166
diode reverse recovery effects, 166
diode reverse recovery softness, 167
diodes, 115, 118, 121, 124, 126
discontinuous mode, 109, 379
epitaxial diodes, 112, 161
flux swing, 111
flyback converter, 92, 111, 113, 123
forward converter, 111, 116, 379
full bridge converter, 111, 125
half bridge converter, 111, 122
high voltage bipolar transistor, 94, 112, 115,
118, 121, 124, 126, 129, 339, 383, 392
isolated, 113
isolated packages, 153
isolation, 108, 111
magnetics design, 191, 197
magnetisation energy, 113
mains filter, 380
mains input, 390
MOSFET, 112, 153, 33, 384
multiple output, 111, 156
non-isolated, 108
opto-coupler, 392
output rectifiers, 163
parasitic oscillation, 149
power-down, 136
power-up, 136, 137, 139
power MOSFET, 153, 339, 384
pulse width modulation, 108
push-pull converter, 111, 119
vii
Index Power Semiconductor Applications
Philips Semiconductors
RBSOA failure, 139
rectification, 381, 392
rectification efficiency, 163
rectifier selection, 112
regulation, 108
reliability, 139
resonant
see resonant power supply
RFI, 154, 158, 167
schottky diode, 112, 154, 173
snubber, 93, 139, 383
soft start, 138
standby, 382
standby supply, 392
start-up, 391
stepdown, 109
stepup, 109
symmetrical, 111, 119, 122
synchronisation, 382
synchronous rectification, 156, 179
TDA8380, 381, 391
topologies, 107
topology output powers, 111
transformer, 111
transformer saturation, 138
transformers, 391
transistor current limiting value, 112
transistor mounting, 154
transistor selection, 112
transistor turn-off, 138
transistor turn-on, 136
transistor voltage limiting value, 112
transistors, 115, 118, 121, 124, 126
turns ratio, 111
TV & Monitors, 339, 379, 399
two transistor flyback, 111, 114
two transistor forward, 111, 117
Switching loss, 230
Synchronous, 497
Synchronous rectification, 156, 179
self driven, 181
transformer driven, 180
Temperature control, 537
Thermal
continuous operation, 557, 568
intermittent operation, 568
non-rectangular pulse, 565
pulse operation, 558
rectangular pulse, composite, 562
rectangular pulse, periodic, 561
rectangular pulse, single shot, 561
single shot operation, 561
Thermal capacity, 558, 568
Thermal characteristics
power semiconductors, 557
Thermal impedance, 74, 568
Thermal resistance, 70, 154, 557
Thermal time constant, 568
Thyristor, 10, 497, 509
’two transistor’ model, 490
applications, 527
asynchronous control, 497
avalanche breakdown, 490
breakover voltage, 490, 509
cascading, 501
commutation, 492
control, 497
current rating, 511
dI/dt, 490
dIf/dt, 491
dV/dt, 490
energy handling, 505
external commutation, 493
full wave control, 499
fusing I
2
t, 503, 512
gate cathode resistor, 500
gate circuits, 500
gate current, 490
gate power, 492
gate requirements, 492
gate specifications, 512
gate triggering, 490
half wave control, 499
holding current, 490, 509
inductive loads, 500
inrush current, 503
latching current, 490, 509
leakage current, 490
load line, 492
mounting, 514
operation, 490
overcurrent, 503
peak current, 505
phase angle, 500
phase control, 498, 527
pulsed gate, 500
resistive loads, 498
resonant circuit, 493
reverse characteristic, 489
reverse recovery, 493
RFI, 497
self commutation, 493
series choke, 502
snubber, 502
speed controller, 531
static switching, 497
structure, 489
switching, 489
viii
Index Power Semiconductor Applications
Philips Semiconductors
switching characteristics, 517
synchronous control, 497
temperature rating, 512
thermal specifications, 512
time proportional control, 497
transient protection, 502
trigger angle, 500
turn-off time, 494
turn-on, 490, 509
turn-on dI/dt, 502
varistor, 503
voltage rating, 510
Thyristor data, 509
Time proportional control, 537
TOPFET
3 pin, 445, 449, 461
5 pin, 447, 451, 457, 459, 463
driving, 449, 453, 461, 465, 467, 475
high side, 473, 475
lamps, 455
leadforms, 463
linear control, 451, 457
motor control, 430, 457, 459
negative input, 456, 465, 467
protection, 445, 447, 451, 469, 473
PWM control, 451, 455, 459
solenoids, 469
Transformer
triac controlled, 523
Transformer core airgap, 111, 113
Transformers
see magnetics
Transient thermal impedance, 559
Transient thermal response, 154
Triac, 497, 510, 518
400Hz operation, 489, 518
applications, 527, 537
asynchronous control, 497
breakover voltage, 510
charge carriers, 549
commutating dI/dt, 494
commutating dV/dt, 494
commutation, 494, 518, 523, 529, 549
control, 497
dc inductive load, 523
dc motor control, 525
dI/dt, 531, 549
dIcom/dt, 523
dV/dt, 523, 549
emitter shorting, 549
full wave control, 499
fusing I
2
t, 503, 512
gate cathode resistor, 500
gate circuits, 500
gate current, 491
gate requirements, 492
gate resistor, 540, 545
gate sensitivity, 491
gate triggering, 538
holding current, 491, 510
Hi-Com, 549, 551
inductive loads, 500
inrush current, 503
isolated trigger, 501
latching current, 491, 510
operation, 491
overcurrent, 503
phase angle, 500
phase control, 498, 527, 546
protection, 544
pulse triggering, 492
pulsed gate, 500
quadrants, 491, 510
resistive loads, 498
RFI, 497
saturable choke, 523
series choke, 502
snubber, 495, 502, 523, 529, 549
speed controller, 527
static switching, 497
structure, 489
switching, 489
synchronous control, 497
transformer load, 523
transient protection, 502
trigger angle, 492, 500
triggering, 550
turn-on dI/dt, 502
varistor, 503
zero crossing, 537
Trigger angle, 500
TV & Monitors
16 kHz black line, 351
30-64 kHz autosync, 399
32 kHz black line, 361
damper diodes, 345, 367
diode modulator, 327, 367
EHT, 352 - 354, 368, 409, 410
high voltage bipolar transistor, 339, 341
horizontal deflection, 341
picture distortion, 348
power MOSFET, 339
SMPS, 339, 354, 379, 389, 399
vertical deflection, 358, 364, 402
Two transistor flyback converter, 111, 114
Two transistor forward converter, 111, 117
Universal motor
back EMF, 531
ix
Index Power Semiconductor Applications
Philips Semiconductors
starting, 528
Vacuum cleaner, 527
Varistor, 503
Vertical Deflection, 358, 364, 402
Voltage doubling, 122
Water heaters, 537
Zero crossing, 537
Zero voltage switching, 537
x
Index Power Semiconductor Applications
Philips Semiconductors
flyback converter, 213
forward converter, 213
half bridge converter, 214
power density, 211
push-pull converter, 213
switched mode power supply, 187
switching frequency, 215
transformer construction, 215
Mains Flicker, 537
Mains pollution, 225
pre-converter, 225
Mains transient, 544
Mesa glass, 162
Metal Oxide Varistor (MOV), 503
Miller capacitance, 139
Modelling, 236, 265
MOS Controlled Thyristor, 13
MOSFET, 9, 19, 153, 253
bootstrap, 303
breakdown voltage, 22, 70
capacitance, 30, 57, 72, 155, 156
capacitances, 24
characteristics, 23, 70 - 72
charge, 32, 57
data sheet, 69
dI/dt, 36
diode, 253
drive, 262, 264
drive circuit loss, 156
driving, 39, 250
dV/dt, 36, 39, 264
ESD, 67
gate-source protection, 264
gate charge, 195
gate drive, 195
gate resistor, 156
high side, 436
high side drive, 44
inductive load, 62
lamps, 435
leakage current, 71
linear mode, parallelling, 52
logic level, 37, 57, 305
loss, 26, 34
maximum current, 69
motor control, 259, 429
modelling, 265
on-resistance, 21, 71
package inductance, 49, 73
parallel operation, 26, 47, 49, 265
parasitic oscillations, 51
peak current rating, 251
Resonant supply, 53
reverse diode, 73
ruggedness, 61, 73
safe operating area, 25, 74
series operation, 53
SMPS, 339, 384
solenoid, 62
structure, 19
switching, 24, 29, 58, 73, 194, 262
switching loss, 196
synchronous rectifier, 179
thermal impedance, 74
thermal resistance, 70
threshold voltage, 21, 70
transconductance, 57, 72
turn-off, 34, 36
turn-on, 32, 34, 35, 155, 256
Motor, universal
back EMF, 531
starting, 528
Motor Control - AC, 245, 273
anti-parallel diode, 253
antiparallel diode, 250
carrier frequency, 245
control, 248
current rating, 262
dc link, 249
diode, 261
diode recovery, 250
duty ratio, 246
efficiency, 262
EMC, 260
filter, 250
FREDFET, 250, 259, 276
gate drives, 249
half bridge, 245
inverter, 250, 260, 273
line voltage, 262
loss, 267
MOSFET, 259
Parallel MOSFETs, 276
peak current, 251
phase voltage, 262
power factor, 262
pulse width modulation, 245, 260
ripple, 246
short circuit, 251
signal isolation, 250
snubber, 276
speed control, 248
switching frequency, 246
three phase bridge, 246
underlap, 248
Motor Control - DC, 285, 293, 425
braking, 285, 299
brushless, 301
control, 290, 295, 303
current rating, 288
v
Index Power Semiconductor Applications
Philips Semiconductors
drive, 303
duty cycle, 286
efficiency, 293
FREDFET, 287
freewheel diode, 286
full bridge, 287
half bridge, 287
high side switch, 429
IGBT, 305
inrush, 430
inverter, 302
linear, 457, 475
logic level FET, 432
loss, 288
MOSFET, 287, 429
motor current, 295
overload, 430
permanent magnet, 293, 301
permanent magnet motor, 285
PWM, 286, 293, 459, 471
servo, 298
short circuit, 431
stall, 431
TOPFET, 430, 457, 459, 475
topologies, 286
torque, 285, 294
triac, 525
voltage rating, 288
Motor Control - Stepper, 309
bipolar, 310
chopper, 314
drive, 313
hybrid, 312
permanent magnet, 309
reluctance, 311
step angle, 309
unipolar, 310
Mounting, transistor, 154
Mounting base temperature, 557
Mounting torque, 514
Parasitic oscillation, 149
Passivation, 131, 162
PCB Design, 368, 419
Phase angle, 500
Phase control, 546
thyristors and triacs, 498
triac, 523
Phase voltage
see motor control - ac
Power dissipation, 557
see High Voltage Bipolar Transistor loss,
MOSFET loss
Power factor correction, 580
active, boost converted, 581
Power MOSFET
see MOSFET
Proportional control, 537
Protection
ESD, 446, 448, 482
overvoltage, 446, 448, 469
reverse battery, 452, 473, 479
short circuit, 251, 446, 448
temperature, 446, 447, 471
TOPFET, 445, 447, 451
Pulse operation, 558
Pulse Width Modulation (PWM), 108
Push-pull converter, 111, 119
advantages, 119
clamp diodes, 119
cross conduction, 119
current mode control, 120
diodes, 121
disadvantages, 119
duty ratio, 119
electronic ballast, 582, 587
flux symmetry, 119, 120
magnetics, 213
multiple outputs, 119
operation, 119
output filter, 119
output ripple, 119
rectifier circuit, 180
switching frequency, 119
transformer, 119
transistor voltage, 119
transistors, 121
Qs (stored charge), 162
RBSOA, 93, 99, 135, 138, 139
Rectification, synchronous, 179
Reset winding, 117
Resistor
mains dropper, 544, 545
Resonant power supply, 219, 225
modelling, 236
MOSFET, 52, 53
pre-converter, 225
Reverse leakage, 169
Reverse recovery, 143, 162
RFI, 154, 158, 167, 393, 396, 497, 529, 530,
537
Ruggedness
MOSFET, 62, 73
schottky diode, 173
Safe Operating Area (SOA), 25, 74, 134, 557
forward biased, 92, 99, 134
reverse biased, 93, 99, 135, 138, 139
vi
Index Power Semiconductor Applications
Philips Semiconductors
Saturable choke
triac, 523
Schottky diode, 173
bulk leakage, 174
edge leakage, 174
guard ring, 174
reverse leakage, 174
ruggedness, 173
selection guide, 176
technology, 173
SCR
see Thyristor
Secondary breakdown, 133
Selection Guides
BU25XXA, 331
BU25XXD, 331
damper diodes, 345
EPI diodes, 171
horizontal deflection, 343
MOSFETs driving heaters, 442
MOSFETs driving lamps, 441
MOSFETs driving motors, 426
Schottky diodes, 176
SMPS, 339
Self Oscillating Power Supply (SOPS)
50W microcomputer flyback converter, 199
ETD transformer, 199
Servo, 298
Single ended push-pull
see half bridge converter
Snap-off, 167
Snubber, 93, 139, 495, 502, 523, 529, 549
active, 279
Softness factor, 167
Solenoid
TOPFET, 469, 473
turn off, 469, 473
Solid state relay, 501
SOT186, 154
SOT186A, 154
SOT199, 154
Space charge, 133
Speed-up capacitor, 143
Speed control
thyristor, 531
triac, 527
Starter
fluorescent lamp, 580
Startup circuit
electronic ballast, 591
self oscillating power supply, 201
Static Induction Thyristor, 11
Stepdown converter, 109
Stepper motor, 309
Stepup converter, 109
Storage time, 144
Stored charge, 162
Suppression
mains transient, 544
Switched Mode Power Supply (SMPS)
see also self oscillating power supply
100W 100kHz MOSFET forward converter,
192
100W 500kHz half bridge converter, 153
100W 50kHz bipolar forward converter, 187
16 & 32 kHz TV, 389
asymmetrical, 111, 113
base circuit design, 149
boost converter, 109
buck-boost converter, 110
buck converter, 108
ceramic output filter, 153
continuous mode, 109, 379
control ic, 391
control loop, 108
core excitation, 113
core loss, 167
current mode control, 120
dc-dc converter, 119
diode loss, 166
diode reverse recovery effects, 166
diode reverse recovery softness, 167
diodes, 115, 118, 121, 124, 126
discontinuous mode, 109, 379
epitaxial diodes, 112, 161
flux swing, 111
flyback converter, 92, 111, 113, 123
forward converter, 111, 116, 379
full bridge converter, 111, 125
half bridge converter, 111, 122
high voltage bipolar transistor, 94, 112, 115,
118, 121, 124, 126, 129, 339, 383, 392
isolated, 113
isolated packages, 153
isolation, 108, 111
magnetics design, 191, 197
magnetisation energy, 113
mains filter, 380
mains input, 390
MOSFET, 112, 153, 33, 384
multiple output, 111, 156
non-isolated, 108
opto-coupler, 392
output rectifiers, 163
parasitic oscillation, 149
power-down, 136
power-up, 136, 137, 139
power MOSFET, 153, 339, 384
pulse width modulation, 108
push-pull converter, 111, 119
vii
Index Power Semiconductor Applications
Philips Semiconductors
RBSOA failure, 139
rectification, 381, 392
rectification efficiency, 163
rectifier selection, 112
regulation, 108
reliability, 139
resonant
see resonant power supply
RFI, 154, 158, 167
schottky diode, 112, 154, 173
snubber, 93, 139, 383
soft start, 138
standby, 382
standby supply, 392
start-up, 391
stepdown, 109
stepup, 109
symmetrical, 111, 119, 122
synchronisation, 382
synchronous rectification, 156, 179
TDA8380, 381, 391
topologies, 107
topology output powers, 111
transformer, 111
transformer saturation, 138
transformers, 391
transistor current limiting value, 112
transistor mounting, 154
transistor selection, 112
transistor turn-off, 138
transistor turn-on, 136
transistor voltage limiting value, 112
transistors, 115, 118, 121, 124, 126
turns ratio, 111
TV & Monitors, 339, 379, 399
two transistor flyback, 111, 114
two transistor forward, 111, 117
Switching loss, 230
Synchronous, 497
Synchronous rectification, 156, 179
self driven, 181
transformer driven, 180
Temperature control, 537
Thermal
continuous operation, 557, 568
intermittent operation, 568
non-rectangular pulse, 565
pulse operation, 558
rectangular pulse, composite, 562
rectangular pulse, periodic, 561
rectangular pulse, single shot, 561
single shot operation, 561
Thermal capacity, 558, 568
Thermal characteristics
power semiconductors, 557
Thermal impedance, 74, 568
Thermal resistance, 70, 154, 557
Thermal time constant, 568
Thyristor, 10, 497, 509
’two transistor’ model, 490
applications, 527
asynchronous control, 497
avalanche breakdown, 490
breakover voltage, 490, 509
cascading, 501
commutation, 492
control, 497
current rating, 511
dI/dt, 490
dIf/dt, 491
dV/dt, 490
energy handling, 505
external commutation, 493
full wave control, 499
fusing I
2
t, 503, 512
gate cathode resistor, 500
gate circuits, 500
gate current, 490
gate power, 492
gate requirements, 492
gate specifications, 512
gate triggering, 490
half wave control, 499
holding current, 490, 509
inductive loads, 500
inrush current, 503
latching current, 490, 509
leakage current, 490
load line, 492
mounting, 514
operation, 490
overcurrent, 503
peak current, 505
phase angle, 500
phase control, 498, 527
pulsed gate, 500
resistive loads, 498
resonant circuit, 493
reverse characteristic, 489
reverse recovery, 493
RFI, 497
self commutation, 493
series choke, 502
snubber, 502
speed controller, 531
static switching, 497
structure, 489
switching, 489
viii
Index Power Semiconductor Applications
Philips Semiconductors
switching characteristics, 517
synchronous control, 497
temperature rating, 512
thermal specifications, 512
time proportional control, 497
transient protection, 502
trigger angle, 500
turn-off time, 494
turn-on, 490, 509
turn-on dI/dt, 502
varistor, 503
voltage rating, 510
Thyristor data, 509
Time proportional control, 537
TOPFET
3 pin, 445, 449, 461
5 pin, 447, 451, 457, 459, 463
driving, 449, 453, 461, 465, 467, 475
high side, 473, 475
lamps, 455
leadforms, 463
linear control, 451, 457
motor control, 430, 457, 459
negative input, 456, 465, 467
protection, 445, 447, 451, 469, 473
PWM control, 451, 455, 459
solenoids, 469
Transformer
triac controlled, 523
Transformer core airgap, 111, 113
Transformers
see magnetics
Transient thermal impedance, 559
Transient thermal response, 154
Triac, 497, 510, 518
400Hz operation, 489, 518
applications, 527, 537
asynchronous control, 497
breakover voltage, 510
charge carriers, 549
commutating dI/dt, 494
commutating dV/dt, 494
commutation, 494, 518, 523, 529, 549
control, 497
dc inductive load, 523
dc motor control, 525
dI/dt, 531, 549
dIcom/dt, 523
dV/dt, 523, 549
emitter shorting, 549
full wave control, 499
fusing I
2
t, 503, 512
gate cathode resistor, 500
gate circuits, 500
gate current, 491
gate requirements, 492
gate resistor, 540, 545
gate sensitivity, 491
gate triggering, 538
holding current, 491, 510
Hi-Com, 549, 551
inductive loads, 500
inrush current, 503
isolated trigger, 501
latching current, 491, 510
operation, 491
overcurrent, 503
phase angle, 500
phase control, 498, 527, 546
protection, 544
pulse triggering, 492
pulsed gate, 500
quadrants, 491, 510
resistive loads, 498
RFI, 497
saturable choke, 523
series choke, 502
snubber, 495, 502, 523, 529, 549
speed controller, 527
static switching, 497
structure, 489
switching, 489
synchronous control, 497
transformer load, 523
transient protection, 502
trigger angle, 492, 500
triggering, 550
turn-on dI/dt, 502
varistor, 503
zero crossing, 537
Trigger angle, 500
TV & Monitors
16 kHz black line, 351
30-64 kHz autosync, 399
32 kHz black line, 361
damper diodes, 345, 367
diode modulator, 327, 367
EHT, 352 - 354, 368, 409, 410
high voltage bipolar transistor, 339, 341
horizontal deflection, 341
picture distortion, 348
power MOSFET, 339
SMPS, 339, 354, 379, 389, 399
vertical deflection, 358, 364, 402
Two transistor flyback converter, 111, 114
Two transistor forward converter, 111, 117
Universal motor
back EMF, 531
ix
Index Power Semiconductor Applications
Philips Semiconductors
starting, 528
Vacuum cleaner, 527
Varistor, 503
Vertical Deflection, 358, 364, 402
Voltage doubling, 122
Water heaters, 537
Zero crossing, 537
Zero voltage switching, 537
x

Thyristors and Triacs

Power Semiconductor Applications Philips Semiconductors

Using Thyristors and Triacs

487

.

2 Thyristor circuit symbol and basic structure MT1 MT1 Gate Gate n p n n p n MT2 MT2 Fig. The most complex common thyristor structure is the bidirectional triode thyristor. Its performance is that of a pair of thyristors in anti-parallel with a single gate terminal. power control device. Forward current On-state characteristic I G> 0 Reverse voltage Avalanche breakdown region IL IH IG = 0 Forward voltage V (BO) Reverse characteristic Off-state characteristic Reverse current Fig. the device is a diode thyristor. The simplest thyristor structure.8A to 25A.1. 1 Thyristor static characteristic Anode Anode p Gate n p n J1 J2 J3 Cathode Cathode Fig. This limits thyristor switching circuits to low frequency applications. Its circuit symbol and basic structure are shown in Fig. The triac (shown in Fig. They have very low on-state voltages but. conducting (large reverse currents at low reverse voltages) and approximate mirror image of the forward characteristic (bidirectional thyristors). is the reverse blocking triode thyristor (usually simply referred to as the ’thyristor’ or SCR ’silicon controlled rectifier’). the device is a triode thyristor. Triac gate triggering circuits must be designed with care to ensure that unwanted conduction. because the minority charge carriers in the devices must be removed before they can block an applied voltage. loss of control.1 Introduction to Thyristors and Triacs Brief summary of the thyristor family The term thyristor is a generic name for a semiconductor switch having four or more layers and is.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors 6. the device is a tetrode thyristor. If an ohmic connection is made to both intermediate regions. Triacs are used almost exclusively at mains supply frequencies of 50 or 60Hz. ie. or as non-isolated or isolated discrete devices. There are three types of thyristor reverse characteristic: blocking (as in normal diodes).c. If an ohmic connection is made to the first p region and the last n region. 2. while the currents (IT(RMS)) range from 0. Gate The devices are available as surface mount components. Thyristors form a large family and it is helpful to consider the constituents which determine the type of any given thyristor. the switching times are comparatively long. a p-n-p-n sandwich. in essence. If an additional ohmic connection is made to the intermediate n region (n gate type) or the intermediate p region (p gate type). All such devices have a forward characteristic of the general form shown in Fig. 1. while in some applications this extends up to the 400Hz supply frequency as used in aircraft. depending on the device rating. does not occur when triggering lasts too long. 3) is able to pass current bidirectionally and is therefore an a. 3 Triac circuit symbol and basic structure 489 . or triac. and the most common. and no other connection is made. Reverse blocking devices usually have four layers or less whereas reverse conducting and mirror image devices usually have five layers. The voltage blocking capabilities of thyristors and triacs are quite high: the highest voltage rating for the Philips range is 800V. Thyristors and triacs are both bipolar devices. but this must be large enough to remove the heat caused by bidirectional current flow. The triac needs only one heatsink.

because of the way that the current then latches. The usual method is by a current applied to the gate. although a thyristor switches faster with VBO turn-on than with gate turn-on. Turn-on by leakage current As the junction temperature of a thyristor rises. The breakover voltage of a thyristor will be greater than the rated maximum voltage of the device. The ’two transistor’ model of Fig. When a thyristor is triggered by exceeding VBO the fall time of the forward voltage is quite low (about 1/20th of the time taken when the thyristor is gate-triggered). Base current for T1 is provided by the external gate current in addition to the collector current from T2. When this condition occurs then both T1 and T2 are driven into saturation and the thyristor is said to be ’latched’. Application of current to the thyristor gate initiates the latching mechanism discussed in the previous section. As thyristor triggering characteristics are temperature dependant. The gate current and voltage requirements which ensure triggering of a particular device are always quoted in the device data. 1 showed that the thyristor will switch to its on-state condition with forward bias voltages less than VBO when the gate current is greater than zero. The thyristor must be turned off by using the external circuit to break the regenerative current loop between transistors T1 and T2. however. Thus a thyristor can be switched on by a signal at the gate terminal but. a current will flow in the device to charge the device capacitance according to the relation: Fig. the thyristor cannot be turned off by the gate. junctions J1 and J3 are forward biased. the density of moving current carriers in the device induces switch-on. When the thyristor cathode is more positive than the anode then junctions J1 and J3 are reverse biased and the device blocks. and most circuit designs attempt to avoid its occurrence. With increasing anode current the loop gain increases sufficiently such that the gate current can be removed without T1 and T2 coming out of saturation. The collector of T1 provides the base current for T2. As J2 is reverse biased. IL. called the holding current value. 4 can be used to consider the p-n-p-n structure of a thyristor as the interconnection of an npn transistor T1 and a pnp transistor T2. Eventually. 2. leakage current would become large enough to initiate latching of the regenerative loop of the thyristor and allow forward conduction. Anode i A Thyristor turn-on methods Turn-on by exceeding the breakover voltage When the breakover voltage. If the reverse voltage across J2 is made to reach its avalanche breakdown level then the device conducts like a single forward-biased junction. then the device still blocks forward voltage.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Thyristor operation The operation of the thyristor can be understood from Fig. the leakage current also increases. 4 ’Two transistor’ model of a thyristor iC = C. the permitted di/dt for breakover voltage turn-on is lower. There are several mechanisms by which a thyristor can be latched. dv dt (1) If the charging current becomes large enough. Reverse biasing the device will initiate turn-off once the anode current drops below a minimum specified value. At a certain critical temperature (above Tj(max)) the thyristor will not support any blocking voltage at all. T2 Turn-on by dV/dt Gate i G Cathode T1 Any p-n junction has capacitance . IH. if the junction temperature is allowed to rise sufficiently. The anode to cathode current is then only limited by the external circuit. Turn-on by gate triggering Gate triggering is the usual method of turning a thyristor on. At the breakover voltage the value of the thyristor anode current is called the latching current. 490 . across a thyristor is exceeded. If a voltage ramp is applied across the anode-to-cathode of a p-n-p-n device. The gains of transistors T1 and T2 are current dependent and increase as the current through T1 and T2 increases. When the anode is more positive than the cathode. If the gain in the base-collector loop of T1 and T2 exceeds unity then the loop current can be maintained regeneratively. This gate current starts the regenerative action in the thyristor and causes the anode current to increase. Breakover voltage triggering is not normally used as a triggering method. VBO. the amplitude and duration of the gate pulse must be sufficient to ensure that the thyristor latches under all possible conditions. the thyristor turns on. The characteristic of Fig.the larger the junction area the larger the capacitance. As a general rule.

The main terminals MT1 and MT2 are connected to both p and n regions of the device and the current path through the layers of the device depends upon the polarity of the applied voltage between the main terminals. 5. whilst the gate trigger current is slightly higher in fourth (MT2-. where the term MT2+ denotes that terminal MT2 is positive with respect to terminal MT1. G I + MT2+ G Quadrant 3 Quadrant 4 Fig. However. followed by the remainder of the device. 5 Anti parallel thyristor representation of a triac I The on-state characteristic of the triac is similar to that of a thyristor and is shown in Fig. The single gate terminal is common to both thyristors. the values of latching current (IL). 7 Triac triggering quadrants For applications where the gate sensitivity is critical and where the device must trigger reliably and evenly for applied voltages in both directions it may be preferable to use a negative current triggering circuit. G+) quadrant. the whole active area of the thyristor (or triac) cannot be turned on simultaneously: the area nearest to the gate turns on first. If dIF/dt is excessive then only a limited area of the device will have been turned on as the anode current increases. Gate power dissipation can also be reduced by triggering the thyristor using a pulsed signal. holding current (IH) and gate trigger current (IGT) vary slightly between the different operating quadrants. Operating quadrants for triacs MT2+ Quadrant 2 + + Quadrant 1 I G I - G G+ G- MT1 Fig. the rate of rise of thyristor anode current dIF/dt is determined by the external circuit conditions. 491 . The resulting localised heating of the device will cause degradation and could lead to eventual device failure. A suitably high gate current and large rate of rise of gate current (dIG/dt) ensures that the thyristor turns on quickly (providing that the gate power ratings are not exceeded) thus increasing the thyristor turn-on di/dt capability.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors During gate turn-on. 6. The device polarity is usually described with reference to MT1. 6 Triac static characteristic Quadrant 1 2 3 4 (1+) (1-) (3-) (3+) Polarity of MT2 wrt MT1 MT2+ MT2+ MT2MT2- Gate polarity G+ GGG+ Triac operation The triac can be considered as two thyristors connected in antiparallel as shown in Fig. Goperation) then the triac is never used in the fourth quadrant where IGT is highest. the latching current is slightly higher in the second (MT2+. Due to the physical layout of the semiconductor layers in a triac. Table 1 and Fig. Forward current On-state T2+ IG > 0 Reverse voltage Off-state V (BO) IL IH IG = 0 Forward voltage IH IL IG = 0 IG > 0 V (BO) Off-state T2On-state Reverse current Fig. for any triac. At turn-on it is important that the rate of rise of current does not exceed the specified rating. If the gate drive circuit is arranged so that only quadrants 2 and 3 are used (i. G-) quadrant than the other quadrants. MT2 Table 1.e. Once the thyristor has latched then the gate drive can be reduced or removed completely. 7 summarise the different gate triggering configurations for triacs. In general.

It is important to remember. Reverse current Thyristor commutation A thyristor turns off by a mechanism known as ’natural turn-off’. To ensure bidirectional conduction. one-way conduction (rectification) results when the trigger angle is smaller than the load phase angle.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors For some applications it is advantageous to trigger triacs with a pulsating signal and thus reduce the gate power dissipation. this is because the charge carriers in the thyristor at the time of turn-off take a finite time to recombine. Figure 9 also shows the continuous power rating curve (PG(AV)=0. that is. For a given thyristor type there will be a spread in forward characteristics of gate junctions and a spread with temperature. 8 Diac static characteristic and circuit symbol Gate requirements for triggering To a first approximation. when the main anode-cathode current drops below the holding value. The trigger circuit load line must not encroach into the failure to trigger region shown in Fig. the diac blocks applied voltages in either direction until the breakover voltage. especially with a very inductive load.1 P = 0. If the 5W peak gate power curve is used. The diac voltage then breaks back to a lower output voltage VO. the trigger pulses must continue until the end of each mains half-cycle. When designing a gate circuit to reliably trigger a triac or thyristor the gate signal must lie on a locus within the area of certain device triggering. Data sheets for Philips thyristors and triacs show the variation of VGT and IGT with temperature. 9 Thyristor gate characteristic The gate triggering characteristic is limited by the gate power dissipation. Fig.self commutation or external commutation. The diac is more strictly a transistor than a thyristor. but has an important role in many thyristor and triac triggering circuits. VBO is reached. 492 . that the thyristor will turn on again if the reapplied forward voltage occurs before a minimum time period has elapsed.5W) for a typical device and the peak gate power curve (PGM(max)=5W).0 Fig. The forward characteristic is as shown in Fig. Gate voltage. A device with a relatively insensitive gate will be more immune to false triggering due to noise on the gate signal and also will be more immune to commutating dv/dt turn-on. As shown in the characteristic of Fig.1 = PGM 5 (2) The diac It is also worthwhile to consider the operation and characteristics of the diac in the context of multilayer bipolar devices. I G (A) = 1. 9. the gate-to-cathode junction of a thyristor or triac acts as a p-n diode. For pulsed operation the triggering locus can be increased.5W curve be used to limit the load line of the gate drive circuit. It is manufactured by diffusing an n-type impurity into both sides of a p-type slice to give a two terminal device with symmetrical electrical characteristics. Thyristor turn-off is achieved by two main methods .5 = 0. If single trigger pulses are used. Breakback voltage Forward current I Reverse voltage V (BO) V O (BO) Forward voltage VO I (BO) V (BO) At the other end of the scale. breakover current and breakback voltage as shown in the figure.5W G(AV) IGT Gate current. Important diac parameters are breakover voltage. The minimum voltage and minimum current to trigger all devices (VGT and IGT) decreases with increasing temperature. however. Philips produce ranges of triacs having the same current and voltage ratings but with different gate sensitivities. V G (V) Gate power ratings Gate-cathode characteristic Failure to trigger V GT PGM(max) 5W = = 0. the level below which triggering becomes uncertain is determined by the minimum number of carriers needed in the gate-cathode junction to bring the thyristor into conduction by regenerative action. 8. the duty cycle must not exceed δmax = PG(AV) 0. Sensitive gate triacs are used in applications where the device is driven from a controller IC or low power gate circuit. Continuous steady operation would demand that the 0. 9 if triggering is to be guaranteed.

