July 12, 2007

Electric Drives:
Understanding Basics

Ned Mohan

Applied Power Electronics Conference &

Exposition (APEC)
Sponsored by: IEEE Power Electronics Society

Course Presenters Biography

Ned Mohan is Oscar A. Schott Professor of Power Electronics in the Department of Electrical
and Computer Engineering at the University of Minnesota, where he has been teaching for 31
years. He has written five textbooks; one of them is translated into several languages.
He is a Fellow of the IEEE. He received the Distinguished Teaching Award of the Institute of
Technology, and the Morse-Alumni Distinguished Teaching Award conferred by the
University of Minnesota.
Ned earned the following degrees: Indian Institute of Technology, Electrical Engineering, B.
Tech. 1967; University of New Brunswick (Canada), Electrical Engineering, MS 1969;
University of Wisconsin - Madison, Nuclear Engineering, MS 1972 ; University of Wisconsin Madison, Electrical Engineering, Ph.D. 1973; University of Wisconsin Madison, Electrical
Engineering, Post-doc 1973-1975.

IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 2 / 17

Course Outline

This course will focus on the basics and analyze induction and permanent-magnet ac
machines in a way that clearly explains how these machines operate on a physical basis, and
hence how they ought to be controlled for optimum performance.
After completing this course you should be able to develop an understanding of:

How ac machines operate on a physical basis

How ac machines can be controlled for optimum performance

IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 3 / 17

Course Summary / Key Points

Course Summary / Key Points

Gained a basic understanding about the steady state operation of electric drives

Examined the physics of operation

Made use of the space vector theory

Related IEEE Expert Now titles Include:

Electric Drives: Understanding Basics by Ned Mohan

Introduction to Power Electronics by David Torrey

IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 4 / 17

Course Transcript

Electric Drives: Understanding Basics

This tutorial is in two modules. Module 1 deals with electric drives under steady state
operation and we have a textbook for it and similarly we have second module which deals
with dynamic operation of this electric drives and their control and again we have a textbook
for it and these two modules relate to two-semester long courses at the University of
Minnesota where the first one for a steady state operation is an undergraduate course and
the second one on the dynamic operation and electric control that is a graduate level course
which also draws quiet a few undergraduate students and by stretching these courses in this
way we are able to teach from very basics to advanced electric control only in two semesters.

Electric Drives: Understanding Basics, Part 1

This module 1 is for basic understanding in steady state operation of electric drives and in
this we have a very unique approach where we look at the physics of operation and rely less
on the mathematics and we also make use of space vector theory which has been made very
easy. So it is no longer reserved for advanced graduate level courses and similarly when it
comes to the second module we will see that we will make use of DQ windings rather than
mathematical transformation and the material that we will cover in this module 1 would
seamlessly transition to vast electric control. So this seamless transition can be represented
by this slide here where we see that the low torque is taking a step jump and the objective
maybe to keep the machine speed, the load speed constant. So normally this would require
sophisticated electric control as we will see in module 2, but what we see here would then
seamlessly apply to that kind of electric control.

What is an Electric-Motor Drive?

So first we should define what an electric drive is, and as you can see in this slide we have a
load on the right hand side, an electric utility normally is the source of energy and power and
in between in the dotted box we have this electric drive which consists of a motor and this
motor is being driven through a power processing unit which consists of power electronics
and we are sensing the parameters like the speed and the current and the position of this
motor load combination and making use of this controller to provide the feedback to this
power processing unit. So in this approach we are looking at the complete electric drive
where motor is really a subset of it and therefore we can talk about this motor and this electric
drive and some very interesting applications, just two examples here, harnessing of wind
energy by windmills and hybrid electric vehicles. Coming back to this power processing unit,
we see here that we will make use of semiconductor switches which are represented here by
three bi-positional switches in a very simple manner and they make up switching power poles

IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 5 / 17

Course Transcript

and there are three of them and each switching power pole on an average basis can be
represented by an ideal transformer with controllable turns ratios, so this makes the
description of the power electronics very easy and that is covered in a separate course at the
University of Minnesota called Switch Mode Power Electronics and there is a textbook again
for that which we use it in this course.

