Professional Documents
Culture Documents
EEWeb
EEWeb.com
TA B L E O F C O N T E N T S
Bill Hall
ON SEMICONDUCTOR
Interview with Bill Hall - Senior Vice President & General Manager of the Standard Products Group
4 9 11
TABLE OF CONTENTS
Featured Products Digital Power Solutions Offer Performance and Integration Enhancement
BY ALEXEI ZERNOV, YNGVE WERNQVIST & CHRIS YOUNG WITH INTERSIL
Why digital power solutions are becoming more widespread for their advantages in power conversion and efficiency.
16 20
Visit www.eeweb.com
INTERVIEW
Bill Hall
How did you get into electrical engineering and when did you start? I actually got my start by mowing the lawn at an electronics distributor when I was in high school in the early 70s. I was eventually offered a parttime job in the warehouse. When I graduated from high school, the manager of the warehouse offered me a full time job at $135/wk. Since I had no idea what I wanted to do in life, I took the job with the goal of working my way into sales. After 2 years of being told I was more valuable in the warehouse then in sales, I decided it was time to go to college. During my time at the distributor, I had contact with many Electrical Engineers. They seemed really smart and drove nice cars and since I was good at math, I decided to become an Electrical Engineer. In addition to starting college in 1975, I also got married, which meant I needed a source of income. I decided to pick Drexel University in Philadelphia because they were one of the few schools in the country
EEWeb | Electrical Engineering Community
FEATURED INTERVIEW
ON Semiconductor
at that time with a Co-op program. This allowed me to work full-time 6 months of the year while getting a BSEE. I co-opd at RCA Missile and Surface Radar Division and got to do cool things like write software for a missile flight simulator along with some mundane jobs like peeling Rubylith. I eventually graduated from Drexel in 1980 and took a full time job with RCA.
Visit www.eeweb.com
INTERVIEW
Can you tell us more about your work experience before you started at ON Semiconductor? My first job was with RCA as a RADAR Design Engineer. I worked on a RADAR digital signal processor for 5 years. We began by designing the system on paper, pretty much using discrete Logic ICs along with some EEPROMs and some bit slice micro-processors. The design was at the module level, so, for example, one module was a complex arithmetic multiplier, another was a target detector and so on. This was cool stuff because we actually used mathematics that you thought were just in textbooks, like real and imaginary numbers, Eulers Theorem, and RADAR countermeasures techniques. After the design was complete, we simulated the modules in FORTRAN and eventually manufactured the modules and began test and integration. Test and integration lasted years and involved rotating shift work since these modules were installed in a commissioned navy ship (actually just a conning tower) in the middle of a corn field in New Jersey. Eventually, the changing shifts got old and I went to work for Fairchild Semiconductor in Maine. I was hired as an applications engineer, but on my 1st day of work, my boss asked if I would be willing to be a marketing engineer for 3 months then I could go back to applications engineering. I said, if the pay is the same, no problem even though I had no idea what marketing did. I actually found that I liked marketing because you got to make a lot of strategic decisions. So after 3 months, I stayed in marketing and began to climb the ladder, working through various manager positions. Marketing exposed me to many aspects of the business, but the next step I wanted was to run a business. In 1987 Fairchild was sold to National and I took doing a good or bad job, just look at the P&L. National Semiconductor spun off Fairchild in 1997 and I stayed with Fairchild, so I worked for 3 companies in the space of 10 years and never changed my phone number. I kept getting more and more product lines folded into my group and got promoted to Vice President of the Interface & Logic Group in 1999 and eventually created and co-ran the Standard Products Group in 2004. In 2006 I moved from Fairchild in Maine to ON Semiconductor in Phoenix as Senior VP & GM of the Standard Products group. As they say; its a dry heat. What have been some of your influences that have helped you get to where you are today? My father instilled a good set of values and a good work ethic in me. The warehouse manager that I worked for taught me not to worry about what other people are doing just focus on doing the best job you can and cream always rises to the top. Do you have any tricks up your sleeve? Somebody once told me that if you are an engineer, you can do almost any job that is out there, which has worked for me. In my career, I have been a radar design engineer, a semiconductor applications engineer, a marketing engineer, a product engineer, and have had management jobs with finance, quality, HR, supply chain and
FEATURED INTERVIEW
In the near future, we can expect to see wide band gap technologies, much more use of integrated passive components in filtering, oscillators and RF amplification. Also, advances in chip singulation will increase the number of die per wafer without changing lithography.