LC Circuit in parallel with the thyristor Initially the capacitor charges to the supply voltage. dIT/dt. The circuit and device waveforms for this method of commutation are shown in Fig. the thyristor can conduct only during the positive half cycle. It is essential for proper commutation that the resonant circuit be less than critically damped. The thyristor conduction interval is half a resonant cycle. Reverse recovery time is the period during which reverse recovery current flows (t1 to t3 in Fig. 10 Commutation using a series LC circuit IR + I thyristor L E C R External commutation If the supply is an alternating voltage. This time increases with increase of forward current and also increases as the forward current decay rate. when this charging current is greater than the thyristor forward current. The circuit diagram and commutation waveforms are shown in Fig. after one resonant half-cycle of the LC circuit). Fig.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Self Commutation In self-commutation circuits the thyristor will automatically turn off at a predetermined time after triggering. it begins to charge in the opposite direction and. the LC circuit starts to reverse the anode current and turns the thyristor off. the resulting main current excites the resonant circuit. 13) and it is the period between the point at which forward current ceases and the earliest point at which the reverse recovery current has dropped to 10% of its peak value. After half a resonant cycle. Fig. decreases. The energy needed for commutation is delivered by a capacitor included in the commutation circuit. The thyristor naturally switches off at the end of each positive half cycle. 11 Commutation using a parallel LC circuit 493 . such as the resonant cycle of an LC-circuit or the Volt-Second capability of a saturable inductor. 12. When the thyristor is triggered the load current flows but at the same time the capacitor discharges through the thyristor in the forward direction. The thyristor conduction period is determined by a property of the commutation circuit. the thyristor turns off.e. 10 shows the circuit diagram and the relevant waveforms for this arrangement. When the capacitor has discharged (i. It is important to ensure that the duration of a half cycle is greater than the thyristor turn-off time. Reverse recovery In typical thyristors the reverse recovery time is of the order of a few micro-seconds. + I thyristor L E C R R leakage LC circuit in series with the thyristor When the thyristor is triggered. 11. Fig.

the reverse voltage distribution can be seriously affected by mismatch of reverse recovery times. The circuit commutated turn-off time increases with: -junction temperature -forward current amplitude -rate of fall of forward current -rate of rise of forward blocking voltage -forward blocking voltage. The amount of stored charge depends upon the reverse recovery characteristics of the triac. if thyristors are connected in series. it must. Thus the turn-off time is specified for defined operating conditions. Data sheet specifications for triacs give characteristics showing the 494 . It is significantly affected by junction temperature and the rate of fall of anode current prior to commutation (dI(com)/dt). If forward voltage is applied to a thyristor too soon after the main current has ceased to flow. Triac-controlled circuits therefore require careful design in order to ensure that the triac does not fail to commutate (switch off) at the end of each half-cycle as expected. circuit I T dI T dt I R V D V R t 0 t t 1 2 t 3 t 4 Fig. Circuit turn-off time is the turn-off time that the circuit presents to the thyristor. In inductive circuits. As the voltage across the triac passes through zero and starts to increase.c. the triac can conduct irrespective of the polarity of the applied voltage. R Triac commutation Unlike the thyristor. the thyristor will turn on. In resistive load applications (e. such as motor control applications or circuits where a dc load is controlled by a triac via a bridge rectifier. dV D dt Fig. 13). Following high rates of change of current the capacity of the triac to withstand high reapplied rates of change of voltage is reduced. lamp loads) current surges at turn-on or during temporary over-current conditions may introduce abnormally high rates of change of current which may cause the triac to fail to commutate.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Reverse recovery current can cause high values of turn-on current in full-wave rectifier circuits (where thyristors are used as rectifying elements) and in certain inverter circuits. it is usually necessary to protect the triac against unwanted commutation due to dv(com)/dt.g. It should also be remembered that. i thyristor Turn-off time Turn-off time is the interval between the instant when thyristor current reverses and the point at which the thyristor can block reapplied forward voltage (t1 to t4 in Fig. It is important to consider the commutation performance of devices in circuits where either dI/dt or dV/dt can be large. 12 Thyristor commutation in an a. Thus the triac does not experience a circuit-imposed turn-off time which allows each anti-parallel thyristor to fully recover from its conducting state as it is reverse biased. be greater than the thyristor turn-off time. then the alternate thyristor of the triac can fail to block the applied voltage and immediately conduct in the opposite direction. of course. 13 Thyristor turn-off characteristics The commutating dv(com)/dt limit for a triac is less than the static dv/dt limit because at commutation the recently conducting portion of the triac which is being switched off has introduced stored charge to the triac.

For the case of an inductive load the current in the triac does not fall to its holding current level until some time later. the ability of a triac to withstand high rates of rise of reapplied voltage is improved by limiting the di/dt using a series inductor. 495 . This is shown in Fig. The device characteristics which determine gate triggering requirements of thyristors and triacs have been presented. -dI/dt Voltage across triac Trigger pulses Conclusions This article has presented the basic parameters and characteristics of triacs and thyristors and shown how the structure of the devices determines their operation. because commutating dv/dt turn-on is dependent upon the rate of fall of triac current. Subsequent articles in this chapter will deal with the use. Current Supply voltage Load current Time dV /dt com VDWM Time The usual method is to place a dv/dt-limiting R-C snubber in parallel with the triac. Important turn-on and turn-off conditions and limitations of the devices have been presented in order to demonstrate the capabilities of the devices and show the designer those areas which require careful consideration. This topic is discussed more fully in the section entitled ’Using thyristors and triacs’. then in circuits with large rates of change of anode current. Time Fig. At the time that the triac current has reached the holding current the mains voltage has risen to some value and so the triac must immediately block that voltage.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors maximum allowable rate of rise of commutating voltage against device temperature and rate of fall of anode current which will not cause a device to trigger. and will present some detailed design and operational considerations for thyristors and triacs in phase control and integral cycle control applications. Additionally. 14. The rate of rise of blocking voltage following commutation (dv(com)/dt) can be quite high. operation and limitations of thyristors and triacs in practical applications. 14 Inductive load commutation with a triac Consider the situation when a triac is conducting in one direction and the applied ac voltage changes polarity.

Thyristors and Triacs

Power Semiconductor Applications Philips Semiconductors

6.1.2 Using Thyristors and Triacs
This chapter is concerned with the uses, operation and protection of thyristors and triacs. Two types of circuit cover the vast majority of applications for thyristors and triacs: static switching circuits and phase control circuits. The characteristics and uses of these two types of circuit will be discussed. Various gate drive circuits and protection circuits for thyristor and triacs are also presented. The use of these circuits will enable designers to operate the devices reliably and within their specified limits. Circuit-generated RFI can be almost completely eliminated by ensuring that the turn-on switching instants correspond to the zero-crossing points of the a.c. mains supply. This technique is known as synchronous (or zero voltage) switching control as opposed to the technique of allowing the switching points to occur at any time during the a.c. cycle, which is referred to as asynchronous control. In a.c. circuits using thyristors and triacs the devices naturally switch off when the current falls below the device holding current. Thus turn-off RFI does not occur.

Thyristor and triac control techniques
There are two main techniques of controlling thyristors and triacs - on-off triggering (or static switching) and phase control. In on-off triggering, the power switch is allowed to conduct for a certain number of half-cycles and then it is kept off for a number of half-cycles. Thus, by varying the ratio of "on-time" to "off-time", the average power supplied to the load can be controlled. The switching device either completely activates or deactivates the load circuit. In phase control circuits, the thyristor or triac is triggered into conduction at some point after the start of each half-cycle. Control is achieved on a cycle-by-cycle basis by variation of the point in the cycle at which the thyristor is triggered.

Asynchronous control
In asynchronous control the thyristor or triac may be triggered at a point in the mains voltage other than the zero voltage crossover point. Asynchronous control circuits are usually relatively cheap but liable to produce RFI.

Synchronous control
In synchronous control systems the switching instants are synchronised with zero crossings of the supply voltage. They also have the advantage that, as the thyristors conduct over complete half cycles, the power factor is very good. This method of power control is mostly used to control temperature. The repetition period, T, is adjusted to suit the controlled process (within statutory limits). Temperature ripple is eliminated when the repetition period is made much smaller than the thermal time constant of the system. Figure 1 shows the principle of time-proportional control. RFI and turn-on di/dt are reduced, and the best power factor (sinusoidal load current) is obtained by triggering synchronously. The average power delivered to a resistive load, RL, is proportional to ton/T (i.e. linear control) and is given by equation 1.

Static switching applications
Thyristors and triacs are the ideal power switching devices for many high power circuits such as heaters, enabling the load to be controlled by a low power signal, in place of a relay or other electro-mechanical switch. In a high power circuit where the power switch may connect or disconnect the load at any point of the mains cycle then large amounts of RFI (radio frequency interference) are likely to occur at the instants of switching. The large variations in load may also cause disruptions to the supply voltage. The RFI and voltage variation produced by high power switching in a.c. mains circuits is unacceptable in many environments and is controlled by statutory limits. The limits depend upon the type of environment (industrial or domestic) and the rating of the load being switched. RFI occurs at any time when there is a step change in current caused by the closing of a switch (mechanical or semiconductor). The energy levels of this interference can be quite high in circuits such as heating elements. However, if the switch is closed at the moment the supply voltage passes through zero there is no step rise in current and thus no radio frequency interference. Similarly, at turn-off, a large amount of high frequency interference can be caused by di/dt imposed voltage transients in inductive circuits.

Pout =

2 V(RMS) ton . RL T

(1)

where: T is the controller repetition period ton is controller ’on’ time V(RMS) is the rms a.c. input voltage. Elsewhere in this handbook the operation of a controller i.c. (the TDA1023) is described. This device is specifically designed to implement time-proportional control of heaters using Philips triacs.

497

Thyristors and Triacs

Power Semiconductor Applications Philips Semiconductors

Triac

Supply voltage

Input voltage

Voltage

O = wt

Trigger signal
Triac Current

Output current
O = wt

tON T
Trigger

Fig. 1 Synchronous time-proportional control
O = wt

Phase control
Phase control circuits are used for low power applications such as lamp control or universal motor speed control, where RFI emissions can be filtered relatively easily. The power delivered to the load is controlled by the timing of the thyristor (or triac) turn-on point. The two most common phase controller configurations are ’half wave control’, where the controlling device is a single thyristor and ’full wave control’, where the controlling device is a triac or a pair of anti-parallel thyristors. These two control strategies are considered in more detail below:

Device triggers

Trigger angle,

Conduction angle,

Fig. 2 Phase controller - resistive load
IT(AV) = 2.IT(MAX) = 0.637 IT(MAX) π IT(MAX)

IT(RMS) =

 √2

= 0.707 IT(MAX)

(2)

where
IT(MAX) = VT(MAX) √2 V(RMS)  = RL RL (3)

Resistive loads
The operation of a phase controller with a resistive load is the simplest situation to analyse. Waveforms for a full wave controlled resistive load are shown in Fig. 2. The triac is triggered at angle δ, and applies the supply voltage to the load. The triac then conducts for the remainder of the positive half-cycle, turning off when the anode current drops below the holding current, as the voltage becomes zero at θ=180˚. The triac is then re-triggered at angle (180+δ)˚, and conducts for the remainder of the negative half-cycle, turning off when its anode voltage becomes zero at 360˚. The sequence is repeated giving current pulses of alternating polarity which are fed to the load. The duration of each pulse is the conduction angle α, that is (180-δ)˚. The output power is therefore controlled by variation of the trigger angle δ. For all values of α other than α=180˚ the load current is non-sinusoidal. Thus, because of the generation of harmonics, the power factor presented to the a.c. supply will be less than unity except when δ=0. For a sinusoidal current the rectified mean current, IT(AV), and the rms current, IT(RMS), are related to the peak current, IT(MAX), by equation 2.

From equation 2 the ’crest factor’, c, (also known as the ’peak factor’) of the current waveform is defined as:
Crest factor, c = IT(MAX) IT(RMS) (4)

The current ’form factor,’ a, is defined by:
Form factor, a = IT(RMS) IT(AV) (5)

Thus, for sinusoidal currents:
a= IT(RMS) = 1.111; IT(AV) c= IT(MAX) = 1.414 IT(RMS) (6)

For the non-sinusoidal waveforms which occur in a phase controlled circuit, the device currents are modified due to the delay which occurs before the power device is triggered. The crest factor of equation 4 and the form factor of equation 5 can be used to describe variation of the current waveshape from the sinusoidal case.

498

 π   IT(RMS) max = IT(MAX) 0.c mains. 3 and 4. for equations 7 and 9 are plotted in Fig. For each case the maximum value occurs when α=180˚ (α=π radians). The variation of rectified mean current. The load current waveform is given in Fig.B. (1 − cosα) 2 IT(AV) max = IT(MAX) π a) b) Supply voltage Thyristor Voltage I T(MAX) Thyristor Current Trigger 1  α − 1 sin 2α  2 2  IT(RMS) = IT(RMS) max. 3b). rms current.4  √2 0.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Half wave controller Figure 3a) shows the simplest type of thyristor half-wave phase controller for a resistive load. T(AV)max Fig. (1 − cosα) 2 IT(AV) max = 2IT(MAX) π 0.B. or a triac. π   IT(RMS) max = IT(MAX) 2 a) Fig.2  α − sin 2α  2 P(out) = P(out) max. IT(AV)/IT(AV)max. rms load current.   π  1 P(out) max = 2 IT(MAX)RL 2 0 (9) 0 I T(AV) I 30 60 I T(RMS) IT(RMS)max 90 P (OUT) P (OUT)max 120 150 Conduction angle 180 N. For each case the maximum value occurs when α=180˚ (α=π radians). 5 Current and power control using conduction angle 499 . and power. IT(AV) = IT(AV) max. IT(RMS). When using equation 7 all values of α must be in radians.571. and load power with trigger angle are given by equation 9. 1 Full wave controller Figure 4 shows the circuit and load current waveforms for a full-wave controller using two antiparallel thyristors.   π  P(out) max = I 2 T(MAX) L Supply voltage R 4 (7) Triac Voltage N. with trigger angle are given in equation 7. as the controlling device. rms current IT(RMS)/IT(RMS)max. P(out)/P(out)max. IT(AV). At α=180˚ the crest factor and form factor for a half wave controller are given by: a= IT(RMS) = 1. 4 Full wave control The variation of normalised average current. The variation of average load current.0 IT(RMS) (8) I T(MAX) Triac Current I T(MAX) Trigger Fig.8 1 Amplitude 0. IT(AV) c= IT(MAX) = 2. IT(AV) = IT(AV) max. Figure 6 shows the variation of current form factor with conduction angle for the half wave controller and the full wave controller of Figs. IT(RMS) and load power over the full period of the a. When using equation 9 all value of α must be in radians. 5. 3 Half wave control b)  α − 1 sin 2α  2 P(out) = P(out) max.6  α − 1 sin 2α  2 2  IT(RMS) = IT(RMS) max. IT(AV).

9a). In Fig. eventually saturating the load inductance. However. is greater than the load phase angle. 8. When the trigger angle. This can be achieved using a delay network of the type shown in Fig. 500 . sinusoidal load current. maximum output. Figure 10 shows several alternative gate drive circuits suitable for typical triac and thyristor applications. This problem can be avoided by using a trigger pulse train as shown in Fig. If the trigger angle is less than the phase angle then the load current in one direction will not have fallen back to zero at the time that the device is retriggered in the opposite direction. that is. δ. 7 the triac is only triggered by the gate pulses when the applied supply voltage is positive (1+ quadrant). Triac Voltage Supply voltage Triac Current Fails to trigger Trigger Device triggers Device triggers Conduction angle Fig.single pulse Inductive loads The circuit waveforms for a phase controller with an inductive load or an active load (for example. This is shown in Fig.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors 6 Form factor 5 4 3 Half-wave rectifier 2 Full-wave rectifier 1 0 Triac Voltage Supply voltage I T(RMS) I T(AV) Triac Current Inductor iron core saturation 0 30 60 90 120 150 Conduction angle 180 Trigger Fig. 9b). In each circuit the gate-cathode resistor protects the device from false triggering due to noise . The trigger pulse train must cease before the mains voltage passes through zero otherwise the triac will continue to conduct in the reverse direction. a thyristor or triac can be triggered into conduction when a voltage of the appropriate polarity is applied across the main terminals and a suitable current is applied to the gate. 7 Triac triggering signals . The triac fails to be triggered as the gate pulse has finished and so the triac then acts as a rectifier. ϕ. The circuit waveforms depend on the load power factor (which may be variable) as well as the triggering angle.this is especially important for sensitive gate devices. the gate pulses which occur one half period later have no effect because the triac is still conducting in the opposite direction.pulse train Gate circuits for thyristors and triacs As discussed in the introductory article of this chapter. a motor) are more complex than those for a purely resistive load. occurs when the trigger angle equals the phase angle. For a bidirectional controller (i. then the load current will become discontinuous and the triac (or thyristor) will block some portion of the input voltage until it is retriggered. Thus unidirectional current flows in the main circuit. 8 Triac triggering signals . This gives a trigger circuit which is suitable for both thyristors and triacs. In addition opto-isolated thyristor and triac drivers are available which are compatible with the Philips range of devices.e triac or pair of anti-parallel thyristors). Greater triggering stability and noise immunity can be achieved if a diac is used (see Fig. The triac triggers on the first gate pulse after the load current has reached the latching current IL in the 3+ quadrant. 6 Variation of form factor with conduction angle Device triggers Device fails to trigger Conduction angle Fig. 7.

This type of circuit is also known as a solid state relay (SSR). A typical solution which involves triggering the smaller device (BT169) from a logic-level controller to turn on the larger device (BT151) is shown in Fig. 11 Master-slave thyristor triggering circuit Fig. IR R + BT169 BT151 b) IR E R R Fig. 9 Basic triac triggering circuits In some applications it may be necessary to cascade a sensitive gate device with a larger power device to give a sensitive gate circuit with a high power handling capability. Figure 12 shows an isolated triac triggering circuit suitable for zero voltage switching applications. 12 Opto-isolated triac triggering circuit 501 . If the BT169 is off then no gate signal is applied to the triac and the load is switched off. The function of R2 R4 Q1 R3 BC547 1K0 BT169 100R R1 + BT138 100R 100nF Fig. 11.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Load 12V 220R Load 180R BC337 1k0 BT145 10k 1k0 BT145 Load 12V 4k7 100nF 12V Load 2:1 BAW62 BT145 1k0 1k0 BT145 10k BC337 10k Fig. 10 Alternative triac triggering circuits a) + E IR the Q1/R2/R3 stage is that the BC547 is on at all instants in time when the applied voltage waveform is high and thus holds the BT169 off.

g.c. Voltage transient protection There are three major sources of transient which may affect thyristor and triac circuits: -the mains supply (e. cosϕ. if the choke is only required to reduce the dv/dt across non-conducting devices then the inductance needs only to be maintained up to quite low currents.c. and the rate of fall of triac current is limited only by the stray inductance in the a. 502 .5W.a snubber network across the device. frequency of the a. For circuits where the load power factor. Alternatively.c circuit will limit the di(com)/dt to an acceptable level. and so for this reason it is usually an air-cored component. ≥ 0. circuit. The snubber resistor needs to be rated at 0. a choke between the power device and external circuit or an overvoltage protection such as a varistor. An alternative topology which avoids triac commutation problems is to control the load on the d. The snubber capacitor should be rated for the full a. The large value of commutating di/dt may cause the triac to retrigger due to  √C 3L (9) where: L is the load inductance f is the supply frequency IT(RMS) is the rms device current dv(com)/dt is the device commutating dv/dt rating. Usually only a few microhenries of inductance are required to limit the circuit di/dt to an acceptable level. lightning) -other mains and load switches (opening and closing) -the rectifying and load circuit (commutation) In order to ensure reliable circuit operation these transients must be suppressed by additional components. 13 Triac protection The following equations can be used to calculate the values of the snubber components required to keep the reapplied dv/dt for a triac within the dv(com)/dt rating for that device. commutating dv(com)/dt.7 the snubber values are given approximately by:  fIT(RMS)  2 C ≥ 25L    dV(com)/dt  R= Series line chokes A series choke may be used to limit peak fault currents to assist in the fuse protection of thyristors and triacs. The presence of a snubber across the device can improve the turn-on performance of the triac by using the snubber capacitor discharge current in addition to the load current to ensure that the triac latches at turn-on. The value of the snubber resistor needs to be large enough to damp the circuit and avoid voltage overshoots. removed at source or allowed for in component ratings. supply with 20µH source inductance gives a maximum di/dt of (220√2)/20=16A/µs. side. If this occurs when the mains voltage is high then Q1 remains on. they have very little effect on the inrush current.c.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors If the input signal is switched high then the photo-transistor turns on. Chokes used to soften commutation should preferably be saturable so as to maintain regulation and avoid deterioration of the power factor.c. The value of the snubber resistor must be large enough to limit the peak capacitor discharge current through the triac to within the turn-on di/dt limit of the device. A small choke in the a. Three types of circuit are commonly employed to suppress voltage transients . Ferrite-cored chokes may be adequate provided that the windings are capable of carrying the full-load current. Choke Load Varistor Snubber Fig. a 220V a. The triac is retriggered every half cycle. When the line voltage passes through its next zero crossing in either direction the photo transistor ensures that Q1 stays off long enough for the BT169 to trigger. The addition of di/dt limiting chokes is especially important in triac circuits where the load is controlled via a bridge rectifier. At the voltage zero-crossing points the conduction transfers between diodes in the bridge network. This protects the devices from turning on too quickly and avoids potential device degradation. For instance. As their impedance reduces at high current. Once the thyristor turns on. The parameters which affect the choice of snubber components are the value of load inductance. This then turns the triac on. If the choke is used in conjunction with fuse protection. supply and rms load current. This is particularly important when considering the commutation behaviour of triacs. the drive circuit is deprived of its power due to the lower voltage drop of the BT169. Snubber networks Snubber networks ensure that the device is not exposed to excessive rates of change of voltage during transient conditions. voltage of the system. it must retain its inductance to very large values of current.c. which has been discussed elsewhere.

Transformer inrush currents are avoided by adjusting the initial trigger angle to a value roughly equal to the load phase angle. and the amount of fuse I2t to clear the circuit must be less than the I2t rating of the triac. However. triacs have an infinite life if they are used within their ratings. The amount of device protection required will depend upon the conditions imposed on the device by the application circuit. they have a current limiting action in the event of a short-circuit. The type of circuit used depends upon the degree of control required and the nature of the load. Short-circuit condition Fuses for protecting triacs should be fast acting. The protection circuits presented here will be suitable for the majority of applications giving a cheap. 503 . incandescent lamp or transformer loads give rise to an inrush condition. Conclusions This paper has outlined the most common uses and applications of thyristor and triac circuits. Several types of gate circuit and device protection circuit have been presented. Inrush condition Motors. as shown in Fig. 13. High voltage fuses exhibit low clearing I2t but the fuse arc voltage may be dangerous unless triacs with a sufficiently high voltage rating are used. fuses) must. they rapidly overheat when passing excessive current because the thermal capacitance of their junction is small. protects the device from transient overvoltages which may occur due to mains disturbances. No damage occurs when the amount of inrush current is below the inrush current rating curve quoted in the device data sheet (see the chapter ’Understanding thyristor and triac data’). efficient overall design which uses the device to its full capability with complete protection and confidence. Lamp and motor inrush currents are avoided by starting the control at a large trigger angle. be fast-acting. Because the fuses open the circuit rapidly. Overcurrent protective devices (circuit breakers.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Varistor The use of a metal oxide varistor (MOV). therefore. Overcurrent protection Like all other semiconductor devices.

.

505 . To overcome this. other ’rules’ have been derived. This section will investigate the effect of pulse duration on the peak current capability of thyristors. and is derived from the surge current from:  ITSM  2  0. If the period over which the energy is delivered is long. Unfortunately little or no real information currently exists to indicate the validity of these rules. In fact it is not the passage of the energy which causes damage. BT152 and BT145 . For very short pulses (<0. This figure assumes a half sine pulse with a width of either 10 ms or 8.to gather the data which would.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors 6. experience and experiments have shown that such an approach is inaccurate. but the heating of the crystal by the energy absorbed by the device which causes damage. decide which was correct. however. the device is subjected to a series of current pulses of increasing magnitude until it receives a surge which causes measurable degradation. However. Tests have been performed on three groups of devices .1 ms) and large crystal.1. This limit is not absolute. I pk = ITSM  0. Test Circuit The technique chosen to measure the peak current capability of the devices was the stepped surge method. This section will discuss some of the factors affecting a thyristor’s peak current capability and review the existing prediction methods.3 The Peak Current Handling Capability of Thyristors The ability of a thyristor to withstand peak currents many times the size of its average rating is well known. Data sheets for thyristors always quote a figure for the maximum surge current that the device can survive. BT152 and BT145 thyristors.like the edges of the crystal. In this test. assuming that the surge is a half sine pulse. the problem is even worse because not all of the active area of a thyristor crystal is turned on simultaneously . which are the conditions applicable for 50/60 Hz mains operation.say a single half sine pulse of current with a duration of <10 ms .01  N    tp  where is N is either constant or a function of the pulse width.the areas to which the energy can spread for the actual duration of the pulse are limited. the plastic encapsulation. It will go on to present the results of an evaluation of the peak current handling capabilities for pulses as narrow as 10 µs for the BT151. It will also propose a method for estimating a thyristor’s peak current capability for a half sine pulse with a duration between 10 µs and 10 ms from its quoted surge rating. hopefully. This number is used to select appopriate fuses for device protection. the absorbed energy has time to spread to all areas of the device capable of storing it .01 I 2t =   √2   This calculates the RMS current by dividing ITSM by √2  Under the simplest of analyses I2t would be assumed to be constant so a device’s peak current capability could be calculated from: 1 I pk = ITSM  0.3 ms.so the current pulse passes through only part of the crystal resulting in a higher level of dissipation and an even more restricted area for absorbing it. However. there is little information about the factors affecting the peak current capability. I2t represents the energy that can be passed by the device without damage.BT151. Another suggestion is that the ’constancy’ continuously changes from I2t to I4t as the pulses become shorter.therefore the temperature rise in the crystal is moderated. narrow pulses with much higher peaks can be handled without damage but little information is available to enable the designer to determine how high this current is.01  2    tp  where Ipk is the peak of a half sine current pulse with a duration of tp. data sheets often quote a figure called "I2t for fusing". for example:  1 N = log   tp  The graph shown in Fig. 1 shows what several of these ’rules’ predict would happen to the peak current capability if they were true. the delivery period is short . This means that the crystal keeps all the energy giving a much bigger temperature rise.conduction tends to spread out from the gate area . One of these ’rules’ suggests that it is not I2t which is constant but I3t or I4t. Expected Results I2t is normally quoted at 10 ms. If. the mounting tab and for very long times the heatsink . All these rules are expressed in the general equation: 1 Energy Handling In addition to the maximum surge current.

Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors 15 14 Choice of L & C The width of the half sine pulse from an LC circuit is: tpulse = π √C L Peak Current Multiplying Factor 13 12 11 10 9 8 7 6 5 4 3 2 1 10us 100us 1ms 10ms and the theoretical peak value of the current is: I peak = V  √L C Width of Half Sine Pulse I t = const. Experimentation indicated that the clearest indication of device damage was obtained by looking for changes in the off-state breakdown voltage. 2. Minimising these effects was considered to be important so care was taken during the building of the circuits to keep the resistance to a minimum. It was decided that steps of approximately 5% would give reasonable resolution. Inductor and Capacitor Values Pushbutton S1 and resistor R2 are a safety feature. 1 Predicted ITSM multiplying factors These equations assume that the circuit has no series resistance to damp the resonant action which would result in a longer but lower pulse. It was decided to test the devices at three different pulse widths . since this work was attempting to determine the current that a device could survive . R2 keeps C discharged until S1 is pressed. The resolution with which the current capability is assessed is defined by the size of each increase in current. The values were selected with the help of a ’spreadsheet’ program running on an PC compatible computer. Pulse Width C (µF) 13.so three sets of L and C were needed. the inductors were wound using heavy gauge wire and the loop C / L / DUT / R3 was kept as short as possible. To this end capacitors with low ESR were chosen. 3 I t = const. 4 Fig. the test method called for each device to be subjected to a series of current pulses of increasing amplitude.not which killed it. The pulse is monitored by viewing the voltage across R3 on an digital storage oscilloscope. 506 . D attempts to prevent any high voltage spikes being fed back into the PSU. The magnitude of the current pulse is adjusted by changing the voltage to which C is initially charged by varying the output of the PSU. This procedure did slow the testing but it was felt that it would result in greater accuracy. Also given in Table 1 are the theoretical peak currents that the L / C combination would produce for a initial voltage on C of 600 V.75 10 154 Ipeak (A) 2564 1885 1244 D R1 S1 L Vak DC PSU DUT R2 C R3 Ia Trigger Pulse 10 µs 100 µs 1 ms 0-600V Fig. 100 µs and 1 ms . log(1/t) I t = const. R1 limits the current from the supply when DUT fails and during the recharging of C. Circuit Description The circuits used to perform the required measurements were of the form shown in Fig. 2 I t = const. So after each current pulse the DUT was removed from the test circuit and checked on a curve tracer. Triggering of the device under test (DUT) itself is used to initiate the discharge.6 100 660 L (µH) 0. R1 and D protect the power supply. the figure actually quoted in the results for a device’s current capability would be the value of the pulse prior to the one which caused damage. 2 Surge current test circuit Table 1. They produce half sine pulses of current from the resonant discharge of C via L. Test Procedure As mentioned earlier. The values which were finally chosen are shown in Table 1. It was also decided that. The gate signal used for all the tests was a 100 mA / 1 µs pulse fed from a pulse generator in single-shot mode. The trigger pulse needs a button on the pulse generator to be pressed which means both hands are occupied and kept away from the test circuit high voltages.10 µs.

the conduction region tends to spread out from the gate. Mean Peak Current Capability (Amps) Pulse Width 10 µs 100 µs 1 ms BT151 912 595 264 BT152 1092 1021 490 BT145 1333 1328 697 Further study of Fig.but not necessarily equal to the time taken to turn on all the active area of the crystal and is calculated from:tcrit = A R Table 3. Table 3 expresses the mean values as factors of the device ITSM rating.6 5.0 3. Preferably. then the current flows through only part of the crystal. This table also gives the factors that the various ’rules’ would have predicted for the various pulse widths. This could be explained by a reduction in the active area of the larger crystals. This will inevitably introduce an error because cathode and crystal areas are not directly proportional.6 10. This is consistent with the known fact that not all areas of a thyristor turn on simultaneously . However. As an alternative the total crystal area can be used if the value of R is adjusted accordingly. This implies that something else will have to be taken into account. may still be correct but that the performance it predicts will ’roll off’ if the pulse duration is less than some critical value. 3 Peak current capability measurements Test Results Figure 3 is a graph showing the measured current capabilities of all of the tested devices. but it should be relatively small.6 4. an inspection of Table 3 clearly shows that there is no 507 . Interpretation of Results It had been hoped that the measurements would give clear indication of which of the ’rules’ would give the most accurate prediction of performance. If the rate at which the conduction area turns on is constant then the time taken for a small device to be completely ON is shorter than for a large device. In fact the variation in the factors between the various device types would indicated that no rule based on an Int function alone can give an accurate prediction.5 5.8 where tcrit is proportional to .4 10.2 1. Measured Current Capability Measured Factor Pulse Width 10 µs 100 µs 1 ms Predicted Factor (by Int rule) n= log(1/t) 4.2 2.1 2.1 6.01  2  tp       tp   tp + tcrit  Table 2. A should be the area of the cathode but this information is not always available.2 which simplifies to:I pk = ITSM   √  0. Proposed Prediction Method The above interpretation leads one to believe that the original energy handling rule.2 2. making them appear to be smaller than they actually are.3 3.0 5.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Peak Current (amps) @ 1000 @ # * * # @ * * * * * correlation between any of the predicted factors and the measured factors.01     tp + tcrit  BT BT BT n=2 n=3 n=4 151 152 145 9. 3 reveals that the difference in the peak current capability of the three device types is becoming less as the pulses become shorter.0 4. which says that I2t is a constant.4 31. Table 2 summarises the measurements by giving the mean of the results for the three device types at each of the pulse widths. Measured and Predicted ITSM Multiplication Factors where: A = crystal area R = constant expressing the rate at which the area is turned on. The equation which was developed to have the necessary characteristics is: 1 1 I pk = ITSM 2  0. This would explain why the performance increase of the BT145 starts falling off before that of the BT151.4 4. If the pulse duration is less than the time it takes for all areas of the device to turn on.6 2. * * @ @ # # # # # # 100 10us 100us 1ms 10ms Width of half-sine pulse # BT151 @ BT152 * BT145 Fig.2 3. reducing the effective size of the device.0 2.