Introduction to AC Machines
Now coming back to machines there are primarily two types of machines, DC machines and
AC machines and we will concentrate in module 1 and 2 only on AC machines and these AC
machines can be further categorized into induction machines and permanent magnet
brushless machines quite often called PMAC machines, sometimes they are also called
brushless DC machines.

Sinusoidally-distributed Stator Windings

As we saw earlier in the cross section these AC machines consists normally of three-phase
windings on the stator, if we just concentrate on one of those three windings. For example,
for phase A we see here that the conductor density in this winding is sinusoidally distributed
and the reason is that we would like each of these phase windings to produce a
cosinusoidally distributed flux density distribution at any given time. So if we concentrate here
on phase A, we see that there is a magnetic axis here which is horizontal here for phase A
and when we pass a current into this phase, if we have ideally a sinusoidally distributed
conductor density then the flux density distribution produced by this current in this phase
winding would be cosinusoidally distributed, that is it will be peaking along the phase
magnetic axis and this radial flux density distribution in the air gap then would cosinusoidally
go down as an angle of cosine theta away from this magnetic axis. So we can write this
expression for flux density, a peak here in terms of the current in this phase.

Space Vector to Represent Sinusoidal Distributions

In any machine as mentioned earlier there are three phase windings with currents in them
simultaneously present. So each current in each phase winding is producing flux density
distribution in the air gap which is peaking along its individual magnetic axis, but we would
like to know at any given time what is the complete flux density distribution in the air gap due
to these three currents flowing simultaneously and if we assume that we are working in a
linear region of the magnetic circuit there is no saturation then we can make use of the
superposition and before we do that we would like to represent each sinusoidally distributed
flux density distribution by means of a space vector and this we represent by putting an arrow
on top so you see here that for phase A there is a space vector which is oriented along the

IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 6 / 17

Course Transcript

phase A axis and similarly for phase B and phase C they are respectively oriented along their
magnetic axis and the amplitude of the space vector is proportional to the current in that
phase. So this space vector is a very simple concept and a very physical concept, its telling
us is that wherever this sinusoidal distribution is peaking thats where we orient the space
vector and the amplitude of that distribution then it determines the length of that space vector.
So now we have three of these space vectors representing each of these three phases and
to find the combined or the resultant of all these three which is represented by a subscript S
we will vectorally add them as shown here in the slide and we can see here is that then this
resultant stator space vector is made up of individual phase space vectors added vectorally.

Space Vectors Representation of Combined Phase Currents and

Voltages
We saw earlier that flux density was sinusoidally distributed in the air gap, but we ought to be
able to represent even the terminal quantities like current and phase voltages by means of a
space vector. So we will define this mathematically. For example, for the stator current space
vector we will define it as the individual phase currents at any given time each multiplied by
their axis orientation and add it together and the same thing we will do for the voltage space
vector.

Physical interpretation
Now we can also give a physical interpretation to this mathematical definition of current
space vector for the stator and that we can do as shown here in the slide by putting a
hypothetical winding which is again sinusoidally distributed and has the same number of
turns as any of the other phase windings, but in this hypothetical winding the current flowing
is amplitude of the stator current space vector which is

Is and this hypothetical winding is

oriented at the same angle as the stator current space vector and the advantage of doing this
is that when we come to calculate the torque on all these three phase windings and then we
will have to add that torque rather than doing that we can just calculate the torque on the
single hypothetical winding representing all three phases.

Balanced Sinusoidal Steady-State Excitation (Rotor Open-Circuited)

Since the emphasis in this module is on balanced sinusoidal steady state operation it will be
very useful to see what these space vectors do as time progresses. So here we have the
currents in three phase windings represented sinusoidally and if we try to combine these
three into a stator current space vector we will find that this combined stator current space
vector would rotate with time at the same speed as the frequency of this excitation. So when
the current and phase A for example is peaking the combined stator current space vector
IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 7 / 17

Course Transcript

would be along the magnetic axis of phase A and from there on it will keep on rotating at the
synchronous speed which is related to the frequency of excitation of these three currents.