on several Product Line Director positions. I liked running product lines because you were essentially running a small company within a larger company. In these positions I was responsible for marketing, product development, design, product engineering, applications engineering, supply chain, and all the financial aspects of the P&L. It was real easy to tell if you were
Visit www.eeweb.com
INTERVIEW
FEATURED INTERVIEW
engineering functions reporting into me. All of these functions came with a learning challenge but kept life interesting. What has been your favorite project that youve worked on? My favorite engineering project was designing anti-counter measures hardware for a naval radar. I had to implement algorithms in hardware for target detection in a jamming environment, or detecting a target in cloud cover, or telling a target from a fake pulse that the target broadcasts back to you. Its a very creative process and these modules that you design are like children. Do you have any note-worthy engineering experiences? One experience that stands out was when I first moved to Fairchild in 1985 and became the Marketing Manager for ACMOS Logic. We were launching the FACT family (Fairchild Advanced CMOS Technology). A large competitor in Texas who shall remain unnamed was late to market with their technology and made a big deal of the noise characteristics (ground bounce) in FACT. Lots of the marketing people at Fairchild said dont acknowledge the competitors claims and the issue would go away, but since I had recently been a system design engineer I was pretty sure the issue wouldnt go away and our slow sales ramp was proving that true. I began to tear down the competitors case one issue at a time. Things like noise in a lumped load test jig doesnt simulate noise in a real good revenue generator for many years. What are you currently working on? My group is focussing on High Performance Power Discretes (IGBTs, MOSFETs, Rectifiers) and cutting edge Protection devices like ESD and EMI Filters. We have the industrys first silicon Common Mode Filter. Also, packaging is real important from CSP to high power density packages and IPMs (Intelligent Power Modules from our Sanyo acquisition). Can you tell us more about ON Semiconductor and the technology they are developing? Most of ON Semiconductor is focused on power efficiency: everything from power switching, power control, and power management to low power consumption. We are working on LED Drivers and ESD protection for LED general lighting. In the automotive realm we are working on sensors for lane departure and vehicle positioning, extreme high temperature controllers for in transmission & on engine applications and voltage regulation for infotainment & driver information. In computing and wireless: cutting edge power management for that increases power efficiency and decreases power density. For example: we are developing
The trick to devices getting smaller is more power dissipation in a smaller surface area. The trick to power efficiency is in complex semiconductor processes and techniques that facilitate faster switching speeds.
system with distributed loads and only specific signals in a digital system would be affected by this noise and so on. I created a demo board that showed these factors and travelled around the world with an oscilloscope (and they were pretty big in 1985) and showed small groups of engineers the truth that this was a good technology, if you take the proper precautions. Again, having just left the design community I knew design engineers dont believe marketing hype and it was critical to show real waveforms in a live demo environment. This worked and FACT went on to be a
Visit www.eeweb.com
INTERVIEW
technology for power supplies that will have less than 10mW of power loss in standby mode. Other areas include power conversion for high end power supplies and motor control for industrial applications. We are also adopting technology that was developed for hearing aids to provide high quality audio output for cell phones. On the packaging side we are working on chip scale packaging which reduces the volume of a diode by more than 99% from the most common SMT packages used today. What direction do you see your business heading in the next few years? One trend that I am seeing in the industry is that everything is getting smaller and more power efficient. The trick to devices getting smaller is more power dissipation in a smaller surface area. The trick to power efficiency is in complex semiconductor processes and techniques that facilitate faster switching speeds. What are some new technologies we can expect to see from ON Semiconductor in the near future? In the near future, we can expect to see wide band gap technologies, much more use of integrated passive components in filtering, oscillators and RF amplification. Also, advances in chip singulation will increase the number of die per wafer without changing lithography. What challenges do you foresee in our industry? The business cycles are getting shorter and more pronounced, supply chains are getting very complex, and it is getting harder and harder to squeeze the cost out of the products. What are some of your hobbies outside of work and design? Having lived in Maine, I must say that my biggest hobby is fishing. I have a vacation home on the coast and I like to go offshore fishing for tuna and sharks.