5A. Device BT151 BT152 BT145 tcrit 148 µs 410 µs 563 µs In this section. is capable of conducting. an equation has been proposed which takes crystal size into account by using it to calculate a factor called tcrit. Using these values in the above equation predicts that the peak current handling capability of the BT151. BT152 and BT145 would be as shown in Fig. Table 3. Calculated Values of tcrit Conclusions The first conclusion that can be drawn from this work is that a thyristor. a peak current greater than 100 times this value in a short pulse.02 m2/s Using this value of R gives the values of tcrit shown in Table 3.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors R was determined empirically to be approximately 0. The results obtained using the test methods indicate that the previously suggested ’rules’ fail to take into account the effect that crystal size has on the increase in performance. Peak Current (amps) 1000 100 10us 100us 1ms 10ms Width of half-sine pulse BT151 BT152 BT145 Fig. This capability has always been known and indeed the surge rating given in the data sheet gives a value for it at pulse widths of around 10 ms. This results in what is believed to be a more accurate means of estimating the capability of a device for a half sine pulse with a duration between 10 µs and 10 ms. This time is then used to ’roll off’ the performance increase predicted by the original energy handling equation . What has been missing is a reliable method of predicting what the peak current capability of a device is for much shorter pulses. 4. with average rating of only 7. Furthermore the power required to trigger the device into conducting this current can be <1 µW. 4 Predicted peak current handling using ’Rolled-off I2t’ rule 508 .I2t = constant. without damage.

having stable on and off states. The thyristor will conduct a load current in one direction only. the anode is connected to the metal tab. However. that is. This present article describes the data sheet descriptions of Philips thyristors and triacs. Only if data is both complete and unambiguous can a true comparison be made between the capabilities of different types. Published ratings and characteristics require supporting information to truly describe the capabilities of devices. turn-on can occur when the forward (anode-to-cathode) voltage is less than V(BO) if the thyristor is triggered by injecting a pulse of current into the gate. for instance. If the device is to remain in the on state. and four-terminal devices are known as silicon controlled switches.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors 6. If a positive voltage is applied. three. However. Once the on state is established. 1 Thyristor circuit symbol and basic structure Forward current On-state characteristic Reverse voltage Avalanche breakdown region IL IH Forward voltage V (BO) Reverse characteristic Off-state characteristic Reverse current Fig. or three-terminal. thyristor is also known as the reverse blocking triode thyristor or the silicon controlled rectifier (SCR). then again a small forward leakage current flows which increases as the forward voltage increases. Manufacturers have been known to quote ratings in such a way as to give a false impression of the capabilities of their devices. only a small reverse leakage current flows. Ratings and characteristics given in published data should always be quoted with the conditions to which they apply. Fig. the thyristor will only conduct this load current when it has been ’triggered’. Anode Anode p Gate Gate n p n Thyristors Thyristor is a generic term for a semiconductor device which has four semiconductor layers and operates as a switch. 2 Thyristor static characteristic Thyristors are normally turned on by triggering with a gate signal but they can also be turned on by exceeding either the forward breakover voltage or the permitted rate of rise 509 . this trigger pulse must remain until the current through the thyristor exceeds the latching current IL. is clear. the holding current IH is the minimum current that can flow through the thyristor and still maintain conduction. turn-on is initiated by avalanche breakdown and the voltage across the thyristor falls to the on state voltage VT. When a small negative voltage is applied to the device. As the reverse voltage is increased. 2 shows the static characteristic of the thyristor. forward voltage reaches the breakover voltage V(BO). A thyristor can have two. or four terminals but common usage has confined the term thyristor to three terminal devices. as will a rectifier diode. together with the advantages of the absolute maximum rating system. A brief survey of short-form catalogues is an insufficient method of comparing different devices. When the Cathode Cathode Fig. the leakage current increases until avalanche breakdown occurs. Fig.1. 1 shows the circuit symbol and a schematic diagram of the thyristor. and these conditions should be those likely to occur in operation. thus comparisons between devices whose performance appears to be similar should not be made on economic grounds alone. this is the essential property of the thyristor. The common.4 Understanding Thyristor and Triac Data The importance of reliable and comprehensive data for power semiconductor devices. Two-terminal devices are known as switching diodes. and aims to enable the circuit designer to use our published data to the full and to be confident that it truly describes the performance of the devices. All Philips thyristors are p-gate types. by reducing the voltage across the thyristor and load to zero. Furthermore. it is important to define the rating or characteristic being quoted. The load current must be reduced to below IH to turn the thyristor off.

or bidirectional triode thyristor. The triac has two main terminals MT1 and MT2 (the load connections) and a single gate. is a device that can be used to pass or block current in either direction. This is the allowable peak value of non-repetitive voltage transients. as shown in Table 1. like the thyristor. Thus the triac saves both cost and space in a.c. mains is usually regarded as a smooth sinewave. The condition when terminal 2 of the triac is positive with respect to terminal 1 is denoted in data by the term ’T2+’. The actual values of gate trigger current. 5). It is equivalent to two thyristors in anti-parallel with a common gate electrode. VRRM: the repetitive peak reverse voltage. Quadrant Polarity of T2 wrt T1 Gate polarity 1 2 3 4 (1+) (1-) (3-) (3+) T2+ T2+ T2T2G+ GGG+ Gate Gate n p n n p n Table 1. these alternative methods of switching to the conducting state should be avoided by suitable circuit design. neglecting transients. Figure 3 shows the triac circuit symbol and a simplified cross-section of the device. In practice. It is therefore an a. some occurring regularly and others only occasionally (Fig. VRWM: the peak working reverse voltage. 3 Triac circuit symbol and basic structure Forward current The voltage of the a. As with the thyristor. MT1 MT1 The on state voltage/current characteristic of a triac resembles that of a thyristor. When terminal 2 is negative with respect to terminal 1 (T2-) the blocking and conducting characteristics are similar to those in the T2+ condition. however. provided that the current through the device exceeds the latching current IL before the trigger pulse is removed. however. the triac can be triggered below V(BO) by a gate pulse. there is a variety of transients. and is quoted with the maximum duration of transient that can be handled (usually t < 10ms). The gate is similarly connected. applications. 4 shows that the triac is a bidirectional switch. The triac. However. The following reverse off-state voltage ratings are given in our published data: VRSM: the non-repetitive peak reverse voltage. The main terminals are connected to both p and n regions since current can be conducted in both directions. thyristors must still handle anode to cathode voltages in excess of the nominal mains value. Operating quadrants for triacs Device data Anode to cathode voltage ratings MT2 MT2 Fig. If the triac is not triggered. power control device.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors of anode voltage dVD/dt. The triac static characteristic of Fig.c.c. but the polarities are reversed. This is the maximum continuous peak voltage rating in the reverse direction. the small leakage current increases as the voltage increases until the breakover voltage V(BO) is reached and the triac then turns on. has holding current values below which conduction cannot be maintained. However. Triacs The triac. This is the allowable peak value of transients occurring every cycle. since a triac can be triggered by both negative and positive pulses. holding current and latching current may be slightly different in the different operating quadrants of the triac due to the internal structure of the device. 4 Triac static characteristic 510 . T2+ Reverse voltage V(BO) Blocking IL IH IH IL Forward voltage V(BO) Blocking T2Reverse current Fig. it only requires one heatsink compared to the two heatsinks required for the anti-parallel thyristor configuration. The triac can be triggered in both directions by either negative (G-) or positive (G+) pulses on the gate. Although some transients may be removed by filters. It corresponds to the peak negative value (often with a safety factor) of the sinusoidal supply voltage.

The surge rating also depends on the conditions under which it occurs. However. the IT(AV) rating should be quoted for a particular mounting-base temperature Tmb. The duration of the inrush transient and the mounting base temperature prior to operation determine the maximum allowable rms inrush current. This rating is the peak current that can be drawn each cycle providing that the average and rms current ratings are not exceeded. ITRM: the repetitive peak forward current. VDSM: the non-repetitive peak off-state voltage applied in the forward direction.3ms should be approximately downrated (multiplied by 0. VRRM=650V and the final R (for Reverse) indicates that the anode of the device is connected to the metal tab.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors V V V V D DSM DRM DWM mounting-base temperature is unrealistically low. An unrealistically high ITSM rating could be quoted if. Our published data quotes the ITSM rating for t=10ms. some manufacturers quote ITSM for t=8. Both the repetitive and non-repetitive voltage ratings are determined partly by the voltage limit that prevents the thyristor being driven into forward or reverse breakdown.83) before comparing them with t=10ms surge ratings. described by the waveforms shown in Fig. This figure forms part of the device type number. and thus surge ratings for devices quoted at t=8. For such conditions the rms current rather than the average current may be the limiting rating. VDWM: the peak working off-state voltage applied in the forward direction. our devices are generally characterised at a mounting-base temperature of at least 85˚C. I I Time V RWM V RRM V RSM V R Fig. 6 Diagrammatic current waveform showing device anode current ratings 511 . It is important for applications when the device current waveform is described by a high value form factor. Our data sheets quote ITSM rating under the worst probable conditions. ratings with no associated mounting-base temperature should be regarded with suspicion.5 times the peak value of the sinusoidal supply voltage. IT(AV): the average value of the idealised mains current waveform taken over one cycle. the duration of a half-cycle of 50Hz mains. followed by reapplied VRWM(max) immediately after the surge. assuming conduction over 180˚. 5 Diagrammatic voltage waveform showing device anode voltage ratings The forward off-state voltages corresponding to VRSM. Tj=Tj(max) immediately prior to the surge. and depends on the duration of the surge. ITSM: the non-repetitive (surge) peak forward current. where 650 corresponds to VDRM. T Anode-to-cathode current ratings The following current ratings. are given in our published data. that is. for example. VRRM and VRWM are listed below.3ms (half-cycle of 60Hz mains). it is advisable to choose a device whose repetitive peak voltage ratings VRRM and VDRM are at least 1. IT(RMS): the rms on-state current. Tj<Tj(max) prior to the surge and then the full rated voltage is not reapplied. This rating is the peak permitted value of non-repetitive transients. A device can have an artificially high current rating if the I TSM TRM I I Mains waveform T(RMS) T(AV) Time Fig. 6. and partly by the instantaneous energy (resulting from an increase in leakage current) that can be dissipated in the device without exceeding the rated junction temperature. When a thyristor is to operate directly from the mains supply. This rating gives the maximum rms current that the thyristor can handle. for example BT151-650R. Published data also includes curves for ITSM against time which show the maximum allowable rms current which can occur during inrush or start-up conditions. VDRM: the repetitive peak off-state voltage applied in the forward direction. For devices mounted on heatsinks. Note that the suffix T implies that the thyristor is in the on state.

3 Duration (s) 1 3 10 Temperature ratings Two temperature ratings are given in the published data. even when excessive power is being dissipated in the device. This rating is required for the selection of fuses to protect the thyristor against excessive currents caused by fault conditions. The gate-to-cathode power ratings should not be exceeded if over-heating of the gate-cathode junction is to be avoided. An excessive rate of rise of current causes local heating and thus damage to the device. = 4702 × 3. Under pulse conditions. averaged over a 20ms period.10−3 = 662. 3. Fig. It is normally only valid over the range 3 to 10ms. Tstg: the storage temperature. The I2t at the rms working voltage must be less than that of the thyristor taken over the fuse operating time. in which case: 2 I 2t = ⌠ i 2. Zth(j-mb): the transient thermal impedance between the junction and mounting-base of the device. 0. The fuse must have an rms current rating equal to. that of the thyristor it is to protect.dt ⌡ (1)  ITSM  2  × 10. For example.1 0. Rth(j-mb): the thermal resistance between the junction and mounting base of the device.001 1. PG(AV): the mean gate power.03 0. ITS(RMS) at 3ms is 470A and therefore I2t at 3ms is given by: 2 I 2t (3ms) = ITS(RMS) × t Thermal characteristics The following thermal resistances and impedances are given in our data. The published data also contains graphs of Zth(j-mb) against time (for non-repetitive conditions) such as those shown in Fig. Values of I2t other than those quoted for 10ms can be estimated by referring to the appropriate published curves of non-repetitive surge current against time. The arc voltage of the fuse must be less than the VRSM rating of the thyristor.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors dI/dt: the rate of rise of on-state current permissible after triggering. From Fig. 8.10−3 (A 2s) =  √2   The user should match the minimum I2t capability of the thyristor to the worst case I2t let-through of a range of nominally rated fuses in order to select a fuse that will protect the device under worst probable conditions. thermal impedances rather than thermal resistances should be considered. In our published data. 7 is the non repetitive surge current curve for a thyristor whose I2t at 10ms is 800A2s. Rth(j-a): the thermal resistance between the junction of the device and ambient (assumed to be the surrounding air). This is one of the principal semiconductor ratings since it limits the maximum power that a device can handle. and thus the critical junction temperature will not be reached instantaneously. and can be limited by additional series inductance in the circuit. or less than.003 0. The value given in the published data is for non-repetitive conditions and a particular pulse duration. 7. The junction temperature rating quoted in our published data is the highest value of junction temperature at which the device may be continuously operated to ensure a long life.7A 2s To summarise. 512 . a value is quoted for 10ms. PGM: the peak gate power dissipation. Rth(mb-h): the thermal resistance between the mounting base of the device and the heatsink (contact thermal resistance). when selecting an appropriate fuse the following conditions must be taken into account. Fig. 7 Non-repetitive surge current as a function of time I t: a dimensional convenience specifying the capability of a thyristor to absorb energy. 2.01 0. Both maximum and minimum values of the temperature at which a device can be stored are given. Higher peak power dissipation is permitted under pulse conditions since the materials in a thyristor have a definite thermal capacity. Tj: the junction temperature. The rate of rise of current is determined by both the supply and load impedances. Gate-to-cathode ratings The following gate-to-cathode ratings are given in the published data. ITS(RMS) 600 500 400 300 200 100 0 0. VRGM: the gate peak reverse voltage.

These characteristics relate the total power P dissipated in the thyristor. a.001 0. 9 Heat flow paths Rth(mb-a) = Rth(mb-h) + Rth(h-a) (2) Where appropriate. The heat generated in a semiconductor chip flows by various paths to the surroundings. the average forward current IT(AV).2 1. our published data contains power graphs such as that in Fig.0001 0.01 0.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Z th(j-mb) (K/W) 10 3 1 0. with the form factor. 50 75 th(mb-a) =0.3 0.1 1 10 T j R th(j-mb) T mb R th(mb-h) T h R th(h-a) R’ th(mb-a) Fig. 8 Thermal impedance between the junction and mounting-base as a function of time The values of the various thermal resistances between the thyristor junction and the surroundings must be considered to ensure that the junction temperature rating is not exceeded.8 40 35 30 25 20 15 10 10 5 0 0 5 10 15 20 0 20 40 60 a=4 2.6 2K/W 1K/W Tmb ( o C) 100 105 110 115 120 80 100 120 125 6 IT(AV) (A) Ta ( o C) Fig. the ambient temperature Ta. When a heatsink is used.001 1E-05 0. 9 shows the various thermal resistances to be taken into account in this process. the heat loss direct to the surroundings from the mounting-base is negligible owing to the relatively high value of Rth(mb-a) and thus: Ta Fig. 10.03 0. and the thermal resistance Rth(mb-a). 10 Derivation of the appropriate Rth(mb-a) for a given value of IT(AV).003 0.5K/W R 80 85 90 3 95 4 P(W) 45 a=2. They enable the designer to work out the required mounting arrangement from the conditions under which the thyristor is to be operated.01 Time (s) 0. With no heatsink. the thermal resistance from the mounting-base to the surroundings is given by Rth(mb-a). as a parameter. Fig.9 1.1 0. a and Tamb 513 .

This parameter is given for the worst probable conditions.2˚C/W and then Equation 2 gives Rth(h − a ) = 4 − 0. This parameter is quoted at a particular value of junction temperature. 4. VT: the forward voltage when the thyristor is conducting. 11). and thus contact will take place on several points only.0 25 C 150 C IT(RMS) a= IT(AV) 2. the reverse voltage VR=VRWM(max) and a high Tj. move vertically upwards to intersect the appropriate form factor curve (interpolating if necessary). The measurement must be performed under pulse conditions to maintain the low junction temperature. 6. when the procedure below should be followed. with a small air-gap over the rest of the contact area. Moving horizontally across from this intersection to the appropriate value of ambient temperature gives the required mounting base to ambient thermal resistance Rth(mb-a). the forward voltage VD=VDWM(max) and a high Tj. IL: the latching current (Fig.8 °C/W Mounting torque Two values of mounting torque are given in the published data. characteristics are a= 19. The maximum anticipated ambient temperature is 25˚C.8 1. at this power and ambient temperature of 25˚C.4 0. the characteristics are designed for use in 50Hz sinusoidal applications.6 12 Figure 10 gives the power as P=20W and the mounting-base temperature as Tmb=105˚C. The published data also contains curves of forward current against forward voltage. forward voltage The surface of a device case and heatsink cannot be perfectly flat.2 = 3. 1. 2). Anode-to-cathode characteristics The following anode-to-cathode included in the published data. 2). This parameter is measured at particular values of forward current and junction temperature. Determine the values of IT(AV) and IT(RMS) for the relevant application. Now. Starting from the appropriate value of IT(AV) on a curve such as Fig. IR: the reverse current. Equation 3 gives. Example The thyristor to which Fig. Also. The required heatsink thermal resistance Rth(h-a) can now be calculated from Equation 2 since the mounting base to heatsink thermal resistance Rth(mb-h) is given in the published data.2 0. 10 applies is operated at an average forward current IT(AV) of 12A and an rms forward current IT(RMS) of 19. that is. that is. 10 gives the value of Rth(mb-a) to be 4˚C/W. 2.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Usually.0 3.2A. ID: the off-state current. 11 Forward current vs. This parameter is quoted at a particular value of junction temperature. for example) since this is the worst case. IH: the holding current (Fig. The junction temperature is usually low (Tj=25˚C. which is given by: IF(A) 3. The use of a soft substance to fill this gap will lower the contact thermal resistance. The published data gives the value of Rth(mb-h) (using a heatsink compound) to be 0.2 = 1. usually for two values of the junction temperature: 25˚C and Tj(max) (Fig.0 (3) MAX TYP 1. This intersection gives the power dissipated in the thyristor on the left-hand axis of the combined graph and the mounting base temperature on the right hand axis.0 VF(V) Fig. 10. 514 . We recommend the use of proprietary heatsinking compounds which consist of a silicone grease loaded with an electrically insulating and good thermal conducting powder such as alumina. Determine the form factor. Fig. A minimum value is quoted below which the contact thermal resistance rises owing to poor contact. and a maximum value is given above which the contact thermal resistance again rises owing to deformation of the tab or cracking of the crystal. 5.6 0. This parameter is again given for the worst probable conditions. 0 0 0.

12 Definition of rate of rise of off-state voltage dVD/dt dV/dt: the rate of rise of off-state voltage that will not trigger any device. Thus curves such as those shown in Fig.63 × 2/3VDRM(max) T 0.63VDM = dt T = = 0. T 2T 3T 4T Time Fig. Tj Fig. 13b) which shows how dVD/dt increases as the ratio VDM/VDRM max decreases. Thus the published data also contains curves such as Fig. Fig.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Fig. This parameter is given at maximum values of junction temperature Tj(max) and forward voltage VD=VDRM(max).000 1.63V DM where T is the exponential time constant. 13a) are provided in the published data so that designers can uprate devices operated at lower junction temperatures. The dVD/dt characteristic can also be increased by operating the device at a low supply voltage. This facilitates the design of RC snubber circuits for device protection when required.750 dt 1.500 Exp l dVD 1.200 dt 1. 13 Derating of maximum rate of rise of off-state voltage 515 . The final voltage applied to the device VDM is chosen as VDRM(max) and the junction temperature is Tj=Tj(max).42VDRM(max) T (V/µs) 0. 12 illustrates the definition of dVD/dt.000 600 750 500 250 0 20 Rating point 400 Rating point 200 40 60 T ( o C) j 80 100 120 140 0 0 20 40 60 80 100 VD /V (%) DRM(max) b) Applied voltage. VD a) Junction temperature.400 Exp l dVD 1.000 1. Note that VDM is unlikely to be greater than 2 /3VDRM(max) (usually owing to the restriction of VDWM(max)) and therefore the fact that dVD/dt approaches zero as VDM increases above the value of 2/3VDRM(max) does not cause problems. 12 shows that dVD/dt is given by the expression: V DM dVD dt dVD 0.250 800 1. 2. The values of dVD/dt quoted in our published data are normally specified assuming an exponential waveform. The dVD/dt capability of a thyristor increases as the junction temperature decreases.

14 Gate characteristics vs. A gate drive circuit must be designed which is capable of supplying at least the required minimum voltage and current without exceeding the maximum power rating of the gate junction. IGT: the gate-to-cathode current that will trigger all devices. Example A thyristor has the VGT/Tj and IGT/Tj characteristics shown in Fig.0W. 15. The following design procedure is recommended to construct a gate drive circuit load-line on the power curves shown in Fig. and then determine the minimum values of VGT and IGT from curves such as Figs.5V. plot a second point on the power curve whose co-ordinates are given by VGT(min) and 5×IGT(min). 14a) and 14b) in the published data. 1. Construct a load line between these two points. Note that it is assumed that at switch-on Tj=Ta. An illustration of how the above design procedure operates to give an acceptable gate drive circuit is presented in the following example. or 2. Determine the maximum average gate power dissipation PG(AV) from the published data (normally 0. Determine the minimum open-circuit voltage of the trigger pulse drive circuit: this is the first co-ordinate on the load line at IG=0. Determine its suitability for this device. 15. The load line must also not intersect the curve which represents the maximum average gate power PG(AV) modified by the pulse mark-space ratio. 2.0W or 2.0W) and then use the appropriate choice of x-axis scaling in Fig. Check the power dissipation by ensuring that the load line must not intersect the curve for the maximum peak gate power PGM(max) which is the outermost (δ=0.5W and PGM(max)=5W.25.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors 3 VGT (V) 150 IGT (mA) 2 100 1 50 0 -50 0 50 Tj ( C) 100 150 0 -50 0 50 Tj ( C) 100 150 a) Minimum VGT that will trigger all devices b) Minimum IGT that will trigger all devices Fig. junction temperature Gate-to-cathode characteristics The following gate-to-cathode characteristics are given in the published data. the δ=0.5W. where: PGM(max) = PAV δ (5) For instance. 5. VGT: the gate-to-cathode voltage that will trigger all devices. The slope of this load gives the maximum allowable source resistance for the drive circuit.1) curve of Fig. 15. 516 . 1. 3. 15. Curves such as those shown in Fig. 14 (which relate the minimum values of VGT and IGT for safe triggering to the junction temperature) are provided in data. Using the appropriate horizontal scaling for the device (PG(AV)=0.5W. Estimate the minimum ambient temperature at which the device will operate.0W). in Fig. 14 and is rated with PG(AV)=0.25 curve can be used for a gate drive with a 1:3 mark-space ratio giving an allowable maximum gate power dissipation of PGM(max)=4W. for a thyristor with PG(AV)=1W. 1. This characteristic should be quoted for particular values of applied voltage VD and low junction temperature. 4. IGT(max)=620mA and Ta(min)=-10˚C is to be designed. VGT(min)=4. A suitable trigger circuit operating with δmax=0. This characteristic should be quoted for the same conditions given above.

power curves 1.25. is 0. tr) and the circuit-commutated turn-off time.5W) (P 2. 4.2 0.5 15.25 curve.75V and 330mA. -Junction temperature. V (V) G 22. -Gate trigger current.8 1.167 = 0. high values reduce tgt.9 1.2 0. the load line ABC does not intersect the δ=0.111 = 0.6 1. high temperatures reduce tgt.8 3.250 = 0. 14. At minimum supply voltage.2 2. Note that point C is the maximum current required at IG=570mA and is within the capability of the drive circuit. The sum of the delay time and the rise time is known as the gate-controlled turn-on time tgt.0W) Gate current.0 (P G(AV) =0.0 17.8 = 0. Gate-controlled turn-on time.2 0.0 (P G(AV) =2.1).4 0. -Rate of rise of gate current.0 12. Therefore PGM(max) = PG(AV)/δ = 0.000 A B 0.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Gate voltage. δ.25 = 2W.5 5. pulse and the onset of the anode current which is known as the delay time td (Fig.5 0 0 0 0 0.125 = 0. From Fig.5W).200 = 0. 15 Gate circuit design procedure . and a rise time.5/0. and load line ABC is constructed as shown.2 0. Switching characteristics Two important switching characteristics are usually included in our published data. td. -On-state current. They are the gate-controlled turn-on time tgt (divided into a turn-on delay time.6 3. high gate currents reduce tgt.143 = 0.6 1.4 0.5 20.0 2. Select the top x-axis scale of Fig. 16). giving point ’A’ in Fig.1 0. The time taken for the anode voltage to fall from 90% to 10% of its initial value is known as the rise time tr.0 2. Circuit-commutated turn-off time When a thyristor has been conducting and is reverse-biased. The gate controlled turn-on time depends on the conditions under which it is measured.7 1. 17). tq. 2.8 0.3 0. that is at 1.6 0. 15.500 = 1.5 1. and IGT(min)=66mA. There is a period which elapses between the application of the trigger 517 . The time from the instant that the anode current passes through zero to the instant that the thyristor is capable of blocking reapplied off-state voltage is the circuit-commutated turn-off time tq (Fig. Point B is plotted at the co-ordinates VGT(min) and 5xIGT(min).0 7. it does not immediately go into the forward blocking state: minority charge carriers have to be cleared away by recombination and diffusion processes before the device can block reapplied off-state voltage.75V.0W) 4. 15 (PG(AV)=0. VGT(min)=1.0 G C 0. The gate drive duty cycle.6 1. I Fig.100 = 0. usually VD=VDWM(max).5 10.8 1. As required the load line does not intersect the PG(max) (δ=0.5V.4 2. 3. -Off-state voltage. and thus the following conditions should be specified in the published data.4 0.333 = 0. As required. the open-circuit gate voltage is 4. tgt Anode current does not commence flowing in the thyristor at the instant that the gate current is applied.0 G(AV) =1.4 (A) 0.

19 Triac commutation waveforms (inductive load) 518 . An excessive rate of fall of current creates a large number of residual charge carriers which are then available to initiate turn-on when the voltage across the triac rises. To guarantee reduction of the current below its holding value. -Gate bias. The triac is thus easily able to block the rising reapplied voltage dVcom/dt. the recovery current in the device would simply switch it on in the opposite direction. negative voltages decrease tq. -Reverse voltage. -Rate of rise of reapplied off-state voltage. I T Fig. 18 Triac commutation waveforms (resistive load) Triac ratings The ratings and characteristics of the triac are similar to those of the thyristor. -Junction temperature. one characteristic requires special attention when choosing triacs. commutation does not present any problems when the load is purely resistive. As shown in Fig. -On-state current. the rate of re-applied voltage that the triac will withstand without uncontrolled turn-on. 16 Thyristor gate-controlled turn-on characteristics I T dI T dt dI/dt = 2πf. given by Equation 6.√2 IT(RMS)  dVcom /dt = 2πf. given by equation 7.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors I GT 10% VD 90% 10% td t gt tr to allow the recombination of any stored charge in the device. However. Voltage across triac VT Trigger pulses Time Fig. the rate of fall of current during the commutation interval (turn-off period) and the rate of rise of re-applied voltage after commutation must both be restricted. 18 the rate of fall of on-state current -dI/dt. since the current and voltage are in phase. high rates increase tq. high temperatures increase tq. -Rate of fall of anode current. high currents increase tq. low voltages increase tq. and the rate of rise of commutating voltage dVcom/dt.√2 V(RMS)  (6) I R (7) V D dV D dt Supply voltage Load current Current V R tq Time -dI/dt dV /dt com VDWM Time Fig. the supply voltage must be reduced to zero and held there for a sufficient time Supply voltage Load current Current Time dV /dt com VDWM Time -dI/dt Voltage across triac Trigger pulses Time Fig. high rates increase tq. are sufficiently low to allow the stored charge in the device to fully recombine. To ensure turn-off. except that the triac does not have any reverse voltage ratings (a reverse voltage in one quadrant is the forward voltage in the opposite quadrant). If a triac is turned off by simply rapidly reversing the supply voltage. With supply frequencies up to around 400Hz and a sinusoidal waveform. 17 Thyristor turn-off characteristics The following conditions should be given when tq is quoted.

the circuit design can remain simple if Hi-Com triacs are employed instead. The characteristic dVcom/dt is specified under the worst probable conditions. The maximum rate of rise of commutating voltage which will not cause the device to trigger spuriously is an essential part of the triac published data.1 and 6. Sensitive gate triacs (i. BT137-600F. IT(RMS) = IT(RMS)(max).2 explain the advantages of using Hi-Com triacs in such inductive circuits.8 1.4 1 0 20 40 60 Tj (C) 80 100 120 Fig. dVcom/dt will be very large. In order that designers may economise their circuits as far as possible. namely: -mounting base temperature.1 3. The triac may switch on immediately unless dV/dt is held less than that quoted in the published data by suitable circuit design. Our published data also contains graphs such as 519 . Tmb=Tmb(max) -reapplied off-state voltage. those which require only a small amount of gate current to trigger the device) have less ability to withstand high values of dVcom/dt before sufficient current flows within the device to initiate turn-on. 20 which relate dVcom/dt to junction temperature with -dIT/dt as a parameter. These different device selections are differentiated by suffices which are added to the device type number eg. particularly the rate of fall of on-state current -dIT/dt. Sections 6. However. VD=VDWM(max) -rms current.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors dV/dt (V/us) 1000 dVD/dt limit 100 -dI/dt= 10 5.0 2. The dV/dt capability can be traded off against the gate sensitivity (IGT(max)) of the device.3. Detailed design considerations for dVcom/dt limiting in inductive circuits when using triacs are considered in separate articles in this handbook. we offer device selections with the same current ratings but with different values of dVcom/dt (at the same value of -dIT/dt) for some of our triac families.3. When the on-state current has fallen to zero after a triac has been conducting in one direction the supply voltage in the opposite direction will have already reached a significant value.9 3. Fig.3 1. with an inductive load (Fig. 20 Rate of rise of commutating voltage with rate of fall of on-state current and temperature However.e. The rate of fall of triac current will still be given by Equation 6 but the rate of rise of reapplied voltage. 19) the current lags behind the voltage and consequently commutation can present special difficulties. dVcom/dt is meaningless unless the conditions which are applicable are provided. Alternatively.

.

Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Thyristor and Triac Applications 521 .

.

Upon turn-off. Control of transformers supplying an inductively loaded bridge rectifier is particularly troublesome because of the added effect of rapid current decay during commutation. The transformer secondary is thus shorted for some time after the zero transitions of the mains voltage and a reverse voltage is applied to the triac. IRRM is abruptly transferred to the snubber elements R and C so the voltage abruptly rises to the level R. Unlike thyristors there is no circuit-imposed turn-off time. During the commutation interval a high rate of decay of current (dIcom/dt) results for two reasons. turning it off. Owing to the high value of both dIcom/dt and dv/dt. 2). To ensure commutation the decay rate of current before turn-off and the rate of rise of reapplied voltage must both be held below specified limits. the commutation behaviour is summarised here. Secondly. The large value of dIcom/dt results in a high rate of rise of voltage. dv/dt. The load inductance forces the rectifier diodes into conduction whenever the instantaneous dc output voltage drops to zero. Saturation should occur at a fraction of the rated load current so that the loss in the rectifier output voltage is minimised. Firstly the rate of fall of current is high because the leakage inductance of most transformers is low. with an inductive rectifier load a substantial current flows when commutation starts to occur. 1). An excessive current decay rate has a profound effect on the maximum rate of rise of voltage that can be sustained. Since the current decays rapidly the peak reverse recovery current IRRM is fairly large. This is necessary to achieve a small dc output voltage loss (represented by the shaded areas in the voltage waveform of Fig. Triacs are bipolar power control elements that may turn on with either polarity of voltage applied between their main terminals. At low currents the total inductance is large.2. Because of transformer leakage inductance the triac does not turn off immediately but continues to conduct over what is called the commutation interval (see Fig.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors 6. Commutation failure is likely to occur owing to circuit inductance imposing a sudden rise of voltage on the triac after conduction. thus softening the commutation and eliminating transients.1 Triac Control of DC Inductive Loads The problem of inductive loads This publication investigates the commutation problem encountered when triacs are used in phase control circuits with inductive loads. 1) in the transformer. Fig. 1 Triac control of transformer suppling rectifier with an inductive load (α = trigger angle) Obtaining reliable commutation A saturable choke in series with the transformer primary proves effective in achieving reliable commutation (Fig. loss of control follows unless measures are taken to prevent it. For a better understanding of the nature of the problem. The choke delays the rise in voltage so a quiescent period of a few tens of microseconds is introduced. during which time 523 . as then a large amount of stored charge is available to initiate the turn-on in the next half cycle. Figure 1 shows the condition for a triac controlled transformer followed by a rectifier with inductive load.IRRM (C is initially discharged).

3 Equivalent circuit diagram Fig. 2 Use of a saturable choke to ensure commutation Circuit analysis Over the commutation interval the transformer secondary is shorted as the load inductance keeps the rectifier diodes in conduction.d i/dt = −Vωt where di/dt is the rate of change of triac current. giving: Fig. The mains voltage is given by vi= Vsinωt. so the simplified diagram of Fig. There is usually no difficulty in designing a choke such that the decay rate of current (dIcom/dt) and the rate of rise of voltage (dv/dt) are sufficiently reduced to ensure reliable control. Period t1 to t2 is shown expanded) Integrating equation (2) gives: i t = It − ˆ vi = −Vωt (1) ˆ Vωt 2 2(Lleak + Lsat ) (3) where It is the current prior to commutation. If the load time constant is much larger than the mains period then the load current can be assumed to be purely dc. 3 applies. Assuming for this analysis that Ls remains in saturation (dashed portion in it waveform) then if Lsat is the saturated inductance. 4. 524 . The waveforms of triac voltage and current are given in Fig.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors the triac can recover. the following expression can be derived: ˆ (Lleak + Lsat ). As the commutation interval is a fraction of the ac period then the rate of change of voltage during the commutation interval can be assumed to be linear. Over the period 0 to t2 the voltage across the saturable choke Ls and leakage inductance Lleak is equal to vi (assuming the triac on-state voltage to be negligible). (2) Fig. 4 Commutation voltage and current waveforms (Dashed portion of It shows current waveform if Ls remains saturated.

Since the motor has a fairly high inductance it may be considered as a constant current source. for a continuous dc load current is: diL i dv = R. vi.DC motor load The motor control circuit of Fig. (7) Choke Lunsat=2. diL/dt. In that case Equations (7) and (9) are still valid by omitting Lleak. so. dvcom/dt. 5 illustrates the use of the design method proposed in the previous section. 525 . 30 turns on 36x23x10mm3 toriod core 6kVA. inductance Motor Series wound DC inductance = 30mH With Lleak = 0.3A/ms and dv0/dt = 0.9mH motor. the triac has fully recovered at time t2.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors At time t1. At time t2 the triac turns off but the voltage across it is still zero. This parameter is decisive for the behaviour of the triac. Using equations (7) and (9) gives dic/dt = -18. the current decay rate at zero current is given by: V2 dic = dt (Lleak + Lunsat ) ˆ sat  2ωVIt (Lleak + L √ ) =− Lleak + Lunsat Fig. the mains voltage V2 at the instant of turn-off is very nearly equal to V1. At time t2. ton is the turn-on time of the triac and R is the snubber resistance. current it passes through zero. Lunsat = 2.6V/µs. dv/dt is equal to the initial rate of rise of voltage dv0/dt. 50Hz supply. leakage Leakage The initial rate of rise of off-state voltage. At t1 the mains voltage has attained the value V1 which is found by combining equations (1) and (4) to give: Example . The voltage drop across Ls and Lleak is equal to V2 and the rate of rise of current carried by these inductances. giving a severe test condition for triac commutation. is given in equation (7). can now be derived.25mH. that is. Transformer 220V/150V. when the off-state voltage has reached a substantial value. some series inductance is still needed to restrict turn-on di/dt. These values can be compared with the commutation limits of the device to ensure that reliable commutation can be expected. from (3): t1 =   √ 2It (Lleak + Lsat ) ˆ Vω (4) In circuits where no transformer is interposed between the triac and rectifier. The inductance in the ac circuit also restricts turn-on di/dt which. Maximum turn-on di/dt occurs at the peak value of input voltage. From equations (7) and (8): d ion vi vi ≈ + dt ton R Lleak + Lunsat (10) dv0 R ˆ = sat  2ωVIt (Lleak + L ) dt Lleak + Lunsat √ (9) where vi is the instantaneous ac input voltage. 0. Since in a practical circuit the delay is only of the order of 50µs. The rate of rise of triac voltage dv/dt is determined by diL/dt and the values of the snubber components R and C.9mH. The initial rise of on-state current depends on the snubber discharge current through R as well as the limiting effect of the circuit inductance. and so the current i to be taken over by the parallel RC snubber network is zero. ˆ V1 = −√ ) sat  2ωVIt (Lleak + L (5) Choke Ls comes out of saturation at low current levels so the triac turn-off point is delayed to time t2.25mH and Lsat<<Lleak the circuit conditions can be calculated for a triac current of It = 20A and a 220V. since a much greater dv/dt can be sustained after carrier recombination. 5 DC Motor test circuit. Denoting the value of unsaturated inductance as Lunsat. + dt dt C (8) When the interval t1 to t2 is long enough. Thus from equation 5: ˆ V2 ≈ −√ ) sat  2ωVIt (Lleak + L (6) The triac conducts until time t2.