Relation Between Space Vectors and Phasors

In sinusoidal steady state its also very useful to see the relationship between the space
vectors and phasors. We know that phasors can only be applied in sinusoidal steady states
although space vectors are more general and they can be applied under even DC conditions,
but we see that there is a definite relationship between space vectors and phasors in
sinusoidal steady state where the amplitude of space vectors is three halves x the amplitude
of the phasors and the orientation of the two is the same. So if space vector is, may be I
should say start with phasors, if the phasor is at an angle then the space vector would be at
an angle as shown here at time zero.

Permanent-Magnet AC (PMAC) Drives

So with this brief introduction of space vectors we can very quickly see how a PMAC drive
operates as shown here. Just to introduce this drive its in a closed loop where we have to
sense the position of the rotor where the permanent magnets are and based on that then we
make the power processing unit feed the appropriate currents into this PMAC motor. So this
we will see very quickly in just a few slides how it operates.

Structure of Permanent-Magnet Synchronous Machines

So in this machine we have a three-phase windings on the stator and on the rotor we have
permanent magnets north and south pole as shown here and we assume here that the flux
density distribution due to the permanent magnets cutting the stator is sinusoidally
distributed, so we can represent that by a space vector Vr as shown here and its orientation is
that m and has this rotor rotates this angle m would change and the amplitude of the space
vector would remain constant because of the permanent magnets.

Principle of Operation
Now, in order to produce torque from this machine what we need to do is to place this stator
current space vector 90 degrees ahead of the rotor flux density vector by 90 degrees in the
direction of rotation as shown here. Then we can make use of this physical interpretation of
what the stator current space vector means. It means is that we have a hypothetical winding.
This hypothetical winding is sinusoidally distributed and the current in that hypothetical
winding is Is , which is producing the same MMF as being produced by the individual phase
currents ia, ib, and ic following through their respective windings.
IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 8 / 17

Course Transcript

Torque Calculation
To calculate torque we can assume that angle m in the previous slide to be zero just to make
it simpler for us, it makes no difference and we can see that the rotor current flux density

space vector is horizontally located over here and is is perpendicular so, we can easily
calculate the torque produced in this machine, but since we have a permanent magnet rotor it
will be much easier to calculate the torque and the stator windings rather than on the rotor
and that is what is done here. We take some angle and in this differential angle d we have
certain number of conductors and these conductors are of length L and the current flowing in
all these conductors is

Is and these conductors are located at some flux density which is

dropping off cosinusoidally from the horizontal axis. So we multiply all those and the radius R
would give us the differential torque and then we can integrate that from /2 to /2 and by
symmetry multiplied by 2 and we can get the total torque that is generated on the stator and
knowing that the torque is equal and opposite on the rotor, we can get the torque being
produced in this machine and that is given by the expression at the bottom here which shows
that this torque is proportional to the amplitude of the stator current space vector x a constant
because all these quantities within the brackets are constant including the flux being
generated by the power magnets.

Similarity Between DC Motor and Brushless DC Motor

So the previous discussion shows the similarity between brush-type DC motors and
brushless DC motors. So as we see here in a brush-type DC motor the feel produced by the
permanent magnets on the stator is constant, is stationary and similarly because of the
brushes and commutators the flux produced by the armature winding on the rotor is also
constant at 90 degrees at all times. In a very similar manner in a brushless DC motor or PM
AC motor whatever you may want to call it, we see that by placing the stator current space
vector 90 degrees ahead of the rotor flux density distribution vector, the angle between the
two always remains 90 degrees and thats why we need to measure the rotor position and
doing this then this very direct similarity between the two types of machines.

Induction Motors: Balanced, Sinusoidal Steady State Operation

So having seen the operation of PM AC motors we can move on to induction machines and
we will initially look at them under balanced sinusoidal steady state operation where they are
being supplied by lets say line frequency rated voltages and currents at the line frequency.

IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 9 / 17

Course Transcript

Structure
We should briefly discuss the structure of these machines, again we have a sinusoidally
distributed three-phase windings on the stator here and the rotor consists of a squirrel cage
that is bars which are placed along the periphery of the rotor and these bars are shorted on
both sides by end rings as shown here.

Electrically Open-circuited Rotor

So initially lets see what happens if somehow we were to cut these bars on the rotor so that
we have an open circuitry rotor and we apply balanced sinusoidal voltages to these three
stator windings, then this structure would just look like a magnetic structure and we will have
what we will call magnetizing currents flowing into these three-phase windings and those are
represented by means of this phasor diagram on the right.

Short-circuiting the Rotor

So in analyzing these machines we will initially assume that the stator windings have zero
resistances and also there is no leakage flux associated with these stator windings and that
would make our analysis very simple and later on then we can introduce the stator winding
resistances and leakage inductances, but it is also very useful to make a transformer analogy
and so in this transformer analogy we will assume that the primary winding has no resistance
and no leakage inductance and we will also put some air gap here to bring this analogy
closer to of an induction machine. And what we see here is that the flux in this core is totally
dictated by the integral of the voltage being applied to the primary winding and there is
nothing we can do to the secondary winding to change the flux. The secondary winding could
be open, could be short-circuited, or could have a load on it or some finite value, but makes
no difference because the Faradays law is telling us that the flux in this core is the integral of
the voltage being applied to the primary winding which is v1.

Induced Voltages in Rotor

So we will make this analogy to our induction machine and initially lets once again assume
for a minute that the rotor bars are open circuited somehow, so currents in the rotor circuit
cannot flow and we are again making that assumption that stator winding resistances and
leakage inductances are zero and we are applying three phase voltages to sinusoidal
voltages balanced to these three phase windings and as a consequence the stator voltage
space vector would rotate at the synchronous speed and so would the flux density distribution
space vector BMS here would also rotate at the synchronous speed and in this case we will
assume that the rotor is rotating at a speed m which is slightly less than the synchronous

IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 10 / 17

Course Transcript

speed. So as a consequence there will be induced voltages in the rotor bars and those are
shown here in this diagram lets say at time t=0.

Induced Currents in Rotor

Now all the rotor bars are rotating at the same relative speed omega sync minus omega m,
but the difference here is that the rotor bars which are at the top and at the bottom are cutting
the maximum flux density and as a consequence the voltages induced in these two rotor bars
would be maximum and if we have the rotor circuit complete then as a consequence there
will be currents which will be flowing through these rotor bars and their distribution would be
based on how much the voltage is induced in the each rotor bar.

Rotor MMF
So initially if we can assume that the rotor circuit has no leakage inductance then at time t=0
the top and the bottom rotor conductors are cutting the maximum amount of flux and
therefore they have maximum induced voltages and also therefore the maximum currents
that is the meaning of larger circles at the top and the bottom rotor conductors and
progressively smaller radii for the other conductors. So again we have a situation where the
current is distributed sinusoidally in these rotor bars and all these rotor currents combined
flowing through these bars would produce MMF and therefore the flux which is shown here,
the result would be that compensating currents would have to be drawn from the three stator
windings because we started out by saying that flux density distribution air gap is completely
dictated by the applied voltages and therefore this flux density distribution cannot change
because the rotor currents and if that is the case then equal and opposite MMF has to be
produced by the additional currents those are drawn from the three stator windings to nullify
the rotor produced MMF and that is shown in this space vector diagram on the right here
which shows that due to the rotor bar currents we have a space vector and due to the
additional currents those are drawn b the three-phase stator windings, this rotor MMF space
vector is nullified and these additional currents are shown by this space vector IR prime and
therefore the resultant stator current would be the vector sum of ir prime and the magnetizing
current IMS.

Slip frequency in the rotor circuit

It is useful to know that in the rotor circuit the voltages and currents are at a slipped
frequency and we should also define the slip speed which is the difference between the
synchronous speed and the rotor speed n and the slipped frequency is related to the
frequency at which we are applying the voltages in a same manner as Omega slip is related
to the synchronous speed.

IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 11 / 17

Course Transcript

Electromagnetic Torque Production

So in this machine the electromagnetic torque can be calculated in an identical manner to PM
AC machines and it is proportional to the peak of the flux density distribution produced by the
stator voltages that we are applying and also to the amplitude of ir prime space vector and
knowing that ir prime space vector depends upon what the flux density distribution peak is
and what the slip speed is, we can see here that this torque is proportional to flux density
distribution peak squared x the slip speed, omega slip.

Torque - Speed Characteristics

So based on the previous expression we see here that the slip speed is directly proportional
to how much electromagnetic torque has to be generated by this machine which in turn
depends up on how much load torque has put on to this machine.

Generator Mode
This physical description also shows how this machine which is so far described as a motor
can go into a generator mode. If you are to rotate this rotor at a speed faster than the
synchronous speed then the direction of voltages and currents would reverse and as a
consequence the torque would be in a direction opposite to the direction of rotation and thats
a generator mode of operation.

Rotor Leakage Inductance

So in the previous discussion we have neglected the rotor leakage flux, but that can be easily
accounted for and that is done in this slide which shows that the additional currents drawn
from the stator cannot nullify the leakage flux because they have no idea that that flux is
reaching the stator and therefore they cannot nullify it. In that case the net flux vr that is
cutting the rotor conductors is really the vectorial sum of the magnetizing flux density
produced by the stator and the rotor leakage flux density and we see here that in this case
the rotor current space vector is 90 degrees ahead of the flux density cutting the rotor bars.

Per-Phase Equivalent Circuit

So based on this analysis its very easy to find the per-phase equivalent circuit as is
commonly done in most conventional textbooks, but the approach here is very different and
we see that again keeping the stator winding resistance and leakage inductance equal to
zero, we see that the equivalent circuit on a per-phase basis consists of a magnetizing
inductance and then a rotor branch which consists of the leakage inductance of the rotor and

IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 12 / 17

Course Transcript

an equivalent resistance and that equivalent resistance depends upon the rotor bar
resistance and the ratio of the synchronous speed to slip speed.

Stator Winding Resistance and Leakage Inductance

And having done that we can easily include the stator winding resistances and the leakage
inductances as done in the slide to make the per-phase equivalent circuit complete.

Characteristics at Rated Voltage and Rated Frequency

Now the previous analysis has shown that the torque versus slip speed relationship is given
by a constant so we get a linear characteristic which is very steep line here, but if the slip
speed goes beyond its rated value then the parasitic effects the second R effects which we
had been able to neglect before become significant and you can see here is that this torquespeed characteristic begins to bend down and actually comes down as shown here by the
dotted line.

Motor Currents, Efficiency, Power Factor As a Function of Load

So to complete this discussion with a motor which is supplied by rated voltages at the rated
line frequency we see here that as a percentage load the efficiency response and the power
factor at which this machine is operating.

Induction Motor Speed Control

So far we have seen this induction machine operated at the rated voltages at rated frequency
which is the line frequency, but we can make use of this induction machine for adjustable
speed operation and get very high efficiencies so that is the intent of showing his block
diagram where the induction machine is supplied through this power processing unit which
consists of power electronics and although it is not necessary, but in general we could say
that there are some sensors which are sensing the speed of this machine and then there is a
reference speed which is going into this controller and this controller is providing the
feedback signal to the power processing unit in order to control this machine.

Operating
To appreciate how this machine can keep on operating at very high efficiency even at low
speeds we will look at the torque-speed characteristics for different applied frequencies, but
maintaining the air gap flux amplitude constant at its rated value. And what we see here is
that we have parallel lines for different frequencies where F1, F2, and F3 are progressively

IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 13 / 17

Course Transcript

smaller compared to the rated frequency and what this slide is showing is that for any torque
that we are demanding from this machine we can operate it at any speed keeping the slip
speed small.