FEATURED INTERVIEW
EEWeb
Electrical Engineering Community
Join Today
www.eeweb.com/register
EEWeb | Electrical Engineering Community Visit www.eeweb.com
Features
Water-resistance (IPX6*) per IEC 60529:2001 Very small PLCC6 package dimensions 3.4 x 2.8 x 1.8mm In-line RGB dies configuration Available in White Surface, Black-Surface and Full Black-Body Wide operating temperature range: -40 to +110
For more information or to request a sample please go to:
Applications
Indoor and outdoor full color display LED advertisement panels Decorative lighting
www.avagoresponsecenter.com/led
ADA4096-2/ADA4096-4 C T S F E AT U R E D P R O D U
PIN CONNECTION DIAGRAMS
OUTA 1 INA 2 +INA 3
8
FEATURED PRODUCTS
+V OUTB
09241-001
09241-002
10 9
ADA4096-2
TOP VIEW (Not to Scale)
7 6
INB
5 +INB ro PMU with V A Buck, Two 300 mA LDOs, 1.2 4 Supervisory, Watchdog, and Manual Reset
nse
The ADP5041 combines one high performance buck regulator and two low dropout regulators (LDO) in a small 20-lead LFCSP to meet 7 OUTB INA 2 demanding performance and board space requirements. The high TOP VIEW switching frequency of the buck regulator enables use of tiny multilayer +INA 3 (Not to Scale) 6 INB external components and minimizes board space. When the MODE pin is set to logic high, the buck regulator operates in forced PWM mode. 5 +INB V 4 When the MODE pin is set to logic low, the buck regulator operates in PWM mode when the load is around the nominal value. When the load current falls below a predefined threshold, the regulator operates NOTES in power save mode (PSM), improving the light load efficiency. For more 1. CONNECT THE EXPOSED Figure 1. LTC2996 PAD TO GROUND. information, please click here.
RFILT = 30 VIN1
ADA4096-2
VBIAS BUCK FB1 PGND EN_BK EN1 MODE LDO1 (DIGITAL) VOUT2 FB2 EN_LDO1 R4 EN2 nRSTO
L1 1H R1
8 +V
PSM/PWM
R2
ON
OFF
FPWM
VIN2
C2 1F
R3
VOUT2 AT 300mA
ON
C5 2.2F
OFF
MR
VBIAS
SUPERVISOR
WDI
VTHR
ON
OFF
EN3
R4
R5
VIN3
EN_LDO2
C3 1F
LDO2 (ANALOG)
VOUT3 FB3
VOUT3 AT 300mA
AGND
block DiagraM
1.2V
09652-001
R3
R7
C6 2.2F
VCC
lator
range of the ADP5041 LDOs extend the battery life of portable TO V V devices. The1ADP5041 LDOsT maintain a power supply rejection CONVERTER 5 greater than 60 dB for frequencies as high as 10 kHz while D GND D operating with a low headroom voltage. 7 4 3
PTAT +
es
Each regulator in the ADP5041 is activated by a high level on the respective enable pin. The output voltages of the regulators and the reset threshold are programmed through external resistor dividers to address a variety of applications. The ADP5041 contains supervisory circuits that monitor power supply voltage levels and code execution integrity in microprocessor-based systems. They also provide power-on reset signals. An on-chip watchdog timer can reset the microprocessor if it fails to strobe within a preset timeout period.