N. 9 Triac voltage and current. Timebase: 100µs/div Upper trace: Triac voltage. Timebase: 2ms/div Upper trace: Triac voltage. 8 Triac voltage and current. 7A motor current. vt (100V/div) Lower trace: Triac current. vt (10V/div) Lower trace: Triac current. vt (20V/div) Lower trace: Triac current. 23A motor current. Series choke added. 10) the quiescent interval is about 30µs. At this current (Fig.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors The oscillograms of Figs. vt (20V/div) Lower trace: Triac current.B. 10 Triac voltage and current. 8 to 10 the choke softens commutation so that dependable control results even at 23A motor current. N. 6 Triac voltage and current. it (1A/div) Fig. slight reverse recovery current Timebase: 50µs/div Upper trace: Triac voltage. No series choke.B. it (1A/div) 526 . Fig.Series choke added 23A motor current. it (5A/div) Fig. As seen from Figs. With no choke added a large dv/dt was observed (Figs. vt (10V/div) Lower trace: Triac current. Timebase: 100µs/div Upper trace: Triac voltage. which is adequate time for the triac to recover. No series choke. 6 to 10 illustrate circuit performance. 7A motor current. 7A motor current. it (5A/div) Fig. 7 Triac voltage and current. 6 and 7) and so consequently commutation failed when motor current was increased to around 9A. it (1A/div) Fig. Series choke added. Snap-off current Timebase: 100µs/div Upper trace: Triac voltage.

The conduction angle. The result of the test is shown graphically in Fig.2 Domestic Power Control with Triacs and Thyristors The increasing demand for more sophisticated domestic products can. the triac will be triggered at the same time after the start of either a positive or negative half cycle. the speed of the motor remains constant and. by using a BT138 triac. 2 Vacuum cleaner . With triacs and thyristors it is possible to produce high performance mains controllers which use only a few simple components. but this is not necessarily the case. The following notes give details of some typical control circuits and highlight areas for special attention when adapting the designs for specific applications. is determined by the adjustment of R2.2. C1 charges until the voltage across it exceeds the diac breakover voltage. Diodes D1 and D2 stabilise the supply voltage to the charging circuit so that its operation is independent of mains voltage fluctuations. During each half cycle of the mains sinewave. a simple but nevertheless effective and reliable suction control circuit (Fig. The diac then switches on and C1 discharges itself into the gate of the triac and switches it on. the cleaner was loaded with a water column. In many cases. Its action is controlled by a diac which is switched on by a charge on C1 under the control of potentiometer R2.) Fig. It might be assumed that the cost of the electronics would be high. 1 the BT138 is the power control element. Circuit Description In Fig. to adjust the suction of a vacuum cleaner. -VBO to +VBO. if suction control is attempted. the brightness of room lighting or the speed of food mixers and electric drills. The measured range of water column height (100 to 1100 mm) translates into a wide air flow range . 5ms/div. Figure 2 shows the current and voltage waveforms for the triac when the conduction angle is 30˚. The zinc oxide voltage dependent resistor (U) minimises the possibility of damage to the triac due to very high voltage transients that may be superimposed on the mains supply voltage. Furthermore its very low thermal impedance minimizes heatsink requirements. In fact. If -VBO and +VBO are equal and opposite. Vacuum cleaner suction control The competitive nature of the vacuum cleaner market has led to the development of a wide variety of machine types and accessories.Triac waveforms Fig. For the test. 1) can be constructed very economically. it consists merely of an adjustable vent in the air flow path. remains within the breakover voltage limits. The width and amplitude of the trigger pulses are kept constant by gate resistor R4. This control can be used. 5ms/div.from little more than a whisper to full suction. and is suitable for all types of cleaner with a power consumption of up to 900W. and therefore the speed of the motor and the cleaner suction. 1 Vacuum cleaner suction control circuit Circuit Performance A laboratory model of the circuit has been tested to determine the range of control that it has over the suction power of a typical vacuum cleaner. This is a glass passivated triac which can withstand high voltage bidirectional transients and has a very high thermal cycling performance. in part.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors 6. for example.) b) Triac Current (1 A/div. The heart of the circuit is the BT138. be met by providing the user with some form of electronic power control. The resistance of the diac is virtually infinite as long as the voltage across it 527 . Preset potentiometer R3 is used to set the minimum suction level. Electronic suction control sounds somewhat expensive and unnecessarily complicated for such an elementary application. 3. a) Triac Voltage (100 V/div.

and in particular the triac and its heatsink. mains supply 198 V (220 V .3 ’limit’ current (A) 24 21 19. A reasonable figure would be the worst case steady state value of Tmb during normal running. 1 2 3 4 5 10 20 time (ms) 20 40 60 80 100 200 400 peak current (A) 49 41 35 32 29 20 14 rms current (A) 22 18 13 14 13 9 6. The current decays to its steady state value in about 1 s. 1 The table shows that the speed setting range is wide.5 14 Table 1. Fig. As suction power is a function of the speed of the vacuum cleaner motor. and R3 set so that the motor just ran. The initial test conditions were: unrestricted flow.) cycle number. with the suction blocked and unrestricted). The ratio of Nmax to Nmin is 3. the curve for the BT138 is reproduced in Fig. 5.82:1 under the same conditions. To ensure that the triac does not overheat. prior to starting. 1 has been shown to work well in a typical vacuum cleaner application. This implies that speed stabilisation is unnecessary. for 220 V mains and unrestricted airflow. The table also shows that the difference between the Nmin for minimum and maximum air flow is quite small.5 18. a second test was carried out to determine the range of motor speed control under conditions of minimum and maximum air flow (i.5 18 15. the speed setting ratio is reduced to 1. This test also checked the motor speed variation due to ±10% variation of a nominal 220 V AC mains supply. R2 at maximum resistance. The variation of motor speed due to variation of the mains input is quite small and represents a negligible change of suction. Motor speed figures for circuit of Fig. 4 Starting current (20 A/div. so some aspects of the design should be looked at carefully before it is finalised. 4. In some areas within the cleaner the temperature can be quite high.e. 528 . 3 Suction power as a function of motor speed Starting current Another factor that may lead to thermal problems is that of inrush current.42:1. Table 2.10%). Nmin is the speed at which the motor just runs and Nmax is the speed of the motor with R2 set at minimum resistance. 50 ms/div. mains voltage (V) 198 220 242 blocked air flow Nmin Nmax (rpm) (rpm) 5300 6250 7400 17100 19000 20000 unrestricted air flow Nmin Nmax (rpm) (rpm) 4300 5000 6000 15400 17100 18200 Fig. The circuit. is particularly important. Tmb. Currents during starting The first step in checking for a problem is to estimate the mounting base temperature. But motors and environments do vary. The rms current during the first 20 ms could be 20 A or more. Table 1 shows the results of the test.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Special Design Considerations The circuit shown in Fig. The starting current of a vacuum cleaner motor is typically as shown in Fig. within the case of the cleaner. reference should be made to the inrush current curves in the triac data sheet. If D1 and D2 are omitted from the circuit. should not be placed in one of these areas if the designer is to avoid problems keeping the temperature of the triac below Tjmax. Circuit positioning The siting of the circuit.

Direct measurement indicated that during normal running the Tmb of a BT138 mounted on a particular heatsink. From other measurements it was estimated that the ambient temperature would not exceed 73˚C. 5. is turned off and then started again before there has been any cooling. This will often be the case with vacuum cleaner motors for two reasons: . If a snubber is found to be necessary then a 100 Ω 0. 4. Since the current is not an ideal sine wave these have been calculated from the peak current by assuming a crest factor (peak to rms) of 2. the measures taken to suppress the electrical ’noise’ of the motor combined with the motor itself will often be more than sufficient to overcome the interference generated by the switching of the triac but this must be checked in all applications. care should be taken to ensure that the triac switches correctly even during starting.a reasonable assumption which covers the case where the motor has been running for a long time. Interference It is. Since ’actual’ inrush current is always less than the ’allowed’ current it is safe to use the BT138 under the proposed conditions to control the motor. This was because the values of dI/dt and dV/dt generated by the circuit were well within the capability of the BT138. These figures give a worst case steady state Tmb of 95˚C.these motors introduce only a small phase shift in the current. Commutation The circuit shown in Fig. necessary to check that the overall equipment complies with local regulations for conducted and radiated interference. As an example consider the performance of the BT138 driving a motor whose starting current is shown in Fig. 5. It can be assumed that this is the highest temperature that the mounting base could be.1µF capacitor will be more than adequate in most circumstances. 10 and 20 of the starting current are given in Table 2. this amounts to a dI/dt well within the capability of the BT138. 1 has no RC snubber. It should be noted that because the crest factor is >√2 the dissipation of the BT138 will be less than assumed by the inrush current curves of Fig. In particular. of course. so the voltage step is small and the dV/dt is low. would be no more than 22˚C above ambient. However. The rms values of cycles 1 to 5. prior to starting .the steady state value of the current is much less than the maximum rating of the BT138.5 W resistor in series with a 0. Also shown are the relevant IO(RMS) figures from the 95˚C line of Fig. 5 Inrush current curve for BT138 Step 2 is to calculate the rms value of one cycle of the starting current at several times during start up and step 3 is to compare these figures with the values taken from the appropriate line of the inrush current curve. However care must be taken to ensure that this is true in all applications. .23. 529 .Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Fig.

The simple LC filter shown within the dashed-lined box in Fig. without filter. triggered via the diac. The glass passivated BT138 triac meets these requirements. as is likely. 7 Interference on mains supply Interference Regulations concerning conducted and radiated interference vary considerably form country to country but it is likely that some form of filter will be needed. when 530 . However. once the prerogative of entertainment centres. The first is to keep mains distortion within the allowed limits. The diac then conducts and C2 discharges into the gate of the triac. then air circulation is going to be very restricted and the ambient temperature around the circuit will be quite high. Each half cycle of the mains charges C2 via R1. Curves (a) and (b) show the level of noise on the mains supply for the circuit. The BT138 is the power control element. limits. The resistance of the diac is very high as long as the voltage across it remains within its breakover voltage limits.P. 6 is often all that is needed.S. The voltage across C1 partially restores the voltage across C2 after triggering and thereby minimizes the hysteresis effect. If C1 were not included in the circuit. Fig. If.R.15 µF capacitor and a low Q inductor of 2. This would cause an undesirable hysteresis effect. the load of this circuit must be restricted to a much lower level. the voltage across C2 would change abruptly after triggering and cause the phase relationship between the mains voltage and voltage across C2 to progressively alter.5 µH was found to be sufficient for the circuit to meet the C. This is illustrated by the plots shown in Fig.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Domestic lamp dimmer The use of light dimmers. Fig. Firstly. The VDR minimizes the possibility of the triac being damaged by high voltage transients that may be superimposed on the mains supply voltage. The width and amplitude of the trigger pulses are kept constant by gate resistor R4. It can also withstand high voltage bidirectional transients and its low thermal impedance minimizes heatsink requirements. The setting of potentiometer R2 determines the phase difference between the mains sine wave and the voltage across C2. R2 and R3 until the voltage being applied to the diac reaches one of its breakover levels. the circuit is to be mounted in the wall in place of a conventional switch. This in turn sets the triac triggering angle and the lamp intensity. It is. but a combination of 0. It is important for reliability reasons to ensure that the temperature of the BT138 never exceeds Tjmax.I. -VBO to +VBO. It is necessary to ensure that the component parts of these units are simple and reliable so that they are compatible with the domestic environment. The values of the filter components will vary. 7. 6 Lamp dimmer circuit Circuit Description A simple circuit of a light dimmer using the BT138 is given in Fig. capable of controlling loads with a rating of 2 kW or more. the triac will be triggered at the same time after the start of either a positive or negative half cycle. There are two reasons for this. 6. has now become widespread in the home. therefore. it has a peak non-repetitive on-state current handling capability of up to 90 A which means it can easily withstand the inrush current that occurs when a cold lamp is switched on. switching it on. Special Design Considerations Circuit rating The BT138 has an rms current rating of 12 A. so the dissipation of the triac must be kept to a low level. If -VBO and +VBO are equal and opposite. The second reason is to limit dissipation. without the necessity of expensive filter networks.

and there will be considerable room for reducing its size.P. This decision is made harder by the need to have a good speed regulation under the widely varying loads that these products are subjected to. R2. R3 150 C1 25uF 25V D1 b) Improved Low Speed Controller R1 5k6 6W Back EMF Feedback Circuits A simple motor speed control circuit that employs back EMF to compensate for changes in motor load and mains voltage is shown in Fig. When the thyristor is not conducting the motor produces a back EMF voltage across the armature proportional to residual flux and motor speed.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors controlling 550 W and 25 W loads respectively. 8(a). This appears as a positive potential at the thyristor cathode. R1 5k6 6W D2 R2 1k0 1W R4 500 BT151 Filter inductor Having selected the value of filter inductor.R. A thyristor fires when its gate potential is greater than cathode potential by some fixed amount. D2 BT151 R2 2k0 1W R3 150 C1 50uF 40V R4 500 D1 c) Improved Low and High Speed Controller Fig. This could cause progressive damage to the triac resulting in premature failure. But they are products where costs have to be tightly controlled so the choice of circuit is very important.I. the circuit may function in several modes. limit. However. 8 Thyristor Speed Control Circuit Using Back EMF Feedback 531 . which is curve (c).R3 and diode D1 provide a positive going reference potential to the thyristor gate via diode D2. has been met. the designer has then to decide how to make it. If the inductor does saturate then the filter capacitor will. R3 150 D1 a) Basic Controller R1 5k6 6W D2 R2 2k0 1W R4 500 BT151 Speed control for food mixers and electric drills Food mixers and electric hand drills are products whose useability is improved by the addition of electronic speed control. effectively. They compensate for load variation by adjusting the firing angle when there is a change in the motor speed . be shorted out by the triac.it is not necessary to achieve a high Q . Curves (d) and (e) are for the circuit with filter connected showing that the C.S.as indicated by a change in its back EMF. In this case the triac current could rise faster than the dI/dt rating allows. Construction will not be too critical . care must be taken to ensure that the inductor does not saturate when the inrush current of a cold lamp flows through it. The circuits to be described provide continuous control of motor speed over a wide speed range by adjusting the conduction angle of a BT151 thyristor. Depending on the waveform shape and amplitude at the gate. Diode D1 is used to reduce the dissipation in the resistor chain by some 50% and diode D2 isolates the trigger circuit with the thyristor in the on-state. The resistor chain R1.

the most positive value of the trigger potential. Thyristor firing would then continue intermittently at the beginning of the positive cycles to maintain some average motor speed. 10(a). Referring to Fig. At lower motor speeds the firing angle might need to be 130˚ for smooth operation. 9). Such a waveform can be produced if there is some phase shift in the trigger network. allowing thyristor triggering to be delayed to this point. 9 Waveforms with DC Gate Supply 3 If. If the maximum firing angle is limited to 90˚ then intermittent firing and roughness of motor operation will result. If. As a result it is impossible for the firing angle to be later than 90˚ .Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors v Mains Voltage 3 2 1 v Mains Voltage 3 2 1 v 0 90 180 back EMF 0 90 180 a) Basic Controller v Mains Voltage Fig. 8(a) the waveform appearing at the thyristor gate will approximate to a half sine wave. The waveform developed across the capacitor has a positive slope to some 140˚. 8(b) has been modified by the addition of a capacitor C1 and diode D1. v 0 3 2 1 90 180 c) Improved Low and High Speed Controller Fig. 10 Gate Voltage Waveforms 532 . 2 1 3 v 2 1 0 90 180 b) Improved Low Speed Controller v Mains Voltage 3 2 1 Stable Firing at Small Conduction Angles The trigger network of the circuit shown in Fig. however. Fig. the thyristor would continue to fire at the beginning of each cycle until the back EMF was large enough to prevent firing. for example. The diode clamps the capacitor potential at zero during the negative going half cycles of the mains input. during positive half cycles a constant DC potential was applied at the gate (see Fig. the waveform at the gate has a positive slope value to an angle of at least 130˚ then it will be possible to have a stable firing point at low speeds.

11 Simplified Firing Circuit 0 0 50 100 150 200 250 Improved Motor Performance With Stable Firing Both the circuits so far discussed have gate voltage waveforms that are of near linear slope from the zero point of each positive half cycle. R1 D2 i Gate Voltage 35 30 R2=1500 25 R2 C1 20 v i2 i1 R2=800 15 D1 10 5 R2=200 Fig. at low angles. 8(a). capacitor C1 is discharged less during negative half cycles but its charging waveform remains substantially unchanged. During negative half cycles the only discharge path for capacitor C1 is via resistors R2 and R3. 8(c) is one way of achieving this. . There are two factors that are important in the circuit operation in order to obtain the above requirements. in conjunction with the gate current from the triggering circuit. Diode D2 isolates the triggering circuit when the thyristor is ON. see Figs. it enables resistor R2 to give consistent speed settings.The value of positive slope of the waveform appearing at the thyristor gate. Hence the result of varying R2 is to shift the DC level of the ramp waveform produced across C1. It will. give a marginally greater speed drop for a given motor loading at low speed settings. 12 Calculated Gate Waveforms Circuit Design If the speed controller is to be effective it must have stable thyristor firing angles at all speeds and give the best possible speed regulation with variations of motor load. This means that the only time that the thyristor can be fired early in the mains cycle. is when the back EMF and hence motor speed is very low. This means that the time to reach the minimum trigger voltage is reduced. Phase Angle C1=32uF C1=50uF C1=64uF Fig. This circuit will give smoother and more stable performance than the circuit of Fig. This effect tends to prevent smooth running at high speeds and high loads. Stable triggering. say at 20˚. the peak of the waveform at the gate will move towards 90˚ as shown in Fig. 8(a) approximates to that of Fig. 10(a) and (b). can be achieved if the gate voltage ramp starts each cycle at a small positive level. Resistor R4 adjusts minimum speed. and by effectively bleeding a constant current. As the value of resistor R2 is increased. however. At the maximum speed settings the circuit of Fig. Diode D1 also prevents C1 from being discharged as the thyristor switches off by the inductively generated pulse from the motor. 8(b).The phase angle at which the positive peak gate voltage is reached during a positive half cycle of mains input. 8(c) gives a motor performance that satisfies both of the above requirements. As the speed increases.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors As the slider of R2 is moved towards R1. In this circuit capacitor C1 is charged during positive half cycles via resistor R1 and diode D1. 533 . 10(b). . the no load firing angle will also advance by a similar amount so stability will be maintained. The circuit of Fig. The circuit of Fig.

Fig. 12. exceeds the back EMF by the gate/cathode voltage.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors As previously described the charging of capacitor C1 by resistor R1 determines the rate of rise of voltage at the thyristor gate during the positive half cycle. 13. at high speeds and torques. it is better. random variations of the firing angle by 20˚ are tolerable. 8(a) circuit. when used to drive an electric drill. and a steeper slope of thyristor gate voltage must be used in order to obtain stable motor operation. giving good torque speed characteristics. 12 may be used as a guide to obtaining the value of this resistor. A motor that gives a smooth back EMF voltage will allow a low slope value to be used. Figure 15 compares the circuits of Fig. It should be noted that the circuits of Figs. The value of resistor R2 in Fig. The value of capacitor C1 is chosen to provide the required positive slope of the thyristor gate voltage. 8(c) simplified to the form of Fig. 14. That of Fig. resistor R1 must also have a value such that several times the maximum thyristor gate current passes through the RC network to D1. Therefore. This current will then give consistent speed settings with the spread of thyristor gate currents when the minimum speed is set by resistor R4. where it is assumed that current flowing to the thyristor gate is small compared with the current flowing through resistor R1.the forward gate/cathode voltage of the thyristor. the back EMF of the motor must be known. It must be chosen so that at 70˚ and at its highest value. there were variations in the back EMF of 1 V. The waveforms of Fig. for example. The thyristor is turned ON when a trigger waveform. This may be measured using the arrangement shown in Fig. 8(a) at low speeds but. 13 Back EMF Measurement Arrangement 534 . if the back EMF varies within a cycle then there will be a cycle to cycle variation in the firing angle. Some calculations have been made on the circuit of Fig. to oscilloscope b) measurement circuit back EMF period A diode conductiong period B Performance The torque speed characteristics of the three circuits. The voltage appearing across the motor is measured during the period when the series diode is not conducting (period A). back EMF b) typical waveform. 8(a) and 8(c) when the load is a food mixer motor. shown in Fig. It may be seen that the circuit of Fig. 8(b) has a poorer performance than the two other circuits. 12. 11. In order to adjust the circuit to suit a given motor. 8(c) may be seen to give a similar performance to the circuit of Fig. since the mains input is a sine wave. the gate voltage is higher than the measured back EMF by about 2 V . there is little increase in the ’no load’ speed when the firing angle is reduced to less than about 70˚. R2 and C1 and is given later. Some motors have coarser back EMF waveforms. The voltage so obtained will be the motor back EMF at its top speed on half wave operation. In practice. the variation of firing angle would be about 12˚. With capacitor values of 50 µF and 64 µF the firing angles variations would be 19˚ and 25˚ respectively. with voltage undulations and spikes. 8(b) and (c) provide low speed operation free from the intermittent firing and noise of the Fig. So. are compared in Fig. Component values have been substituted into the expression to give the thyristor gate waveforms shown in Fig. and corresponds to the back EMF that would be obtained from the unloaded motor at its highest speed when thyristor controlled. a capacitor value of 50 µF would be suitable. The positive slope value of the thyristor gate voltage will have to be fixed according to the motor used. 8(c) determines the motor ’no load’ speed setting. However. If. An expression has been derived for the voltage that would appear at the anode of D2 in terms of R1. then with a firing angle of 70˚and a capacitor of 32 µF. Normally.

000 C1 6.000 the gate voltage waveforms shown in Fig. 11.8(b) Fig. f(E) − v dv v = C1 + R1 dt R2 0 0 0. 6.8(c) Fig.8(a) Fig.5 2 2.000 A 0 0 0. n 2 − 1   neglecting terms of the Fourier series with n > 2..8(a) diode circuit circuit 10. 535 . i1. Y are constants.000 (1) simplifies to 2. i2 are instantaneous currents.000 Fourier analysis of a half sinewave gives:series Fig.5 Motor Torque (Nm) Fig.5 0. d are constants. then 8.500 1. i= f(E) − v = i1 + i2 R1 500 therfore.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Motor Speed (rpm) 3. Simplifying:C1  1 1  f(E) dv + v +  = dt  R1 R2  R1 Motor Speed (rpm) 12.3 0. Put v = a sin(ωt) + b cos(ωt) + c sin(2ωt) + d cos(2ωt) + D B (3) Motor Torque (Nm) Fig. b. 14 Performance with Hand Drill Load where i. is given by: i1 = dq dv = C1 dt dt 2.4 0. 12.5 1 1.000 and i2 = v R2 1.000 1 1 2 n = 0 cos(nθ)  ∑ f(E) = E  + sin(θ) −  π 2 π n = 2.500 2. D. Series Fig.2 0. B. 15 Performance with Food Mixer Load Circuit Calculations The following analysis derives an expression for voltage ’v’ at the anode of D2. c.000 Representing a mains half sine wave by f(E) where E is the peak mains voltage.000   1 1 E 1 1 2 dv + v  +  =  + sin(ωt) − cos(2ωt) dt 3π  R1 R2  R1  π 2  then C1   1 1 E E 1 2 dv + v +  − =  sin(ωt) − cos(2ωt) dt 3π  R1 R2  R1π R1  2  4.1 0. 4. The analysis assumes that the current drawn by the thyristor gate is negligible in comparison with the current flowing in R1.6 dv + Bv − D = X sin(ωt) − Y cos(2ωt) dt (2) where A.. This expression can be used to produce where a. X.8(c) diode circuit circuit circuit The charging current i1 for capacitor C1 in Fig.

(R1 + R2) sin(ωt) − ωC1R1R2 cos(ωt) 2ω C R R + (R1 + R2) 2 2 1 2 1 2 2 2 Conclusions The addition of electronic control can enhance the overall useability of many domestic products. sin(2ωt). 12. Although these circuits are not complex and use only simple components. sin(ωt). it is still important to design with care to ensure that the best performance is achieved. sin(2ωt) 4A 2ω2 + B 2 BY D . cos(2ωt). 1 π(R1 + R2) 4ωC1R1R2 sin(2ωt) + 4(R1 + R2) cos(2ωt) 2 2 2 3π[4ω2C1 R1 R2 + (R1 + R2)2] This may be simplified since (R1 + R2)2 2 2 2ω2C 2R1 R2 So the voltage that the trigger circuit would apply to the gate (assuming the gate draws no current) is given by: 536 . to construct a range of phase control circuits which can meet many of these cost and operational requirements. substituting for the constants in equation (2) gives: v = R2E. cos(2ωt) + B 4A 2ω2 + B 2 + {(R1 + R2)sin(ωt) − ωC1R1R2 cos(ωt) − − 2 ωC R R sin(2ωt) 3π 1 1 2 1 (R + R2)cos(2ωt)} 3π 1 Solving this equation for a different values of C1 and positions of R2 gives the curves shown in Fig. sin(ωt) − 2 2 . This report has given examples of some of these circuits and has highlighted the areas of their design requiring particular care. Cost and performance requirements are major factors when determining the type of control circuit to be used in these applications.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors dv = aω cos(ωt) − bω sin(ωt) dt + 2cω cos(2ωt) − 2dω sin(2ωt) (4) v= R2E π(R1 + R2) R2E 2 2 2 2ω2C1 R1 R2 substituting (3) and (4) in equation (2) and equating terms in cos(ωt). It is possible. cos(ωt) A 2ω2 + B 2 A ω + B2 − − 2AωY . then BX AωX v= . using thyristors and triacs. − R2E. + R2E.

2 0.0 31. Time-proportional control regulates the load power such that there will be no overshoot or undershoot of the desired temperature as is the case with normal on/off systems. has published rules which limit the rate at which domestic heating apparatus may be switched on and off.3 6.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors 6.006.0 32. Simple on/off control with zero-voltage switching avoids generation of RFI but the amount of hysteresis required to prevent temperature oscillations does not give the required control accuracy.8 2. but the high rates of change of current and voltage cause RFI and transients on the mains supply.9 13.5 0.6 The principle of time-proportional control Time proportional control combines the zero-voltage switching of on/off control with the accuracy of proportional control and so eliminates the disadvantages of these two alternative systems.0 24. both phase control and on/off control have disadvantages.3 Design of a Time Proportional Temperature Controller Electronic temperature control is no longer new: phase and on/off controls for heaters have been widely used to replace mechanical switches. the mains voltage changes slightly each time the load is switched on or off.6 3. the ratio of the on and off periods providing the proportional control. There are three states of operation when using time-proportional control: • load switched fully off.6 7.9 1.1 0.0 24.9 2.0 240V 0.2 0.0 10. The TDA1023 is ideally suited for the control of: • Panel heaters • Cooker elements • Electric irons • Water heaters • Industrial applications. to (s) 220V 0.0 4.3 0. • load power proportional to the difference between actual and desired temperatures.0 17. Table 1 gives the minimum repetition period for a range of load powers and common mains voltages from CENELEC publication EN50. CENELEC minimum repetition periods for Domestic Heater Applications Description of the TDA1023 The TDA1023 is a 16-pin dual in-line integrated circuit designed to provide time-proportional power control of electrical heating elements.3 1. Input Ramp voltage Temperature reference Output ON OFF t ON t0 Fig. 537 . phase control is not allowed to be used for domestic heaters. It provides a stabilised reference voltage for the temperature-sensing network. air conditioners. Power (W) 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 Table 1. Figure 1 illustrates the principle.2 0.0 380V 0.0 2. This method of control can cause mains flicker. • load switched fully on.2.3 8. temperature controlled oil baths. However. the load is switched on once and off once in a fixed repetition period.2 0. The TDA1023 Incorporates the following functions: • A stabilised power supply. The TDA1023 has been designed to provide time-proportional control for room heaters and electric heating elements using a minimum number of external components. The TDA1023 may be connected directly to the AC mains using either a dropping resistor or capacitor. Because of this effect. Conventional phase control allows fully-proportional control of the power dissipated in the load.0 16.3 1.0 40. CENELEC.2 0.0 4. e.g. It incorporates additional features to provide fail-safe operation and fine control of the temperature. 1 Duty cycle control Appliance Repetition period. the European Committee for Electro-technical Standardisation.

2 Triac trigger pulse width requirement • A zero-crossing detector to synchronise the output trigger pulses to the zero-crossings of the mains supply. • A control gate circuit to activate the output if there is a mains zero-crossing. the TDA1023 is also suitable for applications requiring on/off control if the timing generator is not used. The repetition period of the timing generator may be set by an external capacitor to conform to the CENELEC specifications for mains load switching. centred on the zero-crossing of the mains voltage. to isolate the voltage translation circuit and comparator from external influences. • A sensor fail-safe circuit to prevent triggering if the thermistor input becomes open or short-circuited. the trigger pulse must last from the time when the load current falls to the value of the triac holding current (IH). • A voltage translation circuit for the potentiometer input. Normally. until the time when the load current reaches the triac latching current (IL). Required Duration of Triac Trigger Pulse The main advantage of triggering at the instant when the applied voltage passes through zero is that this mode of operation renders the use of RF suppression components unnecessary. 3 TDA1023 block diagram and external components 538 . the relatively small temperature variation in a room (5˚C to 30˚C) corresponds to a narrow angle of rotation of a potentiometer shaft. AC line RS R1 6 CS C1 9 Timing generator D1 10 RD 16 Power supply Load 1 Varistor 11 Input buffers 14 Zero crossing detector TDA1023 Voltage translation Comparator Output amplifier 3 RG T1 12 RNTC RP CT 8 7 5 4 13 AC line Fig. with the voltage from the temperature selection dial. the comparator is ON and the fail-safe comparator is OFF. To maintain conduction while the load current is approaching the zero-crossing. • A comparator with adjustable hysteresis. • Input buffers. The detector produces a pulse. Use of this circuit doubles the angle of rotation of the potentiometer shaft for the same temperature range. The amplifier has an output current capability of at least 200mA and is stabilised to 10V while the current limit is not exceeded. Fig. a function of the room temperature. continuous conduction of the triac may be required for many cycles of the mains supply. the duration of which is determined by an external resistor. For time-proportional control. This allows a full 100% control of the load current over a temperature range of only 1˚C or 5˚C. preventing spurious triggering of the output. This compares a thermistor voltage. Although designed specifically for time proportional control.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Triac current IL IH IL Trigger pulse t p(min) IT • A timing generator with an adjustable proportional band. • An output amplifier with a current-limited output.

See Fig. and 200mA for the triac latching current IL and a maximum on-state voltage of VT=1. Vs by R=Vs2/P. The following paragraphs describe the selection of these components to ensure reliable operation under worst-case conditions. Synchronisation Resistor. It compares the current through the synchronisation resistor (RS) with a reference current. P and nominal supply voltage. The graphs of Fig. The trigger pulse width is therefore a function of: • triac latching current (IL) • applied AC voltage (v = Vsinωt) • load resistance (R) • on-state voltage of the triac (VT) at IL.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Fig. and the applied voltage. 4 show tp(MIN) as a function of P for four common mains voltages with values of 30mA. 4 Minimum pulse width as a function of supply voltage and latching current In general. The load resistance is related to the nominal load power. As the supply voltage passes through 539 . Selection of external components The external components required by the TDA1023 determine the operation of the device. 100mA.2V at IL. The current passed by the triac is a function of its on-state voltage. 60mA. the minimum required width of the trigger pulse in the worst case can be calculated. so the minimum trigger pulse duration may be taken as twice the time for the load current in the triac (IT) to rise from zero to the triac’s latching current. the latching current of a triac is higher than the holding current. the load resistance. 2. Assuming that the load resistance has a tolerance of 5% and the AC voltage variation is 10%. RS A current comparator is used as a zero-crossing detector to provide trigger pulse synchronisation.

Figure 7 shows the application of hysteresis to the comparator and the transfer characteristic obtained. Apart from providing a stable two-state output.6 V These figures are for triggering with a positive gate pulse when MT2 is negative with respect to MT1.6V and IGT=98 mA cross the load line for a gate resistor value of 82 Ω. 6. this may be increased by adding a resistor (R4) from Pin 4 to ground which increases the current IH.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors zero. The built-in hysteresis is 20mV. The Proportional Band Resistor R5 The proportional band is the input voltage range that provides control of 0% to 100% of the load power. 6 Gate voltage as a function of output current and gate resistor values 0 100 200 300 400 tp (us) Fig. 10 VGT (V) 8 600 500 6 Vs=380V 240V 220V 4 RG=22R 27 33 400 110V 300 2 39 RG= 390R 0 0 270 180 50 120 100 100 82 68 150 56 IG (mA) 47 200 200 100 0 Fig. Table 2 gives the value of R4 for a range of hysteresis settings. the point at which current limiting occurs is subject to considerable variation between samples of the TDA1023: a gate resistor will reduce the effect of this in production circuits.5kΩ between Pin 1 and Pin 13. Thus. The maximum proportional band of 400mV is obtained by shorting Pin 5 to ground. Pin 4 shorted to ground gives a maximum of 320mV. This rate of change is affected by: • -the AC supply voltage • the supply frequency • the value of the synchronisation resistor. Furthermore. Hysteresis Resistor R4 The comparator of the TDA1023 is designed with built-in hysteresis to eliminate instability and oscillation of the output which would cause spurious triggering of the triac. The rectangular output V/I characteristic of the TDA1023 is shown in Fig. The following example illustrates the use of Fig. The maximum value of gate resistor is therefore 82 Ω. the hysteresis gives the comparator increased noise immunity and prevents half-waving. Gate Termination Resistor RPD The TDA1023 has a resistor approximately 1. 5 Synchronisation resistor values as a function of pulse width and AC voltage Gate Resistor RG The guaranteed minimum amplitude of the output trigger pulse of the TDA1023 is specified as 10V at an output current less than 200mA. 540 . the current in the synchronisation resistor becomes less than the reference current and a trigger pulse is given until the current in RS increases above the reference level. 5 shows the value of RS as a function of trigger pulse width. the duration of the trigger pulse depends upon the rate of change of current in RS at the supply voltage zero-crossing point. Although the output is current-limited. At 0˚C the trigger pulse requirements for a standard BT139 are: IGT = 98 mA VGT = 1. The lines representing VGT= 1. 6. The TDA1023 has a built-in proportional band of Vpb=80mV (corresponding to about 1˚C) which can be increased by the addition of resistor R5 between Pin 5 and ground. R5 (kOhm) The triac to be triggered is a Philips BT139. Inclusion of a gate resistor to limit the gate current to the minimum value required reduces the overall current consumption and the power dissipation in the mains dropping resistor. This is intended for use as a pull-down resistor when sensitive triacs are being used. Fig. Load lines for various values of gate resistor have been plotted on this diagram so that the maximum value of gate resistor can be selected by plotting horizontal and vertical lines to represent the required minimum gate current and voltage. it is still advantageous to include a gate series resistor in the circuit. with the AC supply voltage as a parameter. The output stage is protected against damage due to short-circuits by current-limiting action when the current rises above 200mA.