Maintaining
In the previous slide we were showing the torque-speed characteristics keeping the air gap
flux density constant and what that implies is that as we change the frequency we are to
change the amplitude of applied voltages to the three-stator windings and that can be
computed by looking at this phasor diagram in this slide here.

Adjusting Voltage
So as a rough estimate in order to keep the air gap flux density constant what one should do
is have a somewhat of a linear relationship between the amplitude of the applied voltages
and the frequency at which these voltages are being applied, but if one wants to be more
accurate we can see here that depending upon the torque on the machine the voltage being
applied may be larger or smaller compared to this linear V/F relationship.

Capability Below and Above Rated Speed

Finally we can see that this machine can operate at rated torque so long as the speed is at or
below the rated speed and beyond that this machine goes into rated power capability where
the power output is same as the rated output at the limit but the amount of torque that it can
produce goes down. So this pretty much completes our module 1 and very quickly it shows
how we can analyze and understand PM AC and induction machines and here we have
looked at sinusoidal constant frequency operation and also we have seen how we can get
adjustable speed out of these machines keeping them efficient. This is all done in a
sinusoidal steady state manner, but the topic of the next module would be to show that we
can take this analysis and look at the dynamic operation of this machine where the load
torque may change as a step and we want to then calculate how the speed varies and may
be we dont want to vary that speed then in that case how we are to control these machines
so that we get the desired performance.

IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 14 / 17

Glossary

Electric Motor Drives

Electric Motor Drives (or Electric Drives, for short) include all three sub-blocks (the power
processor unit consisting of power electronics, the electric motor, and the controller) in order
to control the speed and/or position of the mechanical load.

Power Processing Unit

It is a sub-block of the an electric motor drive; it consists of power electronics, in order to
supply the electric motor with voltages and currents of the appropriate magnitude, frequency
and phase.

Space Vectors
At a given time, these represent sinusoidal distribution, for example of the flux density in the
air gap, with the vector orientated along the axis of the distribution peak, and the length
representing the amplitude of the distribution.

AC Motors
Unlike dc motors, ac motors require alternating (ac) voltages and currents.

Induction Motor Drives

These electric drives use an induction motor that consists of a rotor which is a short-circuited
cage, in which the currents are induced by induction.

Synchronous speed
It is the speed at which the flux-density distribution, produced by the stator-applied voltages
and currents, rotates in the air gap.

Slip speed
It is the difference between the synchronous speed and the rotor speed.

Slip frequency
It is the frequency of the voltage and currents induced in the rotor circuit of an induction
motor.

PMAC Motor Drives

These electric drives consist of a Permanent Magnet AC Motor, which consists of a rotor with
permanent magnets.

IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 15 / 17

Glossary

Generator Mode
In this mode, an electric machine behaves as a generator where the electromagnetic torque
is developed in a direction opposite to the speed of rotation.

Voltage Boost
In ac motor drives operating at adjustable speeds, in order to maintain the air gap flux density
at its rated value, a boost in voltage is necessary to overcome the voltage drop across the
stator-winding resistance.

IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 16 / 17

References

Electric Drives: An Integrative Approach by N. Mohan, published by MNPERE, 2003 edition;

ISBN 0-9715292-1-3.
Advanced Electric Drives: Analysis, Control and Modeling using Simulink by N. Mohan,
published by MNPERE, 2001; ISBN 0-9715292-0-5.
First Course on Power Electronics N. Mohan, published by MNPERE, 2005 edition, ISBN 09715292-4-8.
Power Electronics: Converters, Application and Design by Mohan, Undeland and Robbins.
Published by John Wiley & Sons, 1989, 667 pages. The 2nd edition with extensive
revisions became available in January 1995. Media Enhanced 3rd edition printed in
2003; 802 pages. Translated into Chinese, Korean and Greek.
N. Mohan, Techniques for Energy Conservation in AC Motor Driven Systems, Electric Power
Research Institute Final Report EM-2037, Project 1201-1213, September 1981.

IEEE Expert Now Electric Drives: Understanding Basics Transcript pg. 17 / 17