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A. www.analog.com Tel: 781.329.4700 Fax: 781.461.3113 2011 Analog Devices, Inc. All rights reserved.
VTL
VTH
CT2
OT
UVLO
VCC
CT1
400k
UT
2996 BD
The LTC2996 from Linear Technology Corporation measures a remote diodes temperature with 1C accuracy and its own die temperature with 2C accuracy while rejecting errors due to noise and series resistance. The device provides a voltage-proportional-to-absolute-temperature output, as well as individual undertemperature and overtemperature alert outputs, defined by user-adjustable thresholds. No code is required to configure the device. With a 200A quiescent current, the LTC2996 simply provides a precise, space-saving, micropower temperature monitoring solution. For more information, please click here.
Visit www.eeweb.com
Dave Baarman
Director Of Advanced Technologies
"Sed ut perspiciatis unde omnis iste natus error sit voluptatem accusantium laudantium, totam doloremque rem aperiam,
eaque ipsa quae ab illo inventore veritatis et quasi architecto beatae vitae dicta sunt explicabo. Nemo enim ipsam voluptatem quia voluptas sit aspernatur aut odit aut fugit, sed quia consequuntur magni dolores eos qui ratione voluptatem sequi nesciunt. Neque porro quisquam est, qui dolorem ipsum quia dolor sit amet, consectetur, adipisci velit, sed quia non numquam eius modi tempora incidunt ut labore et dolore magnam aliquam quaerat voluptatem. Ut enim ad minima veniam, quis nostrum exercitationem ullam corporis suscipit laboriosam, nisi ut aliquid ex ea commodi consequatur? Quis autem vel eum iure reprehenderit qui in ea voluptate velit esse quam nihil
ARTICLES
JOBS
COMMUNITY
DEVELOPMENT TOOLS
Ele ct ric al En w gi w ne w er .e in ew g
EE
C eb om .c m om un ity
W eb
JOIN
TODAY
TECHNICAL ARTICLE
TECHNICAL ARTICLE
11
TECHNICAL ARTICLE
SDA SCL
Power Management Digital VREF + ADC
L
Controller Interface
DSP
DPWM
Driver
VOUT
compared to analog world controllers. All of this enables digital power technologies to deliver new levels of conversion performance, functionality and integration.
TECHNICAL ARTICLE
Improved Efficiency
Thanks to digital control, the Zl6105 is capable of executing algorithms to optimize the dead-time applied between the gate driver signals for the top and bottom FETs. In a synchronous buck converter the MOSFET drive circuitry must be designed such that top and bottom MOSFETs are never in conducting state at the same time. Conversely, a long period of time which both MOSFTS are off reduces circuit efficiency by allowing current to flow in their parasitic body diodes. The ZL6105 has an algorithm that constantly adjusts dead-time nonoverlap to minimize losses, thus maximizing efficiency. This circuit will null out dead-time differences due to components variation, temperature and loading effects.
DSP While the DSPs were quite . capable in terms of processing power, to achieve the fast processing speeds needed for high frequency switching power supply control, high clock speeds were required. The high clock speed and the inherent high bias current required in those DSPs meant a significant amount of power was consumed in the power conversion process. Furthermore, the DSPs were too expensive for the switching power supply applications. See Fig 1. About 10 years ago, dedicated function state machine based digital power controllers began to appear, first in academia and then in commercial offerings. These state machines were designed specifically to be used as digital switching power supply controllers. Controllers contained specialized HW peripherals for power conversions purposes. They were optimized to the point that digital power started to be economically viable across a wide range of applications. This was that turning point in the history of digital power. A block diagram of the modern digital power converter ZL6105 is shown in Fig 2. A regulation
EN
FC(0,1)
VDD VR
VTRK
BST
Sync Gen
D-PWM NLR
SW
VOUT
SYNC
PLL
ADC ADC
+
DAC
Communication
ADC
MUX
Figure 2: Digital Power Controller ZL6105 Using PID Digital Compensator and Optimized Set of HW Peripherals
Visit www.eeweb.com
12
TECHNICAL ARTICLE
3.3V to 5V CIN VIN
DDC
ZL9117M
COUT
control and monitoring. Input, output voltage and output current monitoring allows ZL6105 to detect system faults and prevent of catastrophic consequences for power supply and load by configurable fault reaction.