Timing capacitor values for 220V operation Fig.8 DC value (µF) 68 47 33 22 15 10 Catalogue Number 2222 016 90129 2222 016 90131 2222 015 90102 2222 015 90101 2222 015 90099 2222 015 90098 R1 V6 V9 RP R NTC Table 3. V6 Increasing temperature V7 Fig.0 4. 7 Temperature control hysteresis characteristic When the proportional band (Vpb) is increased.6 2. This timing is selected using the external capacitor CT.3 1.0 R4 (kΩ) 9.1 0.0 7. the effective DC capacitance is different from the marked (AC) value.0 10. the corresponding minimum hysteresis voltage and the maximum value of hysteresis resistor R4. the values of R5 required for these. • Tolerance for the DC capacitance is ±20%.2 3.8 0. 8 Temperature-sensing bridge circuit TDA1023 0. Typical electrolytic capacitors have wide tolerances: up to -10% to +50%.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors VOUT Ih.Rh Smoothing Capacitor.2 3. a high input ripple voltage can be tolerated.43 0.6 9. usually greater.2 t0(max) (s) 65 45 32 21 14 9.0 0.7 1.95V 11 6 7 Fail-safe Comparator Comparator Buffer V11 Translation circuit Buffer 9 RP R NTC 8 Table 4. A practical preferred value of CS is 220µF. Proportional Band (mV) 80 160 240 320 400 R5 (kΩ) 3.3 2. 16V.6 V11 R1 Fig. the use of standard capacitors may lead to repetition periods far in excess of those required. 9 Temperature-sensing voltage translation circuit 541 . • Very low leakage current (<1µA).3 68 47 33 22 15 10 10 10 t0(nom) (s) 41 28 20 13 9 6 6 6 t0(min) (s) 22 15 11 7 4. Thus.000 hours at 40˚C). Preferred capacitors for use with TDA1023 Power (W) 2000 1800 1600 1400 1200 1000 800 600 CENELEC CT(DC) t0 (s) (µF) 24. V11 Table 2. CT The minimum repetition period required for a particular application was given in Table 1. Table 2 also shows a range of proportional band settings.1 4. • Long lifetime (>100. As the TDA1023 possesses an internal voltage stabilization circuit. it may be necessary to increase the hysteresis voltage (Vh). Moreover.0 16. All further references to CT assume the use of the preferred capacitors which have the following advantages: • DC capacitance is known for each marked AC value. CS The smoothing capacitor is required to provide the supply current to the TDA1023 during the negative half cycles of the mains voltage waveform.8 3. Choice of components R4 and R5 Voltage (V) 25 40 25 40 25 40 AC rating (µF) 47 33 22 15 10 6.6 9. A range of electrolytic capacitors has been developed for use with the TDA1023 (Table 3).8 Hysteresis band (mV) 20 40 60 80 100 Timing Capacitor.

and maximum repetition times are also given. an extra comparator is used. minimum. The advantages of using a triangular waveform are that for a given capacitor value the triangular waveform provides twice the repetition period that the sawtooth gives. This fail-safe comparator will inhibit output pulses if the thermistor input voltage rises above a Input voltage translation circuit Figure 8 shows a temperature sensing network which requires a minimum of components and eliminates performance spreads due to potentiometer tolerances. The published data for the TDA1023 specifies the repetition period as 0. When the thermistor input voltage is zero. To sense the open-circuit thermistor condition.2 s/µF. Short-circuit sensing is automatically obtained from the normal temperature sensing circuit. Table 4 shows the minimum preferred value of CT (DC value) to provide the required minimum repetition time for a range of appliance powers operating at 220V AC.6 s ± 0. For applications where the input voltage variation is very much 542 . This allows the use of smaller capacitors and minimises the effects of the capacitor leakage current thus reducing the spread in repetition periods. The resulting nominal. less than the available voltage then the required temperature will be controlled by a small angle of rotation of the potentiometer shaft. Fail-safe circuits The TDA1023 is fail-safe for both short-circuit and open-circuit conditions. If the voltage translation circuit is not used then pins 9 and 11 must be shorted together to disable the circuit. A block diagram of the translation circuit is shown in Fig. 9. The TDA1023 voltage translation circuit allows the use of 80% of the potentiometer rotation giving accurate control of the temperature. Either of these conditions will prevent production of trigger pulses for the triac.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Fig. 10 Average gate current as a function of RG with RS and AC voltage as parameters The timing circuit The TDA1023 employs a triangular waveform for timing purposes. the triac will never be triggered because the potentiometer slider voltage will be higher.

the maximum average output current of the TDA1023 must be determined. The current consumption of the TDA1023 depends upon the hysteresis and proportional band settings. Determination of required supply current Before any calculations concerning the required supply current can be made. Figure 11 shows the minimum supply current as a function of the average output current for limit settings. Below RG=22Ω there is no further increase in IG as the output current is limited. The current drawn by the temperature-sensing circuit must not be greater than 1mA. For worst-case conditions.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors reference value (see Fig. Figure 10 shows graphs of IG(AV)max as a function of RG and RS for four 50Hz supply voltages. 9). a 5% tolerance for RS and RG and a 10% variation of the mains is assumed. Gate 25 current I3 (mA) 20 15 I4=0 I5=0 V4=0 V5=0 10 5 0 5 10 15 20 25 30 Supply current I16 (mA) Fig. 11 Maximum required input current as function of gate current for limits of hysteresis band settings Fig. The minimum supply current required is the sum of the following currents: • the maximum average output current • the current drawn by the temperature-sense circuit • the current required by the integrated circuit. 12 Input current as a function of RD and power dissipation (with and without series diode) 543 .

When selecting a capacitor for mains voltage reduction. This VDR will also protect the triac against current surges in the mains supply. The power dissipated in RD may be considerably reduced by the addition of a series diode as in Fig. for mains voltages below 200V. the additional cost of the required high-voltage capacitor is not justified. and the minimum voltage at Pin 16. 13. a VDR must be included in the circuit as described above.9 4.1  √2 = 1. giving a reduction of more than 50% of the power dissipated in RD. the maximum peak mains voltage. Economies may be gained by the use of smaller or lower power components and so two versions are described in Table 6.3 7. Fig. • AC voltage rating 544 . a VDR (catalogue number 2322 594 13512) will limit the supply voltage to the required level during current transients of up to about 200A. the power dissipated by the dropping resistor is comparatively small and the use of a capacitor is not considered to be necessary.A voltage-dependent resistor must be connected across the mains input to limit mains borne transients. Mains dropping resistor values CD 16 RSD TDA1023 13 Varistor Triac protection If the mains dropping circuit consists of capacitor CD and resistor RSD. 50Hz. Table 5 shows practical values for RD(min) for four common mains supply voltages Supply voltage Vs (V) 110 220 240 380 RD(min) (kΩ) 2.01A 0. RD The value of the mains dropping resistor must be chosen such that the average supply current to the input of the TDA1023 is at least equal to the required minimum. Fig. CD The power dissipated by the dropping resistor has been computed for four mains voltages as a function of RD and the results plotted on the graphs of Fig. the VDR may be connected directly across the triac. In this case there is no conduction through RD during the negative half-cycle of the supply voltage. However. giving improved protection due to the series resistance of the heater.0 3. For 240V operation. giving 100% control of the power dissipation. For RSD=390Ω this yields a maximum transient voltage of about 740V. Application examples The TDA1023 is intended primarily for room temperature control using electric panel heaters. 14. Version A. A practical value for CD is 680nF. If the mains dropping circuit consists of resistor RD and diode D1. • Limit of Inrush current . and thereby reduce the power dissipation in the voltage reduction components still further. The controllable heater power range is from 400W to 2000W. although the upper limit may be increased by suitable choice of triacs and/or heatsinks. uses a BT138 triac and a 15µF timing capacitor.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors The mains dropping resistor. For mains voltages above 240V. it is recommended that capacitive voltage reduction is only used with mains supplies of 200V(RMS) or 240V(RMS). With a 240V (+10%) supply. 12. version B.5 • Suppression of mains-borne transients . Current surges in the supply will not harm the TDA1023 as the dropping resistor will limit the current to a safe level. For 220V operation.The capacitor CD must not be chosen so large that the input current to the TDA1023 violates the absolute maximum specified in the published data. Table 6 gives the necessary component values under worst case conditions for each of these versions for use with mains supplies of 220V. Use of a mains dropping capacitor It is possible to replace the mains dropping resistor and series diode with a capacitor. a VDR (catalogue number 2322 594 13912) will limit the supply voltage to the required level during current transients of up to about 80A.95 × 390 Table 5. for heaters from 400W to 1200W. The value of the resistor RD is defined by the maximum current that can flow into Pin 16. Resistor RSD must also limit the peak value of the inrush current to less than 2A under worst case operating conditions. 13 Use of a mains dropping capacitor. uses a BT139 triac and a 68µF timing capacitor. The TDA1023 may also be used as a time proportional switch for cooker elements and similar devices. the following points must be considered: 1. For these reasons. for heaters from 1200W to 2000W. the value of 390Ω (-5%) will limit the worst case peak value of the inrush current to: 240 × 1. Domestic panel heater controller Figure 3 showed the design for a time proportional heater control using the TDA1023.

the actual capacitor used should be 47µF. Cat. providing a tp(min) of approximately 70µs. To ensure that 0% and 100% load Table 7. and 220V.2kΩ 82Ω 82Ω 430kΩ 180kΩ 47nF 47nF 220µF. 220V.7kΩ R25=22kΩ. Time proportional power control BT139-500 350V.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors The capacitor C1 has been included in the circuit of Fig. From Table 4 the minimum preferred value of CT to provide this period is 68µF. For a load power of 2kW the BT139 triac must be used. 11. Using this value of input current the required value of RD can be found from Fig.3kΩ 6. Value of R1 and RP For control over the range 5˚C to 35˚C and a thermistor characteristics with R25=22kΩ. Once the maximum average output current is known. 50Hz temperature controller components 2. However. so that load power will be zero before the potentiometer reaches its minimum setting.B=4200k R25=22kΩ. latch current and supply voltage (2000W. The temperature-sensing bridge is replaced by a potentiometer.1200W Version B 1200W . due to the different performance under AC and DC conditions. 2000W. 16V 15µF (DC) 68µF (DC) 680nF 470nF 390Ω 390Ω Notes: 1. 3 to minimise sensor line interference pick-up. From Fig. B=4200k 22kΩ 22kΩ 4. No. 16V 220µF. 50Hz. B=4200k No. the power in the load being proportional to the potentiometer setting. Table 7 gives a summary of the required component values. 25V. Only required if used in place of D1 and RD Table 6. The power dissipation in the resistor when diode D1 is present in the circuit is then 5 W. 4 as a function of the load power. Component T1 VDR 1 D1 R1 2 RNTC 3 RP RD RG RS C1 CS CT CD 4 RSD 4 Version A 400W . 2322 642 12223 22kΩ 6. Value of RD First. A suitable value for RP is 22kΩ. the voltage across it will never reach zero. the maximum average output current must be found. the repetition period must be at least 24s (from Table 1). 6 it can be seen that the operation point of 1. 60mA. As the timing capacitor is charged and discharged by current sources. With minimum hysteresis and proportional band.5 mA.6 kΩ. No. the minimum required supply current can be found from Fig. This effect can be reduced by the addition of resistors R1 and R2. The built-in hysteresis and proportional band provides optimum performance for panel heaters so pins 4 and 5 are not connected. then from Table 3. 2322 642 12223 4. Value of RG The maximum value of RG that may be used is determined by the minimum conditions to reliably trigger all samples of the triac. respectively): tp(min)=64µs. This is only necessary when the sensor is remote from the control circuit. Value of CT For a load of 2kW.8kΩ 82Ω 150kΩ 47nF 220µF. 1mA No. 50Hz temperature controller 545 .7kΩ 18. 1% tolerance 3. BT138-500 BT139-500 350V. the average value of the supply current is 12. maximum load power is reached before the maximum setting of the potentiometer. Cat. Temperature control of 2kW load. 12 giving RD = 5. Proportional power control is thus obtained while the potentiometer voltage lies between the upper and lower limits of the triangular waveform comparator input. The circuit is also that shown in Fig. 1mA 350V. 2322 594 13912 BYX10G 18. A value of RS=135kΩ provides a trigger pulse of the required duration. The next preferred value above this is 150kΩ. Similarly. Component T1 VDR D1 R1 RNTC RP RD RG RS C1 CS CT Value Remarks 3.7kΩ 1% tolerance R25=22kΩ. For this circuit IG(AV)max=5 mA. a suitable value of R1 is 18. 10 the maximum gate current IG is given as a function of the values of resistors RS and RG. 2322 594 13512 2. 16V 47µF (DC) No.7kΩ ±1%. 3. 1mA BYX10G BYX10G 18. In Fig.6V and 98 mA lies on the load line for 82Ω and this is the value chosen. 220V. 2222 016 90129 The TDA1023 may be used to provide proportional control of devices such as electric cooker elements.2000W Value of RS The required trigger pulse width can be found from Fig.

6kΩ 82Ω 220kΩ 220µF. 1mA BYX10G 4. 50Hz supplies. Component T1 VDR D1 R1 R2 RP RD RG RS CS CT Value BT139-500 ZnO.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors power can be selected by the potentiometer setting. 16V 47µF.7kΩ 47kΩ 5. with components suitable for the control of loads from 1kW to 2kW from 220V. Time proportional power controller Fig. is shown in Fig. 25V R2 9 CS RP 6 7 12 R1 CT 11 AC line RS D1 RD 14 10 16 Varistor 3 13 RG T1 Load TDA1023 AC line Fig.7kΩ 4. The circuit uses the TDA1023 control chip and an NE555 timer device to give output phase control proportional to the input voltage command. including the timing capacitor CT. 14 Time-proportional power regulation circuit 4. All the circuit components are calculated in the same way as for the temperature controller. Phase control circuit using the TDA1023 Figure 15 shows an adjustable phase control trigger circuit suitable for thyristor or triac controller applications. 15 Adjustable phase SCR/Triac trigger circuit 546 . Table 8. 350V. the values of R1 and R2 should each be limited to 10% of the value of RP. 14 and Table 8. An example circuit.

Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors Hi-Com Triacs 547 .

.

The probability of any device failing commutation is dependent on the rate of rise of reverse voltage (dV/dt) and the rate of decrease of conduction current (dI/dt). and therefore a smaller probability that a sufficient number will be available to re-trigger the triac. The use of emitter shorts in a triac has two effects on commutation. This means there will be fewer carriers left to recombine when the conduction current falls to zero. The use of a Hi-Com triac greatly simplifies circuit design and gives significant cost savings to the designer. This means that at the end of each half cycle the on-state current in one direction must drop to zero and not resume in the other direction until the device is triggered again. 1 Standard triac triggering quadrants Firstly it reduces the gain of the internal transistors that make up the triac. it is necessary to use an RC-snubber network in parallel with the device to limit the dV/dt. and may be thought of as two antiparallel thyristors monolithically integrated onto the same silicon chip. A higher degree of emitter shorting means the presence of more such paths and lower resistance values in them. If the triac were truly two separate thyristors this requirement would not present any problems. The design features of Hi-Com devices that have made this possible are: Geometric separation of the two antiparallel thyristors Commutation failure can be avoided by physically separating the two ’thyristor halves’ of a triac. However. This "commutation" turn-off capability is at the heart of triac power control applications.3. Within the integrated structure of a Hi-Com triac the two halves of the device are kept further apart by modifying the layout of the chip in order to lessen the chance of conduction in one half affecting the other half. This is described as a "commutation failure" and results in the triac continuing to conduct in the opposite direction instead of blocking. The higher the dI/dt the more unrecombined charge carriers are left at the instant of turn-off. separating them into two discrete chips would remove the advantage of a triac being triggerable in both directions by the same gate connection. These devices are suitable for use with a wide variety of motor and inductive loads without the need for a protective snubber. as the two are on the same piece of silicon there is the possibility that the "reverse recovery current" (due to unrecombined charge carriers) of one thyristor as it turns off. The "High Commutation" triacs (Hi-Com triacs) are designed to have superior commutation capability. This product information sheet explains how the superior characteristics and performance of Hi-Com triacs removes design limitations of standard devices. This is at a penalty of extra circuit complexity and dissipation in the snubber. However. snubber network. thus greatly simplifying the circuit.1 Understanding Hi-Com Triacs Hi-Com triacs from Philips Semiconductors are specifically designed to give superior triac commutation performance in the control of motors for domestic equipment and tools. The higher the dV/dt the more probable it is that some of these carriers will act as gate current.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors 6. Thus the commutation capability of any device is usually specified in terms of the turn-off dI/dt and the re-applied dV/dt it can withstand. so that even at a high rate of turn-off (dI/dt) and a high rate of re-applied dV/dt they can be used without the aid of a Emitter shorting "Emitter shorts" refer to the on-chip resistive paths between emitter and base of a transistor. may act as gate current to trigger the other thyristor as the voltage rises in the opposite direction. MT2+ Quadrant 2 + + Quadrant 1 I G I - G G+ G- I G I + MT2+ G Quadrant 3 Quadrant 4 Fig. In phase control circuits the triac often has to be triggered into conduction part way into each half cycle. Triac commutation explained A triac is an AC conduction device. The second way in which emitter shorts help commutation is that any unrecombined carriers in the conducting thyristor at turn-off 549 . If a triac has to be operated in an inductive load circuit with a combination of dI/dt and dV/dt that exceeds its specification. at any particular junction temperature.

Conduction then spreads to the main thyristor element from this boundary. This gives the designer significant savings in design simplicity. one of the triggering modes. This incurs the penalty that the 3+ trigger mode cannot be used. and lessens the danger of any available carriers acting as gate current for undesirable triggering. 1-.and 3+ (or sometimes quadrants 1. Modified gate structure The gate of a triac allows conduction in both directions to be initiated by either a positive or a negative current pulse between gate (G) and main terminal (MT1). board space and system cost. but it greatly improves the commutation performance of the device. 3+ (or quadrant 4).e changing from Conclusions By modifications to the internal design and layout of the triac it is possible to achieve a high commutation capability triac for use in inductive and motor load applications. Unfortunately the carrier distribution in this triggering mode of operation is very similar to that existing when the triac is commutating in the 1-to-3 direction (i. In particular. This both reduces the number of carriers available.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors will have more chance of flowing out through the emitter shorts (of the opposite thyristor) rather than acting as gate current to trigger that thyristor on. 2. 3. 3 and 4) and are shown in Fig. Hi-Com triacs can be used in existing motor control applications without the snubber network required for a standard device. As the loss of the fourth trigger quadrant can usually be tolerated in most designs. 1. For this reason the Hi-Com triacs have a modified gate design to remove this structure. The Hi-Com triacs have a higher degree of emitter shorting both around the periphery of the device and in the central part of the active area. This triggering versatility arises from the fact that the gate consists of some elements which conduct temporarily during the turn-on phase. The four different modes of triggering are often called 1+. The commutation capability of the devices is well in excess of the operating conditions in typical applications. The devices can be used in all typical motor control applications without the need for a snubber circuit. relies on the main terminal 1 supplying electrons to trigger a thyristor element in the gate-MT1 boundary. These modifications have been implemented in the Hi-Com range of devices from Philips Semiconductors. 550 . conduction with MT2 positive to blocking with MT1 positive). The presence of the element in the gate to allow 3+ triggering will therefore always also undermine commutation capability in the 1-to-3 direction.

Philips Semiconductors Hi-Com Triac range 551 . These devices are suitable for use with a wide variety of motor and inductive loads without the need for a snubber. 3 is a typical example of the simplest type of trigger circuit.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors 6. Hi-Com triacs Hi-Com triacs are specifically designed for use with ac inductive loads such as motors.50 1000 24 50 TO220 BTA216-600B 600 16 2 . characteristics.3. When triac commutation occurs the rate of rise of voltage in the opposite direction can be very rapid and is governed by the circuit and device Parameter Repetitive peak voltage RMS on-state current Gate trigger current Off state dv/dt Commutating di/dt Turn-on di/dt Package VDRM (V) IT(RMS) (A) IGT (mA) dVD/dt (V/µs) dIcom/dt (A/ms) dIT/dt (A/µs) BTA212-600B 600 12 2 . 2).2 Using Hi-Com Triacs Hi-Com triacs from Philips Semiconductors are specifically designed to give superior triac commutation performance in the control of motors for domestic equipment and tools. As commutation capability is not an issue for resistive load applications then standard triacs are still the most appropriate devices for these applications. This high dV/dt means there is a much higher probability of charge carriers in the device re-triggering the triac and causing a commutation failure. Hi-Com triacs are equally suitable for use with microcontroller trigger circuits. The circuit of Fig. Triac commutation For resistive loads the device current is in phase with the line voltage. 1.50 1000 28 50 TO220 BTA216-800B 800 16 2 . This product information sheet explains how the need for a triac snubber arises and how the superior performance of Hi-Com triacs removes design limitations of standard devices. The situation is quite different with inductive or motor loads.50 1000 28 50 TO220 Table 1. The use of a Hi-Com triac greatly simplifies circuit design and gives significant cost savings to the designer.50 1000 24 50 TO220 BTA212-800B 800 12 2 . When using Hi-Com triacs in inductive load applications the trigger circuit cannot trigger the device in the fourth (3+) quadrant (Fig. Fortunately the vast majority of circuit designs do not require this mode of operation and so are suitable for use with Hi-Com triacs without modification. The Hi-Com range is summarised in Table 1. This is not a very severe condition for triac commutation: the slow rising dV/dt gives time for the triac to turn off (commutate) easily. For these circuits conduction current lags behind the line voltage as shown in Fig. This removes the requirement for a snubber circuit in inductive load circuits. An additional advantage of the Hi-Com design is that the off-state (static) dv/dt capability of the device is also significantly improved. Under such conditions triac turn-off (commutation) occurs at the voltage "zero-crossover" point. The significant advantage of a Hi-Com triac is that it has no limitation on the rate of rise of reapplied voltage at commutation.

quadrants. As the loss of the fourth trigger quadrant can usually be tolerated in most designs. IGT Trigger current for the Hi-Com triacs is in the range 2mA to 50mA.and 3. during transients such as start-up and faults such as the stalled motor condition. board space and system cost. As discussed above. Supply voltage Load current -dI/dt Voltage across triac Trigger pulses Time dV /dt com VDWM Time Current Time Hi-Com triac M Fig. The trigger current delivered by the trigger circuit must be greater than 50mA under all conditions in order to guarantee triggering of the device when required. 2 Hi-Com triac triggering quadrants Conclusions The Hi-Com range of devices from Philips Semiconductors can be used in all typical motor control applications without the need for a snubber circuit. triggering is 552 . This corresponds to an RMS current of 63A at 50Hz.Thyristors and Triacs Power Semiconductor Applications Philips Semiconductors only possible in the 1+. This means that gate currents due to noise that are below 2mA in amplitude can be guaranteed not to trigger the devices. Hi-Com triacs can be used in existing motor control applications. This dIcom/dt limit is well in excess of the currents that occur in the device under normal operating conditions. This gives the devices a noise immunity feature that is important in many applications. By removing the snubber the use of a Hi-Com triac gives the designer significant savings in design simplicity. For the 12A Hi-Com triacs the limit commutating current is typically 24A/ms at 125˚C. dIcom/dt Hi-Com triacs do not require a snubber network providing that the rate of change of current prior to commutation is less than the rating specified in the device data sheet. Typical stall currents for an 800W domestic appliance motor are in the range 15A to 20A and so the commutation capability of the Hi-Com triacs is well above the requirement for this type of application. Device limiting values i) Trigger current. Fig. This corresponds to an RMS current of 54A at 50Hz. For the 16A Hi-Com triacs the limit commutating current is typically 28A/ms at 125˚C. The commutation capability of the devices is well in excess of the operating conditions in typical applications. 1. 1 Triac commutation waveforms (inductive load) MT2+ Quadrant 2 + + Quadrant 1 I G I No triggering possible - G G+ G- I G + MT2Quadrant 4 Quadrant 3 Fig. 3 Phase control circuit using Hi-Com triac ii) Rate of change of current.

Hammerton D. Hazel Grove: M.Pichowicz W.J.Gant J. The authors thank Mrs.Hooff J.C.Sharples H.J.P.Woodworth T. The authors also thank Mr.Haslam for his assistance in the formatting and printing of the manuscripts.Miller H.D.J.Bennett M.M.B.Moody S.H.Harper W.Humphreys C.Humphreys P.B.P.Preface Power Semiconductor Applications Philips Semiconductors Acknowledgments We are grateful for all the contributions from our colleagues within Philips and to the Application Laboratories in Eindhoven and Hamburg.5.Rosink D.v.Misdom P.Mellor of the University of Sheffield for contributing the application note of section 3.van de Wouw This book was originally prepared by the Power Semiconductor Applications Laboratory. We would also like to thank Dr.Stork D. of the Philips Semiconductors product division.Houldsworth M.Burley It was revised and updated.Brown R.Oosterling N.R.Mulder E.Brown C.Miller L.Simons T.J.J.F.Fry R.Ham C.J.Mellor R.Gilliam D.1.Bennion D.G. de Ruiter D.Ham C.Verhees F.A.J. Contributing Authors N.Sharples .d.J.Grant N.Hettersheid J. by: N.Buethker L.Burley G.Hammerton D. Nijhof J.Hayes for her considerable help in the preparation of this book. in 1994.A.H.Hammerton D.Tebb H.

Preface

Power Semiconductor Applications Philips Semiconductors

Preface
This book was prepared by the Power Semiconductor Applications Laboratory of the Philips Semiconductors product division, Hazel Grove. The book is intended as a guide to using power semiconductors both efficiently and reliably in power conversion applications. It is made up of eight main chapters each of which contains a number of application notes aimed at making it easier to select and use power semiconductors. CHAPTER 1 forms an introduction to power semiconductors concentrating particularly on the two major power transistor technologies, Power MOSFETs and High Voltage Bipolar Transistors. CHAPTER 2 is devoted to Switched Mode Power Supplies. It begins with a basic description of the most commonly used topologies and discusses the major issues surrounding the use of power semiconductors including rectifiers. Specific design examples are given as well as a look at designing the magnetic components. The end of this chapter describes resonant power supply technology. CHAPTER 3 describes motion control in terms of ac, dc and stepper motor operation and control. This chapter looks only at transistor controls, phase control using thyristors and triacs is discussed separately in chapter 6. CHAPTER 4 looks at television and monitor applications. A description of the operation of horizontal deflection circuits is given followed by transistor selection guides for both deflection and power supply applications. Deflection and power supply circuit examples are also given based on circuits designed by the Product Concept and Application Laboratories (Eindhoven). CHAPTER 5 concentrates on automotive electronics looking in detail at the requirements for the electronic switches taking into consideration the harsh environment in which they must operate. CHAPTER 6 reviews thyristor and triac applications from the basics of device technology and operation to the simple design rules which should be followed to achieve maximum reliability. Specific examples are given in this chapter for a number of the common applications. CHAPTER 7 looks at the thermal considerations for power semiconductors in terms of power dissipation and junction temperature limits. Part of this chapter is devoted to worked examples showing how junction temperatures can be calculated to ensure the limits are not exceeded. Heatsink requirements and designs are also discussed in the second half of this chapter. CHAPTER 8 is an introduction to the use of high voltage bipolar transistors in electronic lighting ballasts. Many of the possible topologies are described.

Contents

Power Semiconductor Applications Philips Semiconductors

Table of Contents CHAPTER 1 Introduction to Power Semiconductors
General
1 3

1.1.1 An Introduction To Power Devices ............................................................ 5

Power MOSFET
1.2.1 1.2.2 1.2.3 1.2.4 1.2.5 1.2.6 1.2.7 1.2.8 1.2.9 PowerMOS Introduction ............................................................................. Understanding Power MOSFET Switching Behaviour ............................... Power MOSFET Drive Circuits .................................................................. Parallel Operation of Power MOSFETs ..................................................... Series Operation of Power MOSFETs ....................................................... Logic Level FETS ...................................................................................... Avalanche Ruggedness ............................................................................. Electrostatic Discharge (ESD) Considerations .......................................... Understanding the Data Sheet: PowerMOS ..............................................

17 19 29 39 49 53 57 61 67 69 77 79 83 91 97

High Voltage Bipolar Transistor
1.3.1 1.3.2 1.3.3 1.3.4 Introduction To High Voltage Bipolar Transistors ...................................... Effects of Base Drive on Switching Times ................................................. Using High Voltage Bipolar Transistors ..................................................... Understanding The Data Sheet: High Voltage Transistors .......................

CHAPTER 2 Switched Mode Power Supplies
Using Power Semiconductors in Switched Mode Topologies
2.1.1 An Introduction to Switched Mode Power Supply Topologies ................... 2.1.2 The Power Supply Designer’s Guide to High Voltage Transistors ............ 2.1.3 Base Circuit Design for High Voltage Bipolar Transistors in Power Converters ........................................................................................................... 2.1.4 Isolated Power Semiconductors for High Frequency Power Supply Applications .........................................................................................................

103 105 107 129 141 153 159

Output Rectification

2.2.1 Fast Recovery Epitaxial Diodes for use in High Frequency Rectification 161 2.2.2 Schottky Diodes from Philips Semiconductors .......................................... 173 2.2.3 An Introduction to Synchronous Rectifier Circuits using PowerMOS Transistors ........................................................................................................... 179 i

Contents

Power Semiconductor Applications Philips Semiconductors

Design Examples

185

2.3.1 Mains Input 100 W Forward Converter SMPS: MOSFET and Bipolar Transistor Solutions featuring ETD Cores ........................................................... 187 2.3.2 Flexible, Low Cost, Self-Oscillating Power Supply using an ETD34 Two-Part Coil Former and 3C85 Ferrite .............................................................. 199

Magnetics Design

205

2.4.1 Improved Ferrite Materials and Core Outlines for High Frequency Power Supplies ............................................................................................................... 207

Resonant Power Supplies

217

2.5.1. An Introduction To Resonant Power Supplies .......................................... 219 2.5.2. Resonant Power Supply Converters - The Solution For Mains Pollution Problems .............................................................................................................. 225

CHAPTER 3 Motor Control
AC Motor Control
3.1.1 Noiseless A.C. Motor Control: Introduction to a 20 kHz System ............... 3.1.2 The Effect of a MOSFET’s Peak to Average Current Rating on Invertor Efficiency ............................................................................................................. 3.1.3 MOSFETs and FREDFETs for Motor Drive Equipment ............................. 3.1.4 A Designers Guide to PowerMOS Devices for Motor Control ................... 3.1.5 A 300V, 40A High Frequency Inverter Pole Using Paralleled FREDFET Modules ...............................................................................................................

241 243 245 251 253 259 273 283

DC Motor Control

3.2.1 Chopper circuits for DC motor control ....................................................... 285 3.2.2 A switched-mode controller for DC motors ................................................ 293 3.2.3 Brushless DC Motor Systems .................................................................... 301

Stepper Motor Control

307

3.3.1 Stepper Motor Control ............................................................................... 309

CHAPTER 4 Televisions and Monitors
Power Devices in TV & Monitor Applications (including selection guides)

317

319

4.1.1 An Introduction to Horizontal Deflection .................................................... 321 4.1.2 The BU25XXA/D Range of Deflection Transistors .................................... 331 ii

....... 5......... 389 Monitor Deflection Circuit Example 397 4..................1.. 345 TV Deflection Circuit Examples 349 4.......................... 379 4...1 An Introduction to the 3 pin TOPFET ................3......... 425 Automotive Lamp Control (including selection guides) 433 5................................. 435 The TOPFET 5....3.......................64 kHz Autosync Monitor .................Contents Power Semiconductor Applications Philips Semiconductors 4........ 5.................. 5....................5 Driving TOPFETs .....................................3..1 A Versatile 30 ...... 5............ 5..1............ 351 4.....3. 5.........................3..................................3.......................2 An Introduction to the 5 pin TOPFET ......................... 5................................1. 5....3............ 5.......................3...............8 PWM Control with TOPFET ........ 339 4......................................................................1 Automotive Motor Control With Philips MOSFETS ........................3.............10 3 pin and 5 pin TOPFET Leadforms ........3 BUK101-50DL ............................2 32 kHz / 100 Hz Deflection Circuits for the 66FS Black Line Picture Tube 361 SMPS Circuit Examples 377 4...2.....1 A 70W Full Performance TV SMPS Using The TDA8380 .12 Negative Input and TOPFET ...................3....................................... 5......................3..........16 High Side Linear Drive with TOPFET .......................................15 An Introduction to the High Side TOPFET .................... 5...2 A Synchronous 200W SMPS for 16 and 32 kHz TV .......6 High Side PWM Lamp Dimmer using TOPFET ..............1 Application Information for the 16 kHz Black Line Picture Tubes .................................................3..........................................4 TV and Monitor Damper Diodes ......................................4 Protection with 5 pin TOPFETs .... 5.....3.........2...........11 TOPFET Input Voltage ..........2......................a Microcontroller compatible TOPFET ..... 399 CHAPTER 5 Automotive Power Electronics Automotive Motor Control (including selection guides) 421 423 5..................3.................3.....................................1 Automotive Lamp Control With Philips MOSFETS ..13 Switching Inductive Loads with TOPFET .....................................................................7 Linear Control with TOPFET ................ 5.. 5......3.... iii 443 445 447 449 451 453 455 457 459 461 463 465 467 469 471 473 475 .3...................3 Philips HVT’s for TV & Monitor Applications .14 Driving DC Motors with TOPFET ......................3.......................9 Isolated Drive for TOPFET .................. 5.........................4.......................

............ 485 487 489 497 505 509 521 Thyristor and Triac Applications 6..........................1 Understanding Hi-Com Triacs ............ 551 CHAPTER 7 Thermal Management Thermal Considerations 553 555 7.....................4......................4...3 Design of a Time Proportional Temperature Controller .2 IGBTs for Automotive Ignition .Philips Transistor Selection Guide .1......2.............................. 537 Hi-Com Triacs 547 6...........................................4 Introduction to Thyristors and Triacs ..........3 An Electronic Ballast .........................1.................................................. 483 CHAPTER 6 Power Control with Thyristors and Triacs Using Thyristors and Triacs 6........3 6..2.........1 Efficient Fluorescent Lighting using Electronic Ballasts ...4..............................1......2 Using Hi-Com Triacs ......1 An Introduction to Electronic Automotive Ignition ............1....2 Domestic Power Control with Triacs and Thyristors . 587 8............3 Electronic Switches for Automotive Ignition ...3....1 Thermal Considerations for Power Semiconductors ..... 479 5...... 527 6..............1....... Using Thyristors and Triacs ........... 579 8...........................3....................Base Drive Optimisation ........ 549 6... Understanding Thyristor and Triac Data ...................2.............................1 6.........................................................Contents Power Semiconductor Applications Philips Semiconductors Automotive Ignition 477 5............................ 557 7..1 Triac Control of DC Inductive Loads ...2 6.......2 Electronic Ballasts ........ 589 iv ...........1........ The Peak Current Handling Capability of Thyristors ............................ 523 6..............................................................1.................................1.........1.................... 481 5...................................................................2 Heat Dissipation ......................... 567 CHAPTER 8 Lighting Fluorescent Lamp Control 575 577 8.............

481. 452. 483 ignition. 138. 551 thyristor. 164 Hi-Com triac. 544 CENELEC. 473. 70 Breakover current diac. 459. 537 Burst pulses. 148 Boost converter. 13 Data Sheets High Voltage Bipolar Transistor. 146 drive transformer. 111. 109 discontinuous mode. 348 losses. 455 motor control. 136 base inductor. 328. 479 screen heater. 345 Darlington. 497 Automotive fans see motor control IGBT. 117 Clamp winding.109 Burst firing. 494. 69 . 117 Current fed resonant inverter. 529 Compact fluorescent lamp. 109 continuous mode. 442 seat heater.Index Power Semiconductor Applications Philips Semiconductors Index Airgap. 483 lamps. 134 Baker clamp. 590 Asynchronous. 143 Damper Diodes. diode assisted. 529 Critical electric field. 110 Buck converter. 108 . 473 Avalanche. 469 TOPFET. 328. 564 Capacitance junction. 572 natural. 138. 492. 303 Buck-boost converter. 589 Current Mode Control. 187 power converters. 109 Bootstrap. 148 base resistor. 144. 145 drive transformer leakage inductance. 347 outlines. 133 triac commutation. 592 thyristor. 492 triac. 557 Converter (dc-dc) switched mode power supply. 61 Avalanche breakdown thyristor. 490 Avalanche multiplication. 471. 580 fluorescent lamp. 537 Cooling forced. 285 Clamp diode. 92. 190 Ballast electronic. 147 base inductor. 457. 120 Current tail. 579 Base drive. 585 Continuous mode see Switched Mode Power Supplies Continuous operation. 187. 570 Crest factor. diode assisted. 143 Base inductor. 537 Charge carriers. 523. 442 reverse battery. 348 selection guide. 589 forward converter. 490 i Bridge circuits see Motor Control . 579 switchstart. 141 speed-up capacitor. 425. 492 Breakdown voltage. 113 Anti saturation diode. 435. 481. 134 Cross regulation. 113 Commutation diode.331 MOSFET. 367 forward recovery. 301. 114. 479. 109 output ripple.AC Brushless motor. 580 Choppers. 345 picture distortion. 345. 549 Choke fluorescent lamp. 149 electronic ballast. 442 solenoids.97. 492 Breakover voltage diac. 107 Cookers. 475 resistive loads. transformer core. 29 Capacitor mains dropper. 303 Breakback voltage diac. 147 Base inductor.

ESD TOPFET. 367 Disc drives. 164 reverse recovery softness. 110. 527. 527 Dropper capacitive. 544 resistive. 588 voltage fed push pull. 116. 162 ii ESD. 213 operation. 455 see RFI. 67 ETD core see magnetics F-pack see isolated package Fall time. 115 Flyback converter (two transistor). 113 rectifier circuit. 113 continuous mode. 114 Food mixer. 587. 111. 113 cross regulation. 589 current fed half bridge. 260. 579 colour temperature. 114 diodes. 587 EMC. 91. 116 clamp diode. 86 Baker clamp. ESD precautions. 114 clamp winding. 143. 587 voltage fed half bridge. 492. 117 cross regulation. 302 Discontinuous mode see Switched Mode Power Supplies Domestic Appliances. 164 forward recovery.Index Power Semiconductor Applications Philips Semiconductors dc-dc converter. 113 magnetics. 173 structure. 473 Emitter shorting triac. 561 EFD core see magnetics Efficiency Diodes see Damper Diodes Electric drill. 156. 116 core loss. 117 . 500. 579 efficacy. 168 lifetime control. 144 Fast Recovery Epitaxial Diode (FRED) see epitaxial diode FBSOA. 544. 161 schottky. 114 coupled inductor. 171 snap-off. 114 electronic ballast. 162 epitaxial. 582 leakage inductance. 580 base drive optimisation. 117 diodes. 579 colour rendering. 162 technology. 169 reverse recovery. 584. 199 synchronous rectifier. 161 characteristics. 116 magnetics. 197 continuous mode. 161 Diode Modulator. 582 transistor selection guide. 580 triphosphor. 162 reverse leakage. 163 dI/dt. 579 Flyback converter. 116 advantages. 589 current fed push pull. 165 passivation. 584. 580 Fluorescent lamp. 117 conduction loss. 213 magnetisation energy. 111. 111. 134 Ferrites see magnetics Flicker fluorescent lamp. 530. 587 flyback. 531 Diac. 114 discontinuous mode. 162. 545 Duty cycle. 138 dI/dt triac. 181 transformer core airgap. 167 stored charge. 162 operating frequency. 167 softness factor. 118 disadvantages. 117 duty ratio. 583. 133 Desaturation networks. 111. 583. 327. 591 Diode. 531 Electronic ballast. 113 transistors. 579. 67 see Protection. 180 self oscillating power supply. 119 Depletion region. 167 selection guide. 117 ferrite cores. 6 double diffused. 113 advantages. 115 disadvantages. 549 Epitaxial diode. 116 core saturation. 531 Forward converter.

537 Heatsink. 150 conductivity modulation. 125 diodes. 8. 589 flux symmetry.Index Power Semiconductor Applications Philips Semiconductors operation. 11 Guard ring schottky diode. 254 drive. 150 critical electric field. 157 transistor voltage. 122 synchronous rectifier. 584. 148 base resistor. 250. 180 reset winding. 587. 97 losses. 122 advantages. 92 carrier concentration. 549. 86 horizontal deflection. 116 rectifier circuit. 91 d-type. 111. 125 advantages. 134 Baker clamp. 174 Half bridge. 259 Full bridge converter. 144 FBSOA. 333 avalanche breakdown. 253. 195 Gold doping. 122 diodes. 99. 86. 99. 552 inductive load control. 262 loss. 122 Heat dissipation. 91 device construction. 145 drive transformer leakage inductance. 139 drive transformer. 567 iii Heat sink compound. 144 base drive. 117 switched mode power supply. 473 High Voltage Bipolar Transistor. 92. 569 Heatsink compound. 341 ‘bathtub’ curves. 157. 125 transistors. 139 operation. 131 avalanche multiplication. 187 switching frequency. 91. 143 current tails. 581. 139 electric field. 331 depletion region.AC Half bridge converter. 86. 169 GTO. 196 synchronous rectifier. 254 FREDFETs motor control. 141. 567 Heater controller. 253 Half bridge circuits see also Motor Control . 125 operation. 551 commutation. 346 data sheet. 122 magnetics. 305 bridge circuit. 138. 385 base drive circuit. 88. 86. 256 reverse recovery. 83. 336. 92. 334 fall time. 92. 168 FREDFET. 133 electronic ballast. 86. 91. 147 base inductor. 195 switching losses. 331. 116 output diodes. 544 Heaters. 162. 117 output ripple. 333. 552 gate trigger current. 436 TOPFET. 585. 589 Fact Sheets. 79. 44. 96. 122 cross conduction. 118 Forward converter (two transistor). 111. 254 diode. 117 Forward recovery. 538 Gate drive forward converter. 430. 135. 136. 126 Gate triac. 143. 519. 98 limiting values. 144. 134 current crowding. 138 base-emitter breakdown. 122 electronic ballast. 122 transistors. 124 voltage doubling. 79 dI/dt. 138. 181 transistors. diode assisted. 151 carrier injection. 551 dIcom/dt. 126 disadvantages. 97. 551 High side switch MOSFET. 133 desaturation. 136 current limiting values. 79. 214 operation. 149 dV/dt. 124 disadvantages. 255 charge. 587. 342 Miller capacitance. 132 current tail. 134 hard turn-off. 92. 321. 146 breakdown voltage. 514 Hi-Com triac. 111. 135. 341 leakage current. 150 . 145 base inductor. 122 clamp diodes.

107 Irons. 352. 86. 479. 134 thermal runaway. 154 Isolation. 138. 133 speed-up capacitor. 152 turn-off. 321. 129. 97 overdrive. 92. 154. 561 rectangular pulse. 332. 341. 92. 342 sub emitter resistance. 138 passivation. 347 s-correction. 99. 86. 11. 162 Lighting fluorescent. 435 PWM control. 346 damper diodes. 91. 482. 146. 92. 98 over voltage. 323. 80 ratings. 564 non-rectangular pulse. 321 transistors. 143. 200. 260. 325. 153 J-FET. 455 switch rate. 136. 367 line output transformer. 199. 79. 199 core losses. 88. 80. 155 thermal resistance. 53 switched mode power supply. 208 core materials. 383 snubber. 91. 139 RC network. 347. 207 EFD core. 342 technology. 207 100W 100kHz forward converter. 191 50W flyback converter. 530 Lamps. 438 TOPFET.Index Power Semiconductor Applications Philips Semiconductors optimum drive. 94. 561 Lamp dimmer. 93. 99. 52. 528. 336 control ic. 523 Lifetime control. 321. 363. 273 see motor control ac current fed. 352. 365. 352. 438 inrush current. 404 TDA2595. 481. 154 stray capacitance. 135 switching. composite. 530 Intrinsic silicon. 53 Induction motor see Motor Control . 149 thermal breakdown. 322 IGBT. 408 waveforms. single shot. 9 Junction temperature. 537 Isolated package. 144. 138. 99. 367 base drive. 436 Magnetics. 141. 481. 88 voltage limiting values. 88 outlines. 207 . 346 over current. 132 process technology.AC Inductive load see Solenoid Inrush current. 323 operating cycle. 327. 139 space charge. 91. 484 ignition. 561 burst pulses. 332. 368 TDA4851. 562 rectangular pulse. 149. 197 100W 50kHz forward converter. 98. periodic. 150 underdrive. 138. 481. 432 Logic level MOSFET. 85. 490 Leakage inductance. 83. 565 rectangular pulse. 85. 133 smooth turn-off. 91. 143 storage time. 137. 134 saturation. 406 east-west correction. 113. 345. 92. 331. 400 TDA8433. 339. 352. 354 linearity correction. electric. 483 clamped. 483 Induction heating. 435 dI/dt. 151 turn-on. 148 reverse recovery. 364. 142. 455 Latching current thyristor. 530 Logic Level FET motor control. 401 d-type transistors. 210 ETD core. 97 RBSOA. 151 safe operating area. 130 Horizontal Deflection. 369 test circuit. 367 diode modulator. 483 iv Ignition automotive. 131 power limiting value. 135. 470. 133 Inverter. 305 automotive. 579 phase control. 150 saturation current. 483 darlington. 341 secondary breakdown. 86 SMPS. 92. 98. 438 MOSFET. 557. 367 drive circuit.

1 Thermal Considerations 553 .Thermal Management Power Semiconductor Applications Philips Semiconductors CHAPTER 7 Thermal Management 7.

.

Thermal Management Power Semiconductor Applications Philips Semiconductors Thermal Considerations 555 .

.

power dissipation below Tmb K). 1 Heat transport in a transistor with power dissipation constant with respect to time Fig. somewhat more complicated. If the transistor is subjected to a mounting-base temperature Tmb 1. The capability of a switch to operate in a particular circuit will therefore depend upon the amount of power dissipated internally and the rise in the operating temperature of the silicon junction that this power dissipation causes. and thus prolonged operation of a power transistor above its junction temperature rating is liable to result in reduced life. These operating areas are usually presented for mounting base temperatures of 25 ˚C. its power dissipation rating (together with careful consideration of thermal resistances associated with the device) ensures that the junction temperature rating is not exceeded. For rectifier products such as diodes. its junction temperature will be less than Tjmax by an amount (TmbK . can operate in its on-state at any voltage up to its maximum rating depending on the circuit conditions. Tmb ≤ Tmb K : Ptot maxK = ∆T j − mb max . and the temperature difference between the junction and mounting base ∆Tj-mb as a thermal voltage. be it a power MOSFET or a bipolar.Tmb 1). as shown by the broken line in Fig. By analogy with Ohm’s law. Operation of a device at. It is therefore necessary to specify a Safe Operating Area (SOA) for transistors which specifies the power dissipation limit in terms of a series of boundaries in the current and voltage plane. in turn.1. Transistors are. 2 shows the dependence of the maximum power dissipation on the temperature of the mounting base. Power dissipation causes a rise in junction temperature which will. in the on-state the voltage drop is well defined. or below. Part One describes the essential thermal properties of semiconductors and explains the concept of a limit in terms of continuous mode and pulse mode operation. All power semiconductors have a power dissipation limitation. Rthj − mb 3 that is. operating conditions must be checked to ensure that junction temperatures are not exceeding the desired operating level. 1 illustrates this heat flow. 2. it follows that: Ptot = T j − Tmb Rthj − mb 1 PART ONE The power dissipation limit The maximum allowable power dissipation forms a limit to the safe operating area of power transistors.Thermal Management Power Semiconductor Applications Philips Semiconductors 7. the power dissipation has a fixed limit value (Ptot max K is the maximum d. In the first case. 557 . Continuous power dissipation The total power dissipation in a semiconductor may be calculated from the product of the on-state voltage and the forward conduction current. All power semiconductors dissipate power internally both during the on-state and during the transition between the on and off states. Fig. The thermal equivalent circuit of Fig. The rate at which these changes proceed is exponentially related to temperature. Ptotmax is limited either by a maximum temperature difference: ∆T j − mb max = T jmax − Tmb K 2 or by the maximum junction temperature Tjmax (Tmb K is usually 25˚C and is the value of Tmb above which the maximum power dissipation must be reduced to maintain the operating point within the safe operating area). The amount of power dissipated internally generally speaking increases in line with the power being switched by the semiconductor. At higher temperatures. start chemical and metallurgical changes. however. the power dissipation rating can be easily translated in terms of current ratings. The heat dissipated in the junction of the device flows through the thermal resistance between the junction and the mounting base. Part Two gives worked examples showing junction temperature calculations for a variety of applied power pulse waveforms.1 Thermal Considerations for Power Semiconductors The perfect power switch is not yet available. thyristors and triacs. Rthj-mb. This chapter is divided into two parts. A transistor. It is therefore important that circuit designers are familiar with the thermal characteristics of power semiconductors and are able to calculate power dissipation limits and junction operating temperatures.c. Ptot can be regarded as a thermal current.

see Fig. 4 we obtain: Fig.c. power dissipation in a transistor as a function of the mounting-base temperature In the second case. Ptot max = T jmax − Tmb . If the power dissipation continues. 2. The power dissipation limit may be extended for intermittent operation. duty factor). However. 4). The materials in a power transistor have a definite thermal capacity. even when excessive power is being dissipated in the device. Rthj − mb 4 that is. the power dissipation must be reduced as the mounting base temperature increases along the sloping straight line in Fig. the device will cool (the heating and cooling laws will be identical. = 47. Tmb > Tmb K : Pulse power operation When a power transistor is subjected to a pulsed load.5 W 2 Provided that the transistor is operated within SOA limits. In this case. Ptot maxK = 75 W Tjmax = 175 ˚C Rthj-mb ≤ 2 K/W The maximum permissible power dissipation for continuous operation at a maximum mounting-base temperature of Tmb = 80 ˚C is required. and thus the critical junction temperature will not be reached instantaneously. From Eq. pulse duration) and the frequency with which operation occurs (that is. When the power dissipation ceases. the steeper is the slope of the line. this value is permissible since it is below Ptot max K. The size of the extension will depend on the duration of the operation period (that is.5 50 40 30 20 10 0 0 20 40 60 80 100 Tmb / C 120 140 160 180 Fig. Some heat energy will be stored by the thermal capacity of the device. a balance will be struck between heat generation and removal resulting in the stabilisation of Tj and ∆Tj-mb. if the power dissipation ceases before the temperature of the transistor stabilises. 5). Tmb is the maximum mounting-base temperature that can occur in operation.Thermal Management Power Semiconductor Applications Philips Semiconductors Ptotmax / W 100 90 80 70 60 47. higher peak power dissipation is permitted. 3 Example of the determination of maximum power dissipation Fig. 3) for the relevant transistor. 2 Maximum d. the peak values of Tj and ∆Tj-mb will be less than the values Ptot max = 175 − 80 W. 4 Heating of a transistor chip If power is applied to a transistor. the device will immediately start to warm up (Fig. Example The following data is provided for a particular transistor. and the stable conditions will be determined by the thermal resistances associated with the transistor and its thermal environment. The same result can be obtained graphically from the Ptot max diagram (Fig. Equation 4 shows that the lower the thermal resistance Rthj-mb. Note that the maximum value of Tmb is chosen to be significantly higher than the maximum ambient temperature to prevent an excessively large heatsink being required. 558 .

Further pulses will build up the temperature until some new stable situation is attained (Fig. This can happen with high-power low-duty-factor pulses. 7 A train of power pulses increases the average temperature if the device does not have time to cool between pulses ID / A A N )= VD S/ ID 100 Fig. 8 shows a typical safe operating area for d. and rectangular pulse operating areas with fixed parameters δ=0. and the duty factor δ. operation of a power MOSFET. This rectangle represents the largest possible pulse operating area. the lower the temperature that the junction reaches. 5 Heating and cooling follow the same law RD O S( BUK553-100 tp = 10 us B 100 us 1 ms 10 ms 100 ms 10 1 DC 0. When the pulse time becomes very short.C. 8 D. the shorter the pulse and the lower the frequency. 6 The peak temperature caused by a short power pulse can be less than the steady-state temperature resulting from the same power Fig. These boundaries represent the largest possible extension of the operating area for particular pulse times. If the second pulse is identical to the first. Zthj − mb 5 where Zthj-mb is the transient thermal impedance between the junction and mounting base of the device. 7). and the pulse time tp as a parameter. Fig. The corresponding rectangular-pulse operating areas with a fixed duty factor. it follows that: Ptot M = T j − Tmb .c. the power dissipation does not have a limiting action and the pulse current and maximum voltage form the only limits. the peak temperature attained by the device at the end of the second pulse will be greater than that at the end of the first pulse. If the upward excursions extend into the region of excessive Tj then the life expectancy of the device may be reduced. even though the average power is below the d. are also shown. rating of the device.Thermal Management Power Semiconductor Applications Philips Semiconductors reached for the same level of continuous power dissipation (Fig.c.1 1 10 VDS / V 100 Fig. The temperature of the device in this stable condition will fluctuate above and below the mean. δ = 0. where: 559 . By analogy with Eq. tp and Tmb=25˚C In general. Fig. 3. It depends on the pulse duration tp. 6).

Zthj − mb T jmax − Tmb . and above Tmb K. 2). This time depends on the material and construction of the case.5. and other factors. 9 Example of the presentation of the transient thermal impedance as a function of the pulse time with duty factor as parameter Again. Zth j-mb / (K/W) BUK553-100A The Zthj-mb curves approach this value asymptotically as tp decreases. 9 shows that Zthj-mb approaches this value as tp becomes large.0 seconds. transient thermal effects die out in most power transistors within 0. the way it is mounted. Zthj − mb 7 Fig. for duty factors in the range 0.1 0. 1E+01 = 1E+00 0. the limit values given by Eq.1 to 1. the junction temperature approaches a stationary value towards the end of a pulse.5 0. 10 have virtually been reached at tp = 10-6 s. Consequently. owing to the internal thermal capacity of the transistor. Fig. 560 . That is. below a mounting-base temperature of Tmb K. the power dissipation must be reduced linearly with increasing mounting-base temperature. that is: lim Zthj − mb = Rthj − mb 11 Ptot(av) = δPtot M The transient thermal impedance becomes: 9 Fig. or by the maximum junction temperature Tjmax. the fluctuations of junction temperature become negligible. Power pulses with a duration in excess of this time have approximately the same effect as a continuous load.1 to 0. Long pulse duration (Fig. 9 shows that. The transient thermal impedance tends to the thermal resistance for continuous power dissipation.05 0. the size of the chip. with duty factor as a parameter. 10b) As the pulse duration increases. 10 Three limit cases of rectangular pulse loads: (a) short pulse duration (b) long pulse duration (c) single-shot pulse Ptot M max = 8 when Tmb > Tmb K. and: ∆T j − mb max .02 P D tp tp T t 1E+01 1E-02 0 = T 1E-03 1E-07 1E-05 1E-03 t/s 1E-01 Fig. that is: tp → ∞ Short pulse duration (Fig. Fig. the maximum power dissipation has a fixed limit value.2 1E-01 0.Thermal Management Power Semiconductor Applications Philips Semiconductors tp δ= . and so by analogy with Eqs. In general. the maximum pulse power dissipation is limited either by the maximum temperature difference ∆Tj-mb max (Eq. 10a) As the pulse duration becomes very short. 9 shows a typical family of curves for thermal impedance against pulse duration. 3 and 4: Ptot maxK = when Tmb ≤ Tmb K. T 6 tp → 0 lim Zthj − mb = δRthj − mb 10 and T is the pulse period. the only factor to be considered is the heating of the junction by the average power dissipation.

Periodic rectangular pulse Fig. the following formula is used: Tj = Tmb + ∆Tj − mb 14 The value for Zth j-mb is taken from the δ=0 curve shown in Fig.05 400 PART TWO Calculating junction temperatures Most applications which include power semiconductors usually involve some form of pulse mode operation. In this example 100W is dissipated for a period of 20µs. To work out the peak junction temperature the following steps are used: 20 40 60 80 100 400 420 440 460 TIME(uS) 480 t = 2 × 10−5s P = 100W δ=0 Tjpeak Tmb 0 20 40 60 80 100 400 420 440 460 TIME(uS) 480 Zth j − mb = 0.05.12 = 12°C Tj = Tmb + ∆Tj − mb = 75 + 12 = 87°C Average Tj : Pav = P × δ = 100 × 0. the junction tends to cool down completely between pulses so that each pulse can be treated individually. 12 (This diagram repeats Fig. Examples are given for a variety of waveforms: (1) Periodic Waveforms (2) Single Shot Waveforms (3) Composite Waveforms (4) A Pulse Burst (5) Non Rectangular Pulses From the point of view of reliability it is most important to know what the peak junction temperature will be when the power waveform is applied and also what the average junction temperature is going to be.05 = 5W ∆Tj − mb(av) = Pav × Zthj − mb(δ = 1) = 5 × 2 = 10°C Tj(av) = Tmb + ∆Tj − mb(av) = 75 + 10 = 85°C The value for Zth j-mb is taken from the δ=0.12K/W ∆Tj − mb = P × Zth j − mb = 100 × 0. Peak junction temperature will usually occur at the end of an applied pulse and its calculation will involve transient thermal impedance. 561 . the Zthj-mb values for δ = 0 (Fig. When considering single pulses. 9 but has been simplified for clarity). representing a duty cycle (δ) of 0. The pulse used is the same as in the previous example.Thermal Management Power Semiconductor Applications Philips Semiconductors Single-shot pulses (Fig.04K/W ∆Tj − mb = P × Zth j − mb = 100 × 0.c. POWER(W) 110 100 90 80 70 60 50 40 30 20 10 0 0 Tj Zth j − mb = 0. The above calculation shows that the peak junction temperature will be 85˚C. which should highlight the differences between periodic and single shot thermal calculations.05 curve shown in Fig. 9) give sufficiently accurate results. 11 Periodic Rectangular Pulse When considering the junction temperature in a device. Single shot rectangular pulse Fig. where ∆Tj-mb is found from a rearrangement of equation 7. 100W is dissipated every 400µs for a period of 20µs. the time period between pulses is infinity. The average junction temperature (where applicable) is calculated by working out the average power dissipation using the d. This type of pulse is commonly found in switching applications. 11 shows an example of a periodic rectangular pulse. This section gives several worked examples showing how junction temperatures can be simply calculated. The peak junction temperature is calculated as follows: Peak Tj : t = 2 × 10−5s P = 100W δ= 20 = 0. 10c) As the duty factor becomes very small.04 = 4°C Fig. thermal resistance. The above calculation shows that the peak junction temperature will be 4˚C above the mounting base temperature. For a single shot pulse. In all the following examples the mounting base temperature (Tmb) is assumed to be 75˚C. 12. ie the duty cycle δ=0. 13 shows an example of a single shot rectangular pulse.

000 0. Table 1a gives the values calculated for this example. The period is 400µs. For instance the first superimposed pulse in Fig. t1 t2 t3 t4 20µs 0.Zth j − mb(t1) + P2.375 0. in addition to suitable values of tp and δ.130 0. The Zth values are taken from Fig. To be able to add the various effects of the pulses at this time. Composite pulse parameters for time tx 562 . but both positive and negative amplitudes.120 For a single shot pulse.040 Tmb 0 20 40 60 80 100 180µs 170µs 150µs Fig.45 = δ. This type of signal can be simulated by superimposing several rectangular pulses which have a common period. while negative pulses decrease it. the equivalent duty cycle must be worked out. 12 Thermal impedance curves for δ=0. the average power dissipated and average junction temperature are not relevant.900 0.000 0.Zth j − mb(t2) − P2. 13 Single Shot Pulse Repetitive T=400µs Single Shot T=∞ δ Zth δ Zth 0. must end at time tx in the first calculation and ty in the second calculation.450 0.05 1E-02 0 P D tp = tp T t 1E+01 T 1E-03 1E-07 1E-05 1E-03 t/s 1E-01 Fig.425 0. a power device frequently has to handle composite waveforms. 14 lasts for a time t1 = 180µs. Positive pulses increase the junction temperature.Thermal Management Power Semiconductor Applications Philips Semiconductors 1E+01 Zth j-mb / (K/W) Composite rectangular pulse In practice. 9 to find a value for Zth.050 0. P2=20W and P3=100W. both positive and negative. P2 and P3 are known: P1=40W. all the pulses.05 and δ=0 POWER(W) 110 100 90 80 70 60 50 40 30 20 10 0 0 20 40 60 80 100 Calculation for time tx ∆Tj − mb@x = P1. representing a duty cycle of 180/400 = 0. the values for P1.800 0.9K/W. 20W for 150µs and 100W for 20µs.Zth j − mb(t4) 15 Tj Tjpeak In equation 15. 14. To show the way in which the method used for periodic rectangular pulses is extended to cover composite waveforms. which in this case is 0.Zth j − mb(t3) + P3. These values can then be used in conjunction with Fig.130 0. consider the composite waveform shown in Fig. 9.125 0.000 0. The peak junction temperature may be calculated at any point in the cycle.12 0. By way of an example. namely 40W for 10µs. = 1E+00 0. rather than the simple rectangular pulses shown so far. Table 1a.04 1E-01 0. For each term in the equation.000 0.850 0. the waveform shown has been chosen to be an extension of the periodic rectangular pulse example. and the waveform consists of three rectangular pulses.Zth j − mb(t4) − P1.

Thermal Management

Power Semiconductor Applications Philips Semiconductors

POWER (W)
100 80 60 40 20 0 0 200 180 160 140 120 100 80 60 40 20 0 20 40 60 20 40 60 80 100 120 140 160

tx

ty

180

200

220

360

380

400

420

440

460

480

500

POWER (W) t1 t2 t3 t4 Time (uS)

P3

P2 P1 P1 P2 Time (uS)

POWER (W)
160 140 120 100 80 60 40 20 0 20 40 60 80 100 120

P1

P3 P2 P2 Time (uS)

P3 t5 t6 t7 t8

Fig. 14 Periodic Composite Waveform

Substituting these values into equation 15 for Tj-mb@x gives
Repetitive: ∆Tj − mb@x = 40 × 0.9 + 20 × 0.85 + 100 × 0.13 − 40 × 0.85 − 20 × 0.13 = 29.4°C Tj = Tmb + ∆Tj − mb = 75 + 29.4 = 104.4°C

Single Shot:

∆Tj − mb@x = 40 × 0.13 + 20 × 0.125 + 100 × 0.04 − 40 × 0.125 − 20 × 0.04 = 5.9°C Tj = Tmb + ∆Tj − mb = 75 + 5.9 = 80.9°C

Hence the peak values of Tj are 104.4˚C for the repetitive case, and 80.9˚C for the single shot case.

563

Thermal Management

Power Semiconductor Applications Philips Semiconductors

Calculation for time ty
∆Tj − mb@y = P2.Zth j − mb(t5) + P3.Zth j − mb(t6) + P1.Zth j − mb(t8) − P2.Zth j − mb(t6) − P3.Zth j − mb(t7) 16

Burst pulses
Power devices are frequently subjected to a burst of pulses. This type of signal can be treated as a composite waveform and as in the previous example simulated by superimposing several rectangular pulses which have a common period, but both positive and negative amplitudes, in addition to suitable values of tp and δ.
POWER (W) T=240us

where Zthj-mb(t) is the transient thermal impedance for a pulse time t. t5 t6 t7 380µs 250µs 230µs Repetitive T=400µs Single Shot T=∞ δ Zth δ Zth 0.950 0.625 1.950 1.300 0.000 0.000 0.200 0.160 0.575 1.250 0.000 0.150 t8 10µs 0.025 0.080 0.000 0.030

150 100 50 0 350 300 250 200 150 100 50 0 50 100 150 200 250 300 350 0 20 40 60 80 100 120 140 160 t5 t3 t1 t2 t4 Time (us) 240 260 280 300 Time (us)

Table 1b. Composite pulse parameters for time ty Substituting these values into equation 16 for Tj-mb@y gives
Repetitive: ∆Tj − mb@y = 20 × 1.95 + 100 × 1.3 + 40 × 0.08 − 20 × 1.3 − 100 × 1.25 = 21.2°C Tj = Tmb + ∆Tj − mb = 75 + 21.2 = 96.2°C Single Shot: ∆Tj − mb@y = 20 × 0.2 + 100 × 0.16 + 40 × 0.03 − 20 × 0.16 − 100 × 0.15 = 3°C Tj = Tmb + ∆Tj − mb = 75 + 3 = 78°C

Fig. 15 Burst Mode Waveform Consider the waveform shown in Fig. 15. The period is 240µs, and the burst consists of three rectangular pulses of 100W power and 20µs duration, separated by 30µs. The peak junction temperature will occur at the end of each burst at time t = tx = 140µs. To be able to add the various effects of the pulses at this time, all the pulses, both positive and negative, must end at time tx. Positive pulses increase the junction temperature, while negative pulses decrease it.
∆Tj − mb@x = P.Zth j − mb(t1) + P.Zth j − mb(t3) + P.Zth j − mb(t5) − P.Zth j − mb(t2) − P.Zth j − mb(t4) 17

where Zthj-mb(t) is the transient thermal impedance for a pulse time t. The Zth values are taken from Fig. 9. For each term in the equation, the equivalent duty cycle must be worked out. These values can then be used in conjunction with Fig. 9 to find a value for Zth. Table 2 gives the values calculated for this example. t1 δ Zth δ Zth t2 t3 t4 t5

Hence the peak values of Tj are 96.2˚C for the repetitive case, and 78˚C for the single shot case. The average power dissipation and the average junction temperature can be calculated as follows:
Pav = 25 × 10 + 5 × 130 + 20 × 100 400

= 7.25W ∆Tj − mb(av) = Pav × Zthj − mb(δ = 1) = 7.25 × 2 = 14.5°C Tj(av) = Tmb + ∆Tj − mb(av) = 75 + 14.5 = 89.5°C

120µs 100µs 70µs 50µs 20µs Repetitive T=240µs Single Shot T=∞ 0.500 0.420 0.290 0.210 0.083 1.100 0.800 0.600 0.430 0.210 0.000 0.000 0.000 0.000 0.000 0.100 0.090 0.075 0.060 0.040

Clearly, the junction temperature at time tx should be higher than that at time ty, and this is proven in the above calculations.

Table 2. Burst Mode pulse parameters Substituting these values into equation 17 gives

564

Thermal Management

Power Semiconductor Applications Philips Semiconductors

Repetitive:

∆Tj − mb@x = 100 × 1.10 + 100 × 0.60 + 100 × 0.21 − 100 × 0.80 − 100 × 0.43 = 68°C Tj = 75 + 68 = 143°C

50 40 30 20 10 0 0 20 40 60 80 100

50 40 30 20 10 0 0 20 40 60 80 100

100 90 80 70 60 50 40 30 20 10 0 10

100 90 80 70 60 50 40 30 20 10 0 10 20 30 P1 t1 P2 t2 t3 P3

Single Shot:

∆Tj − mb@x = 100 × 0.10 + 100 × 0.075 + 100 × 0.04 − 100 × 0.09 − 100 × 0.06 = 6.5°C Tj = 75 + 6.5 = 81.5°C

Hence the peak value of Tj is 143˚C for the repetitive case and 81.5˚C for the single shot case. To calculate the average junction temperature Tj(av):
Pav = 3 × 100 × 20 240

20 30

Fig. 16 Non Rectangular Waveform
∆Tj − mb = Ptot M × Zthj − mb Single Shot ∆Tj − mb = 50 × 0.065 = 3.25°C Tjpeak = 75 + 3.25 = 78.5°C 10% Duty cycle ∆Tj − mb = 50 × 0.230 = 11.5°C Tjpeak = 75 + 11.5 = 86.5°C 50% Duty cycle ∆Tj − mb = 50 × 1.000 = 50°C Tjpeak = 75 + 50 = 125°C

= 25W ∆Tj − mb(av) = Pav × Zthj − mb(δ = 1) = 25 × 2 = 50°C Tj(av) = 75 + 50 = 125°C

The above example for the repetitive waveform highlights a case where the average junction temperature (125˚C) is well within limits but the composite pulse calculation shows the peak junction temperature to be significantly higher. For reasons of improved long term reliability it is usual to operate devices with a peak junction temperature below 125˚C.

When the waveform is split into two rectangular pulses:
∆Tj − mb = P3.Zth j − mb(t3) + P1.Zth j − mb(t1) − P2.Zth j − mb(t2) 18

Non-rectangular pulses
So far, the worked examples have only covered rectangular waveforms. However, triangular, trapeziodal and sinusoidal waveforms are also common. In order to apply the above thermal calculations to non rectangular waveforms, the waveform is approximated by a series of rectangles. Each rectangle represents part of the waveform. The equivalent rectangle must be equal in area to the section of the waveform it represents (ie the same energy) and also be of the same peak power. With reference to Fig. 16, a triangular waveform has been approximated to one rectangle in the first example, and two rectangles in the second. Obviously, increasing the number of sections the waveform is split into will improve the accuracy of the thermal calculations. In the first example, there is only one rectanglular pulse , of duration 50µs, dissipating 50W. So again using equation 14 and a rearrangement of equation 7:

For this example P1 = 25W, P2 = 25W, P3 = 50W. Table 3 below shows the rest of the parameters: t1 75µs Single Shot T=∞ 10% Duty Cycle T=1000µs 50% Duty Cycle T=200µs D Zth D Zth D Zth 0.000 0.085 0.075 0.210 0.375 0.700 t2 t3

50µs 37.5µs 0.000 0.065 0.050 0.140 0.250 0.500 0.000 0.055 0.037 0.120 0.188 0.420

Table 3. Non Rectangular Pulse Calculations

565

14 = 7.5 × 2 = 25°C Tj(av) = 75 + 25 = 100°C Conclusion A method has been presented to allow the calculation of average and peak junction temperatures for a variety of pulse types.25 = 78.75 = 82. To calculate the average junction temperature: 10% Duty Cycle Pav = 50 × 50 1000 = 2.5W ∆Tj − mb(av) = Pav × Zthj − mb(δ = 1) = 12.42 + 25 × 0.085 − 25 × 0.055 + 25 × 0. care must be taken to ensure that the calculation gives the peak junction temperature. as it may not occur at the end of the pulse. The method for non rectangular pulses can be applied to any wave shape. In this instance several calculations must be performed with different endpoints to find the maximum junction temperature.5 × 2 = 5°C Tj(av) = 75 + 5 = 80°C 566 .5W ∆Tj − mb(av) = Pav × Zthj − mb(δ = 1) = 2.7 − 25 × 0.5°C 50% Duty cycle ∆Tj − mb = 50 × 0.12 + 25 × 0. Several worked examples have shown calculations for various common waveforms.5°C 10% Duty cycle ∆Tj − mb = 50 × 0.065 = 3. For pulses such as these.5 = 26°C Tjpeak = 75 + 26 = 101°C 50% Duty Cycle Pav = 50 × 50 200 = 12.Thermal Management Power Semiconductor Applications Philips Semiconductors Sustituting these values into equation 18 gives: Single shot ∆Tj − mb = 50 × 0.75°C Tjpeak = 75 + 7.21 − 25 × 0. allowing temperature calculations for waveforms such as exponential and sinusoidal power pulses.25°C Tjpeak = 75 + 3.

2 Heat Dissipation All semiconductor failure mechanisms are temperature dependent and so the lower the junction temperature. Care should be taken when drilling holes in heatsinks to avoid burring and distorting the However. 1a). and the heatsink loses heat to the surroundings by convection and radiation. when a heatsink is not used. However. The formulae and diagrams given in this section can only be considered as a guide for determining the nature of a heatsink. and heating from these components. 1b shows that. Thus our data specifies a maximum junction temperature which should not be exceeded under the worst probable conditions. Fig. the total thermal resistance between junction and ambient is given by: Rth j − amb = Rth j − mb + Rthmb − amb 1 Contact thermal resistance Rth mb-h The thermal resistance between the transistor mounting base and the heatsink depends on the quality and size of the contact areas. Normally. Heat flow path The heat generated in a semiconductor chip flows by various paths to the surroundings.1 is usually the surrounding air. and the contact pressure. The contact thermal resistance Rth mb-h is usually small with respect to (Rth j-mb + Rth h-amb) when cooling is by natural convection. This is because the thermal resistance of a heatsink depends on numerous parameters which cannot be predetermined. or by conduction to cooling water (Fig. the additional thermal resistance between the inside and outside of the equipment case should be taken into account. Generally air cooling is used so that the ambient referred to in Fig. are usually mounted on heatsinks because of the higher power dissipation they experience. the ratio of the lengths of the sides of the heatsink. They include the position of the transistor on the heatsink.Thermal Management Power Semiconductor Applications Philips Semiconductors 7. the screening effect of nearby components. Power transistors. the total thermal resistance is given by: Rth j − amb = Rth j − mb + Rthmb − h + Rthh − amb 2 567 . Thus careful consideration of these thermal resistances (or impedances) allows the user to calculate the maximum power dissipation that will keep the junction temperature below a chosen value. the heat flows from the junction to the mounting base which is in close contact with the case. The more complex the heat dissipation conditions. However. the heatsink thermal resistance Rth h-amb can be very small when either forced ventilation or water cooling are used. Small signal devices do not usually require heatsinking. Paint finishes of normal thickness. and so contact will take place on several points only. power transistors are generally mounted on a heatsink since Rth j-amb is not usually small enough to maintain temperatures within the chip below desired levels. 1 Thermal resistances in the heat flow process: (a) Without a heatsink (b) With a heatsink Thermal resistance calculations Fig. with a small air-gap over the rest of the area. the higher the reliability of the circuit. the more important it becomes to carry out such checks. Transistor case and heatsink surfaces can never be perfectly flat. The use of a soft substance to fill this gap lowers the contact thermal resistance. however. Heat flows from the transistor case to the heatsink by way of contact pressure.1. Fig. 1a shows that. Heat is then lost by the case to the surroundings by convection and radiation (Fig. metal. the extent to which air can flow unhindered. the type of any intermediate plates used. Proprietary heatsinking compounds are available which consist of a silicone grease loaded with some electrically insulating good thermally conducting powder such as alumina. up to 50 um (as a protection against electrolytic voltage corrosion). It is always advisable to check important temperatures in the finished equipment under the worst probable operating conditions. derating the operating temperature from Tjmax is always desirable to improve the reliability still further. Note that if this is the air inside an equipment case. and thus a close thermal contact between the transistor case and heatsink becomes particularly important. when a heatsink is used. the gap is filled with a heatsinking compound which remains fairly viscous at normal transistor operating temperatures and has a high thermal conductivity. The junction temperature depends on both the power dissipated in the device and the thermal resistances (or impedances) associated with the device. 1b). barely affect the thermal resistance. and both mating surfaces should be clean. The use of such a compound also prevents moisture from penetrating between the contact surfaces.

1 are inappropriate for intermittent operation. Tmb = T j − PtotM .Rth j − mb 8 Intermittent operation The thermal equivalent circuits of Fig.Thermal Management Power Semiconductor Applications Philips Semiconductors Note that the direct heat loss from the transistor case to the surroundings through Rth mb-amb is negligibly small. Rth j − amb = and Rth j − mb = T j − Tamb Ptot(av) T j − Tmb Ptot(av) 3 Fig. 3) and Eq. Thus. and the heatsink thermal resistance can be obtained from Eq. if the pulse duration tp exceeds one second. the transistor is temporarily in thermal equilibrium since such a pulse duration is significantly greater than the thermal time-constant of most transistors. it is no longer valid to assume that the mounting base temperature is constant since the pulse time is also no longer small with respect to the thermal time constant of the heatsink. the temperature difference Tj . In addition. This is because heatsinks have a high thermal capacity and thus a high thermal time-constant.Zth j − mb The mounting-base temperature has always been assumed to remain constant under intermittent operation. The value of Tmb can be calculated from Eq.Tmb reaches a stationary final value (Fig. 4 Combining equations 2 and 3 gives: Rthh − amb = T j − Tamb − Rth j − mb − Rthmb − h Ptot(av) 5 and substituting Eq 4 into Eq 5 gives: Rthh − amb = Tmb − Tamb − Rthmb − h Ptot(av) 6 Now. This assumption is known to be valid in practice provided that the pulse time is less than about one second.6.7 should be replaced by: The values of Rth j-mb and Rth mb-h are given in the published data. The first stage in determining the size and nature of the required heatsink is to calculate the maximum heatsink thermal resistance Rth h-amb that will maintain the junction temperature below the desired value Continuous operation Under dc conditions. Consequently. the maximum heatsink thermal resistance can be calculated directly from the maximum desired junction temperature. provided that the pulse time is less than one second. 3 Variation of junction and mounting base temperature when the pulse time is not small compared with the thermal time constant of the heatsink 568 . for pulse times of more than one second. 2. either Eq. Fig. and the thermal impedance Zth j-mb should be considered. PtotM = thus: T j − Tmb Zth j − mb 7 Tmb = T j − PtotM . 7. Thus Eq.6 is valid for intermittent operation. 2 Variation of junction and mounting base temperature when the pulse time is small compared with the thermal time constant of the heatsink The thermal time constant of a transistor is defined as that time at which the junction temperature has reached 70% of its final value after being subjected to a constant power dissipation at a constant mounting base temperature. 5 or Eq. The mounting-base temperature does not change significantly under these conditions as indicated in Fig.6 can be used to find the maximum heatsink thermal resistance.

6 is replaced by the peak power dissipation to obtain the value of the thermal impedance between the heatsink and the surroundings In Fig. With finned heatsinks. and the thermal resistance of a flat plate heatsink painted gloss white will be only about 3% higher than that of the same heatsink painted matt black. Finned heatsinks Finned heatsinks may be made by stacking flat plates. The material normally used for heatsink construction is aluminium although copper may be used with advantage for flat-sheet heatsinks. Heatsink finish Heatsink thermal resistance is a function of surface finish. The thermal resistance obtained depends on the thickness. the graph is accurate to 25%.6. although it is usually more economical to use ready made diecast or extruded heatsinks.25:1. Since most commercially available finned heatsinks are of reasonably optimum design. 4. Heatsinks Three varieties of heatsink are in common use: flat plates (including chassis). A plate mounted horizontally will have about twice the thermal resistance of a vertically mounted plate. This comparison is made in Fig. 4 Generalised heatsink characteristics: flat vertical black aluminium plates. area and orientation of the plate. although they are approximately ten times better than a bright aluminium metal finish. The colour of the paint used is relatively unimportant. 5 Generalised heatsink characteristic: blackened aluminium finned heatsinks 569 . approximately square Flat-plate heatsinks The simplest type of heatsink is a flat metal plate to which the transistor is attached. Zthh − amb Tmb − Tamb = − Rthmb − h PtotM 9 The value of Zth h-amb will be less than the comparable thermal resistance and thus a smaller heatsink can be designed than that obtained using the too large value calculated from Eq. Fig. Metallic type paints. painting is less effective since heat radiated from most fins will fall on adjacent fins but it is still worthwhile. where about one third of the heat is dissipated by radiation. such as aluminium paint. 3mm thick. Fig. it is possible to compare them on the basis of the overall volume which they occupy. as well as on the finish and power dissipated. 5 for heatsinks with their fins mounted vertically. have the lowest emissivities. This is particularly important where the equipment chassis itself is used as the heatsink. The average power dissipation in Eq. Both anodising and etching will decrease the thermal resistivity. The graph is accurate to within 25% for nearly square plates.Thermal Management Power Semiconductor Applications Philips Semiconductors Smaller heatsinks for intermittent operation In many instances. Small finned clips are sometimes used to improve the dissipation of low-power transistors. diecast finned heatsinks. A painted surface will have a greater emissivity than a bright unpainted one. again. where the ratio of the lengths of the sides is less than 1. The effect is most marked with flat plate heatsinks. and extruded finned heatsinks. the thermal resistance of a blackened heatsink is plotted against surface area (one side) with power dissipation as a parameter. the thermal capacity of a heatsink can be utilised to design a smaller heatsink for intermittent operation than would be necessary for the same level of continuous power dissipation. Such heatsinks are used both in the form of separate plates and as the equipment chassis itself.

The conditions to which the nomogram applies are as follows: 570 . 7). The appropriately-sized heatsink is found as follows.whether flat or extruded (finned) can be derived from the nomogram in Fig. Consequently. 6 Heatsink nomogram 1 Heatsink dimensions The maximum thermal resistance through which sufficient power can be dissipated without damaging the transistor can be calculated as discussed previously. particularly where the thermal conditions are critical. 6. not affected by nearby heatsinks). • ambient temperature about 25˚C. 8). Move vertically upwards to intersect the appropriate power dissipation curve in section 2. Enter the nomogram from the right hand side of section 1 at the appropriate Rth h-amb value (see Fig. Move horizontally to the left. • natural air cooling (unimpeded natural convection with no build up of heat). measured about 50mm below the lower edge of the heatsink (see Fig. This section explains how to arrive at a type and size of heatsink that gives a sufficiently low thermal resistance. it is advisable to take temperature measurements in the finished equipment. The actual values for the heatsink thermal resistance may differ by up to 10% from the nomogram values. • single mounting (that is. 2. 7). • atmospheric pressure about 10 N/m2. the nomogram does not give exact values for Rth h-amb as a function of the dimensions since the practical conditions always deviate to some extent from those under which the nomogram was drawn up. • distance between the bottom of the heatsink and the base of a draught-free space about 100mm (see Fig. • transistor mounted roughly in the centre of the heatsink (this is not so important for finned heatsinks because of the good thermal conduction). 1. Like all heatsink diagrams.Thermal Management Power Semiconductor Applications Philips Semiconductors Length of Extruded Heatsink (mm) 10 100 1000 1mm FLAT PLATE 1W 2W 5W 10W 20W 50W 100W Bright Horizontal Bright Vertical Blackened Horizontal Blackened Vertical 30D EXTRUDED 2mm 3mm 40D 5 10 Area of one side 2 mm 100 4 10 SOT93 TO220 SOT82 10 3 10 Fig. until the appropriate curve for orientation and surface finish is reached. Natural air cooling The required size of aluminium heatsinks .

Fig. If a flat-plate heatsink is to be used. 7 Conditions applicable to nomogram in Fig. move vertically upwards to obtain its length (Figs. 9a and 9b give the outlines of the extrusions). If the maximum ambient temperature is less than 25˚C.6 ˚C/W at the following powers: (a)Ptot(av) = 5W (b)Ptot(av) = 50W Enter the nomogram at the appropriate value of the thermal resistance in section 1. any Fig.25:1. 6 An extruded heatsink mounted vertically and with a painted surface is required to have a maximum thermal resistance of Rth h-amb = 2. the greater the power dissipation. 7. Case (b) requires a shorter length since the temperature difference is ten times greater than in case (a). Move horizontally to the left to obtain heatsink area. This is illustrated by the following example. 8 Use of the heatsink nomogram 571 . the smaller is the required size of heatsink. so does the temperature of the heatsink and thus the thermal resistance (at constant power) decreases owing to the increasing role of radiation in the heat removal process. and via either the 50W or 5W line in section 2. The heatsink dimensions should not exceed the ratio of 1. As the ambient temperature increases beyond 25˚C. Thus. a heatsink with dimensions derived from Fig. Move horizontally to the left into section 3 for the desired thickness of a flat-plate heatsink. Consequently. If an extruded heatsink is required. 4.Thermal Management Power Semiconductor Applications Philips Semiconductors 3. or the type of extrusion. move vertically downwards to intersect the appropriate curve for envelope type in section 4. However. at a constant value of the heatsink thermal resistance. 5. 6. 9a Outline of Extrusion 30D Th Fig. 6 at Tamb > 25˚C will be more than adequate. the appropriate lengths of the extruded heatsink 30D are found to be: (a) length = 110mm and (b) length = 44mm. Example Fig. then the thermal resistance will increase slightly. 9b Outline of Extrusion 40D Approx 50mm Approx 100mm Tamb Tamb The curves in section 2 take account of the non linear nature of the relationship between the temperature drop across the heatsink and the power dissipation loss.

For heatsinks with relatively small areas. of these areas or lengths.5m/s). 12. the thermal resistance hardly depends on the power dissipation and the orientation of the heatsink. too. The effect of forced air cooling in the case of flat heatsinks is seen from Fig. If a flat heatsink is used. 10 Arrangement of two equally loaded transistors mounted on a common heatsink If several transistors are mounted on a common heatsink. It also shows the reduction in thermal resistance or length of heatsink which may be obtained with forced air cooling.5 0 0 50 100 150 200 Length (mm) 250 300 350 Fig. Apart from the size of the heatsink. Note that turbulence in the air current can result in practical values deviating from theoretical values. that is. Provided that the cooling air flows parallel to the fins and with sufficient speed (>0. even when the length is small. provided that the air flow is sufficiently fast.Thermal Management Power Semiconductor Applications Philips Semiconductors increase will lie within the limits of accuracy of the nomogram and within the limits set by other uncertainties associated with heatsink calculations. The area of extruded heatsinks is always large with respect to the surface of the transistor case. or the heatsink not too large.5 Blackened 2 Natural Convection 1. From the sum Rth h-amb (K/W) 2. Fig. each transistor should be associated with a particular section of the heatsink (either an area or length according to type) whose maximum thermal resistance is calculated from equations 5 or 6. Forced air cooling If the thermal resistance needs to be much less than 1˚C/W. the thermal resistance now only depends on the speed of the cooling air. without taking the heat produced by nearby transistors into account. the size of the common heatsink can then be obtained. the dissipated power and the orientation of the heatsink have only a slight effect on the thermal resistance. forced air cooling by means of fans can be provided. Fig. Here.5 (Vertical) P=3W 1 Forced Cooling P=10W P=30W P=100W 1m/s 2m/s 5m/s 0. a considerable part of the heat is dissipated from the transistor case. 11 Thermal Resistance of a finned heatsink (type 40D) as a function of the length with natural and forced air cooling 572 . 11 shows the form in which the thermal resistances for forced air cooling are given in the case of extruded heatsinks. the transistors are best arranged as shown in Fig. The maximum mounting base temperatures of transistors in such a grouping should always be checked once the equipment has been constructed. This is why the curves in section 4 tend to flatten out with decreasing heatsink area. 10.

and thus temperatures should always be checked in the finished equipment. (a) blackened (b) bright Summary The majority of power transistors require heatsinking. 12 Thermal Resistances of heatsinks (2mm thick copper or 3mm thick aluminium) under natural convection and forced cooling conditions.Thermal Management Power Semiconductor Applications Philips Semiconductors Fig. or one with a very low thermal resistance. in which case forced air cooling by means of fans should be provided. a heatsink of appropriate type and size can be chosen. some applications require a small heatsink. Finally. and once the maximum thermal resistance that will maintain the device’s junction temperature below its rating has been calculated. with a SOT93 envelope. 573 . The practical conditions under which a transistor will be operated are likely to differ from the theoretical considerations used to determine the required heatsink.

Buethker L.R.Gilliam D.Hammerton D.Burley It was revised and updated.J.Gant J.Ham C.Misdom P.Brown R.D.Bennett M.Harper W.Verhees F.Burley G.Sharples H. de Ruiter D.Ham C.van de Wouw This book was originally prepared by the Power Semiconductor Applications Laboratory. The authors also thank Mr.Hammerton D.1.G.Mulder E. Contributing Authors N.J.Miller H.Moody S.J.Sharples .J.J.Rosink D.Fry R. Hazel Grove: M.Oosterling N.H.F. We would also like to thank Dr.J.Grant N.Houldsworth M. of the Philips Semiconductors product division.P.B.A. Nijhof J.Preface Power Semiconductor Applications Philips Semiconductors Acknowledgments We are grateful for all the contributions from our colleagues within Philips and to the Application Laboratories in Eindhoven and Hamburg.Mellor R.Tebb H.A. The authors thank Mrs.M.C.5.B.Stork D.Simons T.Hayes for her considerable help in the preparation of this book.Humphreys C.Bennion D.v.Mellor of the University of Sheffield for contributing the application note of section 3.d. by: N.J.Hettersheid J.Hooff J. in 1994.Woodworth T.P.J.Miller L.H.Haslam for his assistance in the formatting and printing of the manuscripts.Humphreys P.Brown C.Pichowicz W.Hammerton D.

CHAPTER 2 is devoted to Switched Mode Power Supplies. Hazel Grove. Part of this chapter is devoted to worked examples showing how junction temperatures can be calculated to ensure the limits are not exceeded. CHAPTER 8 is an introduction to the use of high voltage bipolar transistors in electronic lighting ballasts. Specific design examples are given as well as a look at designing the magnetic components. CHAPTER 4 looks at television and monitor applications. The book is intended as a guide to using power semiconductors both efficiently and reliably in power conversion applications. dc and stepper motor operation and control. CHAPTER 6 reviews thyristor and triac applications from the basics of device technology and operation to the simple design rules which should be followed to achieve maximum reliability. Power MOSFETs and High Voltage Bipolar Transistors. Deflection and power supply circuit examples are also given based on circuits designed by the Product Concept and Application Laboratories (Eindhoven). Many of the possible topologies are described. A description of the operation of horizontal deflection circuits is given followed by transistor selection guides for both deflection and power supply applications. It begins with a basic description of the most commonly used topologies and discusses the major issues surrounding the use of power semiconductors including rectifiers. Heatsink requirements and designs are also discussed in the second half of this chapter. .Preface Power Semiconductor Applications Philips Semiconductors Preface This book was prepared by the Power Semiconductor Applications Laboratory of the Philips Semiconductors product division. CHAPTER 7 looks at the thermal considerations for power semiconductors in terms of power dissipation and junction temperature limits. This chapter looks only at transistor controls. It is made up of eight main chapters each of which contains a number of application notes aimed at making it easier to select and use power semiconductors. CHAPTER 5 concentrates on automotive electronics looking in detail at the requirements for the electronic switches taking into consideration the harsh environment in which they must operate. phase control using thyristors and triacs is discussed separately in chapter 6. The end of this chapter describes resonant power supply technology. CHAPTER 3 describes motion control in terms of ac. Specific examples are given in this chapter for a number of the common applications. CHAPTER 1 forms an introduction to power semiconductors concentrating particularly on the two major power transistor technologies.

.. Understanding the Data Sheet: PowerMOS ........................... Logic Level FETS .................................................................2.3..........8 1....... Effects of Base Drive on Switching Times ..1................................................2..........2................................................................................ Electrostatic Discharge (ESD) Considerations ..................................2.1........2..........3..........3.............2..2......3..................1 An Introduction To Power Devices ....................... Using High Voltage Bipolar Transistors ....................................6 1.. Understanding The Data Sheet: High Voltage Transistors ...................2.7 1....................... 17 19 29 39 49 53 57 61 67 69 77 79 83 91 97 High Voltage Bipolar Transistor 1........1 1............... 103 105 107 129 141 153 159 Output Rectification 2............................................1 1.................................................... 173 2....................................................3 Base Circuit Design for High Voltage Bipolar Transistors in Power Converters .......................2..................1.. 179 i ............................................... Avalanche Ruggedness ........ Parallel Operation of Power MOSFETs .............................9 PowerMOS Introduction .............................. Understanding Power MOSFET Switching Behaviour ... 2....5 1... Power MOSFET Drive Circuits ..............2.....................1 An Introduction to Switched Mode Power Supply Topologies ................3 An Introduction to Synchronous Rectifier Circuits using PowerMOS Transistors ...............4 Isolated Power Semiconductors for High Frequency Power Supply Applications ....................2 1.......1 Fast Recovery Epitaxial Diodes for use in High Frequency Rectification 161 2....................................4 Introduction To High Voltage Bipolar Transistors .......1...2...2 The Power Supply Designer’s Guide to High Voltage Transistors .1.........................Contents Power Semiconductor Applications Philips Semiconductors Table of Contents CHAPTER 1 Introduction to Power Semiconductors General 1 3 1....... CHAPTER 2 Switched Mode Power Supplies Using Power Semiconductors in Switched Mode Topologies 2................................... 2............... 2..........................2 1......... 5 Power MOSFET 1.......... Series Operation of Power MOSFETs ....................3 1................................3 1................................2...2 Schottky Diodes from Philips Semiconductors ......4 1................................

.. 207 Resonant Power Supplies 217 2.1..........2..................4 A Designers Guide to PowerMOS Devices for Motor Control .........................3 Brushless DC Motor Systems ........................... Motor Control: Introduction to a 20 kHz System ..1... 219 2................ Resonant Power Supply Converters ......2............... 3..................... 331 ii .........3 MOSFETs and FREDFETs for Motor Drive Equipment .......................The Solution For Mains Pollution Problems ....2 The Effect of a MOSFET’s Peak to Average Current Rating on Invertor Efficiency ..1 Noiseless A..............................................................2............. An Introduction To Resonant Power Supplies .............1 Mains Input 100 W Forward Converter SMPS: MOSFET and Bipolar Transistor Solutions featuring ETD Cores ............. 187 2.1 An Introduction to Horizontal Deflection ........... 241 243 245 251 253 259 273 283 DC Motor Control 3....... 40A High Frequency Inverter Pole Using Paralleled FREDFET Modules .1................................. 199 Magnetics Design 205 2.............1................ 301 Stepper Motor Control 307 3.3...........2 The BU25XXA/D Range of Deflection Transistors ............................C.................5........ 285 3...................1......2 A switched-mode controller for DC motors .........................................................................3.......................1 Chopper circuits for DC motor control .............................5 A 300V...................................................................... Low Cost..1 Stepper Motor Control .....1........... 3...........1. 321 4..........Contents Power Semiconductor Applications Philips Semiconductors Design Examples 185 2.................... 309 CHAPTER 4 Televisions and Monitors Power Devices in TV & Monitor Applications (including selection guides) 317 319 4............................................................................. Self-Oscillating Power Supply using an ETD34 Two-Part Coil Former and 3C85 Ferrite . 3.....................3...................1.............................................................. 293 3.......................................... 3.......................4... 225 CHAPTER 3 Motor Control AC Motor Control 3......5............................................................2 Flexible.....1 Improved Ferrite Materials and Core Outlines for High Frequency Power Supplies ...2............

...................1 Automotive Lamp Control With Philips MOSFETS .............10 3 pin and 5 pin TOPFET Leadforms ....1 Application Information for the 16 kHz Black Line Picture Tubes .....3....................................3.................9 Isolated Drive for TOPFET ................................ 5...............................................3 Philips HVT’s for TV & Monitor Applications .......................14 Driving DC Motors with TOPFET ........................................... 379 4.............. 389 Monitor Deflection Circuit Example 397 4...................3..........a Microcontroller compatible TOPFET .......................3................... 5...........................1 A 70W Full Performance TV SMPS Using The TDA8380 ...................2 32 kHz / 100 Hz Deflection Circuits for the 66FS Black Line Picture Tube 361 SMPS Circuit Examples 377 4.1 An Introduction to the 3 pin TOPFET ..............3.......................3 BUK101-50DL .................. 435 The TOPFET 5... 5....................... 5....2.........................2 A Synchronous 200W SMPS for 16 and 32 kHz TV ......3...............................7 Linear Control with TOPFET .........................2 An Introduction to the 5 pin TOPFET ............................1 A Versatile 30 ....4 TV and Monitor Damper Diodes .........2. 5......................3................ iii 443 445 447 449 451 453 455 457 459 461 463 465 467 469 471 473 475 .........................5 Driving TOPFETs . 399 CHAPTER 5 Automotive Power Electronics Automotive Motor Control (including selection guides) 421 423 5....................................3............................................................ 5........64 kHz Autosync Monitor ... 339 4.3.3................6 High Side PWM Lamp Dimmer using TOPFET ..................................4..............3.................... 5...............................1............................3...................................... 5................1.2.4 Protection with 5 pin TOPFETs .......16 High Side Linear Drive with TOPFET ...8 PWM Control with TOPFET .............. 351 4.3..................................... 345 TV Deflection Circuit Examples 349 4............3............................................1.........3........ 5..12 Negative Input and TOPFET ...................11 TOPFET Input Voltage .................1 Automotive Motor Control With Philips MOSFETS ... 5................... 425 Automotive Lamp Control (including selection guides) 433 5.....13 Switching Inductive Loads with TOPFET .................. 5. 5..............................15 An Introduction to the High Side TOPFET .....3........3............. 5........................ 5..3........................ 5.......................Contents Power Semiconductor Applications Philips Semiconductors 4..........

..1 An Introduction to Electronic Automotive Ignition .... The Peak Current Handling Capability of Thyristors ................1 Understanding Hi-Com Triacs ...................4....1..............4...........................2 Domestic Power Control with Triacs and Thyristors ..........................Contents Power Semiconductor Applications Philips Semiconductors Automotive Ignition 477 5................. 479 5.................. 485 487 489 497 505 509 521 Thyristor and Triac Applications 6.........................1 Thermal Considerations for Power Semiconductors ......................................1.............1...................... 527 6.....................3 Design of a Time Proportional Temperature Controller .........................................4.......................................... 567 CHAPTER 8 Lighting Fluorescent Lamp Control 575 577 8..............2 Heat Dissipation ..2 Electronic Ballasts ..........................3 6..............1 Efficient Fluorescent Lighting using Electronic Ballasts .................................Philips Transistor Selection Guide ....................2 Using Hi-Com Triacs .........2..................... 589 iv ......................... 557 7...............................1. Understanding Thyristor and Triac Data ......Base Drive Optimisation .......2.....3......2 IGBTs for Automotive Ignition ......................................1 Triac Control of DC Inductive Loads ...3 An Electronic Ballast ..........1..........3...........1..2 6........2.......................... 549 6.................... 537 Hi-Com Triacs 547 6................3 Electronic Switches for Automotive Ignition ............. 481 5..........................1......................................... 579 8..1............................. 483 CHAPTER 6 Power Control with Thyristors and Triacs Using Thyristors and Triacs 6......... 551 CHAPTER 7 Thermal Management Thermal Considerations 553 555 7... Using Thyristors and Triacs ...1 6...1................................. 587 8....4 Introduction to Thyristors and Triacs . 523 6.............

490 i Bridge circuits see Motor Control . 473 Avalanche. 551 thyristor. 492 Breakdown voltage. 549 Choke fluorescent lamp. 303 Breakback voltage diac. 138. 117 Clamp winding. 523. 347 outlines. transformer core. 133 triac commutation. 589 Current Mode Control. 492 Breakover voltage diac. 492. 109 continuous mode.109 Burst firing. 120 Current tail. 459. 589 forward converter. 483 ignition. 144. 148 base resistor. 190 Ballast electronic. 61 Avalanche breakdown thyristor. 148 Boost converter. 490 Avalanche multiplication. 572 natural. 348 losses. 164 Hi-Com triac. 109 discontinuous mode. 345 Darlington. 70 Breakover current diac. 285 Clamp diode. 481. 442 reverse battery. 590 Asynchronous. 580 fluorescent lamp. 134 Baker clamp. 481. 108 . 579 switchstart. 537 Charge carriers. 29 Capacitor mains dropper. 143 Damper Diodes. 328. 117 Current fed resonant inverter. 475 resistive loads. 328. 442 seat heater. 345 picture distortion. 69 . 592 thyristor. 564 Capacitance junction. 13 Data Sheets High Voltage Bipolar Transistor. 134 Cross regulation. diode assisted. 497 Automotive fans see motor control IGBT. 92. 537 Cooling forced. 529 Compact fluorescent lamp. 113 Commutation diode. 479. 537 Burst pulses. 187. 442 solenoids. 579 Base drive.AC Brushless motor. 529 Critical electric field. 457. 187 power converters. 109 Bootstrap. 455 motor control. 471. 494. 303 Buck-boost converter. 143 Base inductor. 145 drive transformer leakage inductance. 435. 149 electronic ballast. diode assisted. 147 base inductor. 113 Anti saturation diode. 138. 114. 580 Choppers. 146 drive transformer. 110 Buck converter. 425. 147 Base inductor. 483 lamps. 585 Continuous mode see Switched Mode Power Supplies Continuous operation. 557 Converter (dc-dc) switched mode power supply. 570 Crest factor. 109 output ripple. 473. 469 TOPFET.331 MOSFET.Index Power Semiconductor Applications Philips Semiconductors Index Airgap. 348 selection guide. 141 speed-up capacitor. 301.97. 492 triac. 107 Cookers. 452. 111. 136 base inductor. 345. 544 CENELEC. 479 screen heater. 367 forward recovery.

110. 113 cross regulation. 327. 111. 115 Flyback converter (two transistor). 113 rectifier circuit. 580 base drive optimisation. 113 advantages. 367 Disc drives. 67 ETD core see magnetics F-pack see isolated package Fall time. 114 diodes. 169 reverse recovery. 527 Dropper capacitive. 114 coupled inductor. 171 snap-off. 167 stored charge. 116 advantages. 587. 473 Emitter shorting triac. 584. 67 see Protection. 117 diodes. 162 operating frequency. 589 current fed half bridge. 579 colour rendering. 530. 116 clamp diode. 111. 589 current fed push pull. 164 reverse recovery softness. 549 Epitaxial diode. 580 Fluorescent lamp. 162 technology. 86 Baker clamp. 111. 213 operation. 117 duty ratio. 162 ii ESD. 114 clamp winding. 116 magnetics. 6 double diffused. 162 epitaxial. 531 Diac. 161 schottky. 199 synchronous rectifier. 138 dI/dt triac. 114 electronic ballast. 582 transistor selection guide. 544 resistive. 115 disadvantages. 582 leakage inductance. 579 efficacy. 116. 117 ferrite cores. 133 Desaturation networks. 180 self oscillating power supply. 162 reverse leakage. 455 see RFI. 588 voltage fed push pull. 113 continuous mode. 579 Flyback converter. 583. 167 selection guide. 111. 583. 117 . 531 Electronic ballast. ESD TOPFET. 197 continuous mode. 587 EMC. 492.Index Power Semiconductor Applications Philips Semiconductors dc-dc converter. 167 softness factor. 162. 164 forward recovery. 587 flyback. 144 Fast Recovery Epitaxial Diode (FRED) see epitaxial diode FBSOA. 561 EFD core see magnetics Efficiency Diodes see Damper Diodes Electric drill. 580 triphosphor. 114 discontinuous mode. 181 transformer core airgap. 531 Forward converter. 591 Diode. 500. 117 conduction loss. 302 Discontinuous mode see Switched Mode Power Supplies Domestic Appliances. 213 magnetisation energy. 161 characteristics. 587 voltage fed half bridge. 260. 156. 134 Ferrites see magnetics Flicker fluorescent lamp. 118 disadvantages. 91. 163 dI/dt. 168 lifetime control. 544. 165 passivation. 584. 116 core loss. 173 structure. 116 core saturation. 579. 579 colour temperature. 114 Food mixer. 117 cross regulation. 143. 113 transistors. 545 Duty cycle. 527. 113 magnetics. 161 Diode Modulator. 119 Depletion region. ESD precautions.

552 inductive load control. 149 dV/dt. 136 current limiting values. 150 conductivity modulation. 122 cross conduction. 589 flux symmetry. 254 FREDFETs motor control. 124 disadvantages. 138 base-emitter breakdown. 86. 569 Heatsink compound. 125 advantages. 253. 135. 538 Gate drive forward converter. 584. 125 transistors. 79. 551 commutation. 333 avalanche breakdown. 162. 436 TOPFET. 122 Heat dissipation. 169 GTO. 139 operation. 8. 131 avalanche multiplication. 44. 122 clamp diodes. 333. 331 depletion region. 195 switching losses. 138. 139 drive transformer. 132 current tail. 111. 321. 148 base resistor. 117 Forward recovery.AC Half bridge converter. 135. 341 ‘bathtub’ curves. 141. 111. 134 Baker clamp. 254 diode. 86 horizontal deflection. 150 critical electric field. 581. 146 breakdown voltage. 567 iii Heat sink compound. 334 fall time. 116 output diodes. 157. 136.Index Power Semiconductor Applications Philips Semiconductors operation. 587. 181 transistors. 342 Miller capacitance. 122 diodes. 144. 92. 134 hard turn-off. 122 transistors. 133 desaturation. 473 High Voltage Bipolar Transistor. 117 output ripple. 98 limiting values. 124 voltage doubling. 589 Fact Sheets. 254 drive. 116 rectifier circuit. 430. 97. 552 gate trigger current. 143. 122 advantages. 331. 262 loss. 195 Gold doping. 91. 187 switching frequency. 86. 250. 259 Full bridge converter. 99. 138. 305 bridge circuit. 145 base inductor. 122 synchronous rectifier. 92. 92 carrier concentration. 168 FREDFET. 144 base drive. 133 electronic ballast. 341 leakage current. 139 electric field. diode assisted. 91. 118 Forward converter (two transistor). 122 electronic ballast. 92. 126 disadvantages. 255 charge. 549. 145 drive transformer leakage inductance. 253 Half bridge circuits see also Motor Control . 346 data sheet. 122 magnetics. 97 losses. 79. 111. 83. 551 dIcom/dt. 180 reset winding. 96. 151 carrier injection. 117 switched mode power supply. 544 Heaters. 88. 126 Gate triac. 125 operation. 144 FBSOA. 150 . 147 base inductor. 385 base drive circuit. 585. 196 synchronous rectifier. 587. 79 dI/dt. 86. 551 High side switch MOSFET. 91 device construction. 336. 92. 214 operation. 537 Heatsink. 11 Guard ring schottky diode. 174 Half bridge. 519. 256 reverse recovery. 134 current crowding. 143 current tails. 125 diodes. 86. 91 d-type. 157 transistor voltage. 567 Heater controller. 514 Hi-Com triac. 99.

483 darlington. 562 rectangular pulse. 321. 133 speed-up capacitor. 341 secondary breakdown. 148 reverse recovery. 352. 530 Logic Level FET motor control. 323. 483 clamped. 134 thermal runaway. 400 TDA8433. 321. 136. 98 over voltage. 273 see motor control ac current fed. 561 Lamp dimmer. 137. 365. 53 Induction motor see Motor Control . 481. 557. 92. 92. 347 s-correction. 154 stray capacitance. 561 rectangular pulse. 142. 207 . 130 Horizontal Deflection. 162 Lighting fluorescent. 346 over current. 151 safe operating area. 332. 482. 154 Isolation. 131 power limiting value. 404 TDA2595. 138. 579 phase control. 91. 438 MOSFET. 98. 435 dI/dt. 368 TDA4851. 135. 99. 210 ETD core. 99. 483 Induction heating. 53 switched mode power supply. 138. 207 100W 100kHz forward converter. 138. 347. 91. 92. 530 Lamps. 208 core materials. 199. 564 non-rectangular pulse. 481. 80 ratings. 438 inrush current. 530 Intrinsic silicon. 9 Junction temperature. 85. 138 passivation. 383 snubber. 191 50W flyback converter. 354 linearity correction. 200. 336 control ic. 86. 561 burst pulses. 133 smooth turn-off. 52. 133 Inverter. 321 transistors. 88. 342 technology. periodic. 432 Logic level MOSFET. 490 Leakage inductance. 455 switch rate. 325. 367 diode modulator. 352. 134 saturation. 80. 149 thermal breakdown. 305 automotive. 98. 88 voltage limiting values. 91. 88 outlines. single shot. 484 ignition. 86 SMPS. 364. 155 thermal resistance. 146.Index Power Semiconductor Applications Philips Semiconductors optimum drive. 199 core losses. 323 operating cycle. 97 RBSOA. electric. 129. 481. 152 turn-off. 139 space charge. 406 east-west correction. 479. 352. 150 saturation current. 339. 322 IGBT. 151 turn-on. 345. 83. 91. 470. 332. 341. 99. 352. 342 sub emitter resistance. 207 EFD core.AC Inductive load see Solenoid Inrush current. 528. 565 rectangular pulse. 537 Isolated package. 455 Latching current thyristor. 367 drive circuit. 141. 153 J-FET. 132 process technology. 435 PWM control. 523 Lifetime control. 149. 144. 97 overdrive. 139 RC network. 197 100W 50kHz forward converter. 86. 260. 93. 154. 150 underdrive. 143. 363. 143 storage time. 331. 85. 436 Magnetics. 401 d-type transistors. 367 base drive. 438 TOPFET. 92. 79. 327. 135 switching. 107 Irons. 369 test circuit. 94. 483 iv Ignition automotive. 11. 92. 367 line output transformer. 408 waveforms. 346 damper diodes. composite. 113.

57. 71 linear mode. 30.72 charge. 339. 24. 531 starting. 225 pre-converter. 250 snubber. 273 anti-parallel diode. 262 pulse width modulation. 57 data sheet. 250 duty ratio. 26. 19. 70 threshold voltage. 25. 245. 285. 285. 384 solenoid. 276 peak current. 44 inductive load. 250 dV/dt. 73. 250. 9. 249 diode. 155. 36 diode. 276 gate drives. 245 control. 267 MOSFET. 49. 264 drive circuit loss. 61.Index Power Semiconductor Applications Philips Semiconductors flyback converter. 57. 155. 23. 39. 259. 36. 71 package inductance. 69 dI/dt. 156 high side. 544 Mesa glass. 153. 69 motor control. 246 efficiency. 425 braking. 32. 265 parasitic oscillations. 26. 35. 305 loss. 57. 34 maximum current. 74 series operation. 156 driving. 211 push-pull converter. 265 MOS Controlled Thyristor. 250. 32. 293. 303 current rating. 74 thermal resistance. 260 filter. 250 carrier frequency. 436 high side drive. 246 short circuit. 245. 19 switching. 251 signal isolation. 259 Parallel MOSFETs. 58. 248 Motor Control . 24 characteristics. 253 drive. 34. 303 breakdown voltage. 251 phase voltage. 162 Metal Oxide Varistor (MOV). 213 half bridge converter. 70 . 253 bootstrap. 34. 537 Mains pollution. 72 turn-off. 21. 503 Miller capacitance. 528 Motor Control . 214 power density. 213 forward converter. parallelling. 62 structure. 39. 264 gate charge. 47. 156 capacitances. universal back EMF. 70 capacitance. 225 Mains transient. 264 ESD. 273 line voltage. 253 antiparallel diode. 194. 251 Resonant supply. 248 switching frequency. 52 logic level. 262 power factor. 215 Mains Flicker. 22. 256 Motor. 260 ripple. 260. 435 leakage current. 249 half bridge. 195 gate resistor. 213 switched mode power supply. 429 modelling. 67 gate-source protection. 70 transconductance. 295. 215 transformer construction. 262 loss. 250 FREDFET. 29. 246 three phase bridge. 245 inverter. 37. 62 lamps. 290. 262.AC. 196 synchronous rectifier. 265 on-resistance. 36 turn-on. 51 peak current rating. 259. 73 parallel operation. 195 gate drive. 53 SMPS. 236. 139 Modelling. 73 v safe operating area. 179 thermal impedance. 301 control. 49. 276 speed control. 187 switching frequency. 21. 13 MOSFET. 262 dc link.DC. 262 EMC. 53 reverse diode. 299 brushless. 248 current rating. 246 underlap. 73 ruggedness. 288 . 262 switching loss. 72. 261 diode recovery.

286 full bridge. 111. 293. 162 PCB Design. 537 Ruggedness MOSFET. 119 transistor voltage. 236 MOSFET. 53 pre-converter. 312 permanent magnet. 285 PWM. 558 Pulse Width Modulation (PWM). 135. 293. 557 Mounting torque. 119 clamp diodes. 173 Safe Operating Area (SOA). 73 schottky diode. 162 RBSOA. 93. 514 Parasitic oscillation. 459. 473. 287. 537 Protection ESD. 446. 431 TOPFET. 305 inrush. 557 forward biased.Stepper. 582. 525 voltage rating. 498 triac. 135. 287 freewheel diode. 119 duty ratio. 119 transistors. 121 Qs (stored charge). 294 triac. 139 . 446. 119 transformer. 119 output filter. 99. 119 cross conduction. 546 thyristors and triacs. 529. 446. 225 modelling. 309 unipolar. 419 Phase angle. 429 motor current. 445. 149 Passivation. 447. 167. 119 output ripple. 430 inverter. 497. synchronous. 99. 457. 108 Push-pull converter.ac Power dissipation. 213 multiple outputs. 134. 430 permanent magnet. 479 short circuit. 457. 286 torque. 430. 500 Phase control. 25. 309 bipolar. 286. 451 Pulse operation. 301 permanent magnet motor. 287 high side switch. boost converted. 119. 119 rectifier circuit. 138. 92. 475 logic level FET. 286 efficiency. 448 temperature. 121 disadvantages. 302 linear. 429 IGBT. 120 diodes. 131. 471 servo. 288 Motor Control . 303 duty cycle. 154. 544. 447. 587 flux symmetry. 139 Rectification. 482 overvoltage. 295 overload. 452. 446. 448. 432 loss.Index Power Semiconductor Applications Philips Semiconductors drive. 138. 93. 143. transistor. 431 stall. 74. 293 FREDFET. 314 drive. 580 active. 158. 557 see High Voltage Bipolar Transistor loss. 396. 219. 475 topologies. 309 reluctance. 523 Phase voltage see motor control . 99. 225 Reverse leakage. 313 hybrid. 62. 288 MOSFET. 52. 545 Resonant power supply. 179 Reset winding. MOSFET loss Power factor correction. 251. 117 Resistor mains dropper. 581 vi Power MOSFET see MOSFET Proportional control. 287 half bridge. 180 switching frequency. 471 TOPFET. 393. 119 electronic ballast. 120 magnetics. 162 RFI. 530. 298 short circuit. 448. 285. 154 Mounting base temperature. 310 chopper. 119 operation. 368. 311 step angle. 469 reverse battery. 310 Mounting. 169 Reverse recovery. 119 advantages. 134 reverse biased. 119 current mode control. 459.

126 discontinuous mode. 174 edge leakage. 298 Single ended push-pull see half bridge converter Snap-off. 156 non-isolated. 384 pulse width modulation. 426 Schottky diodes. 113 base circuit design. 383. 113 core loss. 125 half bridge converter. 339 Self Oscillating Power Supply (SOPS) 50W microcomputer flyback converter. 339. 192 100W 500kHz half bridge converter. 173 selection guide. 469. 339.Index Power Semiconductor Applications Philips Semiconductors Saturable choke triac. 174 ruggedness. 144 Stored charge. 139. 166 diode reverse recovery softness. 120 dc-dc converter. 111 magnetics design. 108 opto-coupler. 392 isolated. 167 Solenoid TOPFET. 108. 118. 109. 391 control loop. 174 guard ring. 469. 591 self oscillating power supply. 154 SOT186A. 153 100W 50kHz bipolar forward converter. 441 MOSFETs driving motors. 523. 343 MOSFETs driving heaters. 531 triac. 126. 111. 442 MOSFETs driving lamps. 115. 473 Solid state relay. 176 technology. 119 diode loss. 113 isolated packages. 33. 108 core excitation. 137. 309 Stepup converter. 173 bulk leakage. 390 MOSFET. 93. 122 high voltage bipolar transistor. 11 Stepdown converter. 136. 279 Softness factor. 527 Starter fluorescent lamp. 94. 115. 191. 109 buck-boost converter. 121. 523 Schottky diode. 136 power-up. 379 full bridge converter. 392 output rectifiers. 199 Servo. 549 active. 111 flyback converter. 109 Stepper motor. 111. 143 Speed control thyristor. 109 vii Storage time. 174 reverse leakage. 111. 389 asymmetrical. 379 epitaxial diodes. 108 ceramic output filter. 92. 153. 166 diode reverse recovery effects. 153 continuous mode. 121. 379 control ic. 153 isolation. 129. 154 Space charge. 112. 580 Startup circuit electronic ballast. 111. 167 diodes. 111. 149 boost converter. 529. 118. 502. 149 power-down. 201 Static Induction Thyristor. 112. 124. 199 ETD transformer. 187 16 & 32 kHz TV. 108 push-pull converter. 111. 167 current mode control. 173 SCR see Thyristor Secondary breakdown. 109. 124. 123 forward converter. 163 parasitic oscillation. 162 Suppression mains transient. 331 damper diodes. 167 Snubber. 153. 176 SMPS. 544 Switched Mode Power Supply (SMPS) see also self oscillating power supply 100W 100kHz MOSFET forward converter. 112. 113 mains filter. 116. 133 Speed-up capacitor. 154 SOT199. 139 power MOSFET. 119 . 384 multiple output. 331 BU25XXD. 111. 133 Selection Guides BU25XXA. 345 EPI diodes. 380 mains input. 501 SOT186. 197 magnetisation energy. 113. 171 horizontal deflection. 495. 473 turn off. 110 buck converter. 161 flux swing.

490 gate power. 498. 112 transistor mounting. 509 ’two transistor’ model. 121. 497 avalanche breakdown. 111 transformer. 493 reverse characteristic. 511 dI/dt. 154. 505 phase angle. 111 transformer saturation. 500 gate circuits. 497 self commutation. 399 two transistor flyback. 126 turns ratio. 492 gate specifications. 124. 111 TV & Monitors. 139 resonant see resonant power supply RFI. 70. 527 pulsed gate. 500 inrush current. 500 resistive loads. 382 standby supply. 107 topology output powers. periodic. 381. 490 load line. 391 stepdown. 122 synchronisation. 489 reverse recovery. 108 reliability. 118. 503 peak current. 490 applications. 139. 558. 138 standby. 514 operation. 493 RFI. 565 pulse operation. 392 start-up. 111. 498 resonant circuit. 493 series choke. 531 static switching. 115. 109 symmetrical. 173 snubber. composite. 491 dV/dt. 502 snubber. 156. 109 stepup. 505 external commutation. 562 rectangular pulse. 512 gate triggering. 568 Thermal resistance. 154. 568 non-rectangular pulse. 489 . 568 Thyristor. 499 fusing I2t. 492 gate requirements. 112 transistors. 509 leakage current. 112 regulation. 502 speed controller. 119. 561 rectangular pulse. 501 commutation. 490 half wave control. 490 dIf/dt. 489 switching. 158. 381. 509 inductive loads. 537 Thermal continuous operation. 497 structure. 136 transistor voltage limiting value. 117 Switching loss. 492 control. 490 overcurrent. 74. 509 cascading. 392 rectification efficiency. 379. 111. 557 Thermal time constant. 179 self driven. 497 current rating. 500 gate current. 230 Synchronous. 490 energy handling. 154. 568 intermittent operation. 503. 527 asynchronous control. 490 breakover voltage. 561 Thermal capacity. 512 gate cathode resistor. 493 full wave control. 93. 490. 138 transistor turn-on. 111. 557. 383 soft start. 156. 112. 163 rectifier selection. single shot. 114 two transistor forward. 497 Synchronous rectification. 382 synchronous rectification. 167 schottky diode. 391 topologies. 561 single shot operation. 391 transistor current limiting value. 503 latching current. 558 rectangular pulse. 154 transistor selection. 10. 490. 181 transformer driven. 339.Index Power Semiconductor Applications Philips Semiconductors RBSOA failure. 112 transistor turn-off. 557 Thermal impedance. 568 viii Thermal characteristics power semiconductors. 492 mounting. 179 TDA8380. 499 holding current. 490. 500 phase control. 139 rectification. 497. 180 Temperature control. 138 transformers.

154 Triac. 451. 447. 497 transformer load. 549 dIcom/dt. 379. 544 pulse triggering. 451. 409. 492. 489 switching. 469 Transformer triac controlled. 503. 402 Two transistor flyback converter. 465. 489 synchronous control. 510. 475 lamps. 503 zero crossing. 494. 497 breakover voltage. 461 5 pin. 510 operation. 114 Two transistor forward converter. 540. 500 gate circuits. 499 fusing I2t. 457. 518. 518 400Hz operation. 497 saturable choke. 510 Thyristor data. 368. 497. 494 commutating dV/dt. 523 dc motor control. 549 speed controller. 455 leadforms. 509 Time proportional control. 537 asynchronous control. 523. 537 TOPFET 3 pin. 523 Transformer core airgap. 497 transient protection. 529. 498. 549. 111. 467 protection. 399 vertical deflection. 545 gate sensitivity. 498 RFI. 500 gate current. 537 Trigger angle. 494 commutation. 111. 459. 453. 111. 538 holding current. 559 Transient thermal response. 492 pulsed gate. 339. 339. 449. 523 dV/dt. 341 picture distortion. 475 high side. 491. 497 dc inductive load. 500 phase control. 491 ix gate requirements. 503 isolated trigger. 445. 352 . 512 gate cathode resistor. 463 linear control. 527. 451. 525 dI/dt. 449. 500 turn-off time. 361 damper diodes. 501 latching current. 354. 491. 549 control. 410 high voltage bipolar transistor. 523 series choke. 345. 351 30-64 kHz autosync. 451. 389. 447. 495. 399 32 kHz black line. 529. 523 transient protection. 500 TV & Monitors 16 kHz black line. 491 overcurrent. 546 protection. 489. 457. 459 negative input. 500 inrush current. 473. 510 resistive loads. 117 Universal motor back EMF.Index Power Semiconductor Applications Philips Semiconductors switching characteristics. 502 varistor. 500 quadrants. 509 turn-on dI/dt. 512 thermal specifications. 502 snubber. 497 temperature rating. 473 PWM control. 510 Hi-Com. 503 phase angle. 341 horizontal deflection. 503 voltage rating. 467. 491 gate triggering. 461. 455. 550 turn-on dI/dt. 549 commutating dI/dt. 502. 531 . 497 structure. 367 diode modulator. 494 turn-on. 445. 527 static switching. 456. 469. 463 driving. 492 gate resistor. 502 trigger angle. 327. 339 SMPS. 348 power MOSFET. 367 EHT. 502 trigger angle. 364. 500 triggering. 113 Transformers see magnetics Transient thermal impedance. 549 emitter shorting. 517 synchronous control. 510 charge carriers. 549 full wave control. 551 inductive loads. 512 time proportional control. 430. 358. 518 applications. 531. 502 varistor.354. 459 solenoids. 527. 490. 491. 457 motor control. 523. 465. 523.

528 Vacuum cleaner. 503 Vertical Deflection. 364. 537 Zero voltage switching. 537 Zero crossing. 402 Voltage doubling. 358. 527 Varistor. 122 Water heaters.Index Power Semiconductor Applications Philips Semiconductors starting. 537 x .

215 Mains Flicker. 71 package inductance. 301 control. 73 parallel operation. 30. 179 thermal impedance. 429 modelling. 155. 52 logic level. 139 Modelling. 250. 37. 531 starting. 245. 70 transconductance. 285. 58. 51 peak current rating. 250 carrier frequency. 156 driving. 73 v safe operating area. 69 motor control. 264 drive circuit loss. 25. 250 FREDFET. 32. 74 thermal resistance. 26. 537 Mains pollution. 262. 276 gate drives. 213 half bridge converter. 36. 264 ESD. 57. 73 ruggedness. 9. 13 MOSFET. 303 breakdown voltage. 250 snubber. 73. 253 antiparallel diode. 248 switching frequency. 213 switched mode power supply. 249 half bridge. 195 gate drive. 339. 62 lamps. 265 parasitic oscillations. 295. 196 synchronous rectifier. 251 Resonant supply. 253 bootstrap. 61. 24 characteristics. 53 SMPS. 19. 70 capacitance. 246 short circuit. 245. 21. 259. 225 Mains transient. 70 threshold voltage. 288 . 262 switching loss. 425 braking.DC. 290. 26. 265 on-resistance. 236. 259 Parallel MOSFETs. 71 linear mode. 72. 260 filter. 262 loss. 276 speed control. 250 duty ratio.AC. universal back EMF. 34. 214 power density. 39. 267 MOSFET. 276 peak current. 260 ripple. 23. 436 high side drive.72 charge. 246 underlap. 285. 53 reverse diode. 544 Mesa glass. 213 forward converter. 22. 384 solenoid. 245 control. 195 gate resistor. 156 high side. 187 switching frequency. 299 brushless. 303 current rating.Index Power Semiconductor Applications Philips Semiconductors flyback converter. 246 three phase bridge. 194. 162 Metal Oxide Varistor (MOV). 293. 72 turn-off. 69 dI/dt. 44 inductive load. 256 Motor. 251 phase voltage. 250. 36 diode. 57. 251 signal isolation. 246 efficiency. 57 data sheet. 34 maximum current. 70 . 503 Miller capacitance. 273 line voltage. 24. 262 power factor. 528 Motor Control . 262 EMC. 248 current rating. 21. 156 capacitances. 57. 261 diode recovery. 305 loss. 260. 249 diode. 262 dc link. 259. 35. 74 series operation. 435 leakage current. 49. 211 push-pull converter. 32. 39. 49. 248 Motor Control . 215 transformer construction. 265 MOS Controlled Thyristor. 19 switching. 62 structure. 273 anti-parallel diode. 29. 225 pre-converter. 67 gate-source protection. 36 turn-on. 34. 262 pulse width modulation. 250 dV/dt. 155. 264 gate charge. 47. 245 inverter. 153. parallelling. 253 drive.

117 Resistor mains dropper. 448. 446. 469 reverse battery. 225 Reverse leakage. 309 reluctance. 92.Index Power Semiconductor Applications Philips Semiconductors drive. 301 permanent magnet motor. 138. 446. 93. 293. 135. 294 triac. 459. 431 TOPFET. 119 transformer. 225 modelling. 119 advantages. 121 disadvantages. 313 hybrid. 295 overload. 448. 287. 119 transistor voltage. 475 logic level FET. 134 reverse biased. 479 short circuit. 99. 448 temperature. 451 Pulse operation. 419 Phase angle. 134.ac Power dissipation. 500 Phase control. 396. 119 rectifier circuit. 432 loss. 446. 459. 302 linear. 169 Reverse recovery. 305 inrush. 393. 120 diodes. 139 . 213 multiple outputs. 119 cross conduction. 429 motor current. 93. 447. 154. 162 RFI. 138. MOSFET loss Power factor correction. 311 step angle. 471 servo. 73 schottky diode. 143. 582. 286 full bridge. 557 see High Voltage Bipolar Transistor loss. 119 duty ratio. 473. 309 bipolar. 119 transistors. 557 Mounting torque. 219. 431 stall. 74. 108 Push-pull converter. 446. 457. 310 Mounting. 309 unipolar. 537 Ruggedness MOSFET. 430. 119 output ripple. 162 RBSOA. 447. 167. 310 chopper. 430 inverter. 99. 288 Motor Control . 537 Protection ESD. 558 Pulse Width Modulation (PWM). 429 IGBT. 121 Qs (stored charge). 312 permanent magnet. 557 forward biased. 314 drive. 298 short circuit. 120 magnetics. 523 Phase voltage see motor control . synchronous. 119 output filter. 545 Resonant power supply. 293 FREDFET. 587 flux symmetry. 119 electronic ballast. 180 switching frequency. 111. 135. 162 PCB Design. 131. 287 high side switch. 471 TOPFET. 286 torque. 52. 139 Rectification. 498 triac. 288 MOSFET. 482 overvoltage. 544. 287 half bridge. 25. 236 MOSFET. 179 Reset winding. 546 thyristors and triacs. 303 duty cycle. 158. 452. 475 topologies. 154 Mounting base temperature. 293. 99. 525 voltage rating. 581 vi Power MOSFET see MOSFET Proportional control. 251. 53 pre-converter. 430 permanent magnet. 62. 119 current mode control. transistor. 149 Passivation. 530. 368. 119 clamp diodes. 119. 285 PWM. 529. 497. 173 Safe Operating Area (SOA).Stepper. 119 operation. 457. 580 active. boost converted. 514 Parasitic oscillation. 287 freewheel diode. 445. 285. 286. 286 efficiency.

126. 109 Stepper motor. 92. 109. 379 control ic. 11 Stepdown converter. 118. 298 Single ended push-pull see half bridge converter Snap-off. 591 self oscillating power supply. 124. 118. 111. 502. 111. 197 magnetisation energy. 174 ruggedness. 391 control loop. 339. 115. 501 SOT186. 119 diode loss. 544 Switched Mode Power Supply (SMPS) see also self oscillating power supply 100W 100kHz MOSFET forward converter. 171 horizontal deflection. 121. 129. 109 vii Storage time. 380 mains input. 111. 153. 139 power MOSFET. 192 100W 500kHz half bridge converter. 176 SMPS. 154 SOT186A. 143 Speed control thyristor. 162 Suppression mains transient. 339. 133 Speed-up capacitor. 121. 137. 133 Selection Guides BU25XXA. 112. 527 Starter fluorescent lamp. 384 pulse width modulation. 153 isolation. 33. 379 full bridge converter. 473 Solid state relay. 108 push-pull converter. 113 core loss. 163 parasitic oscillation. 389 asymmetrical. 136 power-up. 110 buck converter. 108. 154 SOT199. 111. 441 MOSFETs driving motors. 112. 123 forward converter. 119 . 93. 199 ETD transformer. 111. 111. 144 Stored charge. 166 diode reverse recovery softness. 174 reverse leakage. 529. 392 output rectifiers. 125 half bridge converter. 442 MOSFETs driving lamps. 191. 383. 379 epitaxial diodes. 153 continuous mode. 201 Static Induction Thyristor. 531 triac. 187 16 & 32 kHz TV. 167 current mode control. 113 base circuit design. 339 Self Oscillating Power Supply (SOPS) 50W microcomputer flyback converter. 167 Snubber. 112. 166 diode reverse recovery effects. 279 Softness factor. 174 edge leakage. 426 Schottky diodes. 136. 111. 199 Servo. 111 flyback converter. 174 guard ring. 154 Space charge. 126 discontinuous mode.Index Power Semiconductor Applications Philips Semiconductors Saturable choke triac. 469. 120 dc-dc converter. 124. 473 turn off. 345 EPI diodes. 167 Solenoid TOPFET. 113 isolated packages. 176 technology. 108 opto-coupler. 580 Startup circuit electronic ballast. 139. 94. 392 isolated. 549 active. 331 BU25XXD. 167 diodes. 469. 108 ceramic output filter. 113 mains filter. 109 buck-boost converter. 122 high voltage bipolar transistor. 161 flux swing. 111 magnetics design. 108 core excitation. 153 100W 50kHz bipolar forward converter. 173 bulk leakage. 523 Schottky diode. 149 boost converter. 149 power-down. 331 damper diodes. 153. 523. 113. 115. 109. 116. 173 SCR see Thyristor Secondary breakdown. 390 MOSFET. 384 multiple output. 309 Stepup converter. 173 selection guide. 156 non-isolated. 343 MOSFETs driving heaters. 495.

499 holding current. 490 half wave control. 489 . 107 topology output powers. 139 resonant see resonant power supply RFI. 114 two transistor forward. 490 energy handling. 490 applications. 492 gate requirements. 111. 383 soft start. 109 symmetrical. 111. 381. 167 schottky diode. 490 load line. 154. 111. 93. 379. 70. 111 transformer. 493 full wave control. 112 transistor turn-off. 154. 501 commutation. 126 turns ratio. 561 single shot operation. 124. 112 regulation. 382 synchronous rectification. 74. 568 Thermal resistance. 339. 382 standby supply. 490 dIf/dt. 505 external commutation. 499 fusing I2t. 118. 491 dV/dt. 391 transistor current limiting value. single shot. 500 gate current. 503 latching current. 112 transistors. 514 operation. 498. 493 reverse characteristic. 156. 497. 490 gate power. 109 stepup. 502 speed controller. periodic. composite. 558 rectangular pulse. 158. 568 Thyristor. 112. 500 inrush current. 392 start-up. 391 stepdown. 121. 557. 399 two transistor flyback. 509 ’two transistor’ model. 138 transistor turn-on. 490. 139. 230 Synchronous. 493 RFI. 531 static switching. 561 rectangular pulse. 154. 391 topologies. 136 transistor voltage limiting value. 561 Thermal capacity. 512 gate triggering. 527 asynchronous control. 492 mounting. 179 TDA8380. 500 gate circuits. 527 pulsed gate. 139 rectification.Index Power Semiconductor Applications Philips Semiconductors RBSOA failure. 111 TV & Monitors. 562 rectangular pulse. 180 Temperature control. 112 transistor mounting. 108 reliability. 505 phase angle. 503. 509 leakage current. 557 Thermal time constant. 509 inductive loads. 568 viii Thermal characteristics power semiconductors. 509 cascading. 500 resistive loads. 568 non-rectangular pulse. 511 dI/dt. 115. 497 avalanche breakdown. 10. 497 structure. 498 resonant circuit. 181 transformer driven. 497 self commutation. 117 Switching loss. 156. 500 phase control. 490 overcurrent. 119. 492 gate specifications. 173 snubber. 381. 502 snubber. 489 switching. 138 transformers. 138 standby. 163 rectifier selection. 493 series choke. 111 transformer saturation. 568 intermittent operation. 557 Thermal impedance. 512 gate cathode resistor. 497 current rating. 492 control. 179 self driven. 490. 490. 537 Thermal continuous operation. 558. 392 rectification efficiency. 565 pulse operation. 122 synchronisation. 490 breakover voltage. 154 transistor selection. 503 peak current. 497 Synchronous rectification. 489 reverse recovery.

500 TV & Monitors 16 kHz black line. 549. 364. 494 turn-on. 510 charge carriers. 455 leadforms. 497 temperature rating. 497. 529. 341 horizontal deflection. 527. 341 picture distortion. 502 snubber. 500 triggering. 549 speed controller. 463 linear control. 500 inrush current. 509 turn-on dI/dt. 465. 399 32 kHz black line. 494 commutation. 537 asynchronous control. 473 PWM control. 467. 495. 500 gate circuits. 451. 549 control. 531. 502 varistor. 512 thermal specifications. 551 inductive loads. 491 overcurrent. 529. 339. 449. 447. 461 5 pin. 503 voltage rating. 113 Transformers see magnetics Transient thermal impedance.Index Power Semiconductor Applications Philips Semiconductors switching characteristics. 540. 510 operation. 445. 492 gate resistor. 518. 537 Trigger angle. 451. 461. 503 zero crossing. 503. 500 quadrants. 339 SMPS. 518 applications. 457. 352 . 509 Time proportional control. 550 turn-on dI/dt. 459 solenoids. 451. 367 diode modulator. 537 TOPFET 3 pin. 502. 500 gate current. 457. 463 driving. 491. 500 phase control. 546 protection. 497 structure. 348 power MOSFET. 491. 489 synchronous control. 351 30-64 kHz autosync. 491 ix gate requirements. 354. 492. 111. 492 pulsed gate. 525 dI/dt. 475 high side. 358. 447. 491 gate triggering. 367 EHT. 499 fusing I2t. 527. 502 varistor. 510 resistive loads. 490. 469 Transformer triac controlled. 523 Transformer core airgap. 114 Two transistor forward converter. 544 pulse triggering. 545 gate sensitivity. 489. 494 commutating dV/dt. 512 time proportional control. 467 protection. 475 lamps. 389. 491. 523 dV/dt. 523 transient protection. 510 Thyristor data. 445. 409. 517 synchronous control. 497 dc inductive load. 111. 497 transformer load. 559 Transient thermal response. 523 series choke. 361 damper diodes. 457 motor control.354. 497 breakover voltage. 523. 549 full wave control. 469. 368. 451. 549 dIcom/dt. 523 dc motor control. 510 Hi-Com. 430. 345. 497 saturable choke. 503 isolated trigger. 489 switching. 531 . 455. 494. 501 latching current. 523. 402 Two transistor flyback converter. 410 high voltage bipolar transistor. 459 negative input. 549 commutating dI/dt. 518 400Hz operation. 538 holding current. 500 turn-off time. 399 vertical deflection. 465. 502 trigger angle. 498. 497 transient protection. 456. 503 phase angle. 459. 527 static switching. 549 emitter shorting. 502 trigger angle. 327. 512 gate cathode resistor. 473. 449. 510. 117 Universal motor back EMF. 379. 111. 154 Triac. 498 RFI. 453. 339. 523.

364. 402 Voltage doubling. 537 x . 537 Zero crossing. 527 Varistor.Index Power Semiconductor Applications Philips Semiconductors starting. 537 Zero voltage switching. 122 Water heaters. 358. 528 Vacuum cleaner. 503 Vertical Deflection.

Sign up to vote on this title
UsefulNot useful