TECHNICAL ARTICLE
CIN DDC
ZL9117M
Figure 3: Two ZL9117 Modules Using Communication via DDC Bus for Current Sharing
eliminates dozens of components from the design. Further HW integration, and therefore reliability improvement can be found in integrated FET controllers like the ZL2101 and even fully integrated power supply modules like the ZL9117. Fig 3 shows how easily two ZL9117 modules can be combined to build a two phase current sharing rail. Digital power robustness is enhanced by ability to monitor and respond, in an optimal way,
Stability is a critical operational requirement for power supplies. In regulated power supplies, stability is controlled by the characteristics of the feedback path. Power supply engineers need to ensure stable operation over all load conditions, environmental conditions and component characteristic variations. Design of the feedback loop to be stable under all of these conditions is a time consuming task. Digital power solutions provide an alternative to analog compensation. Digital compensation has no external components and can be tuned just by changing the gain values stored in digital registers. Digital filters are not simply replacements of analog filters. Digital filters can perform functions
to environmental changes. As an example, the ZL6105 monitors both the internal die temperature and external temperatures. This allows the controller to compensate temperature sensitive measurements for accurate
that go far beyond the capabilities of analog filters. For example, in high Q (>0.5) second order circuits, the poles in the plant are complex conjugate poles which may require complex conjugate zeros in the compensation network to effectively
Visit www.eeweb.com
13
TECHNICAL ARTICLE
All these converters use a dedicated state machine for the digital PWM controller and an embedded microcontroller to monitor the circuit, environmental conditions, and configuration profile to setup and modify the state machine operation in real time.
Figure 4: Transient Respond of the Power Supply System Before (Upper Graph) and After Auto Compensation (Lower Graph)
amount of time, producing a more stable power supply and potentially improves the reliability and robustness of the power supply system.
TECHNICAL ARTICLE
Conclusion
Digital power control offers many advantages over traditional analog controllers in terms of optimal performance, reliability, high number of features, and ease of use. Switching from traditional analog power control to digital power control is easy and rewarding. About the Authors Chris Young serves as Sr. Manager for Digital Power Technology at Intersil Zilker Labs. Previously he served Chief Technical Officer for Zilker Labs. He has over 20 years of experience in the management of research and development of state of the art electronic systems. He was one of the founders and vice president of technology at ColdWatt, Inc. Prior to that, he held technical and engineering management positions at leading companies including Dell, Astec Power, Lucent/Bell Labs and Unison Industries. He has had numerous publications and patents in the areas of pulsed power, power control and conversion, and stability analysis. Mr. Young holds Bachelors degree in Physics from the University of Texas and a Masters degree in Electrical Engineering from Texas Tech University where he also serves on the Industrial Advisory Board.
compensate. Conventional analog compensators only provide real zeros for compensation. On the other hand, digital filters can easily provide the complex conjugate zeros to compensate high Q power supplies. Nonetheless, even this advantage, in many cases is not enough to stabilize and optimize a power supply over all conditions. Initial inductor and capacitor values can vary by +/- 10% . This can significantly change the control loop even to the extent that the power supply has substantially degraded stability. For example, electrolytic capacitor characteristics such as the capacitance and ESR can greatly change with temperature. What is really needed is a method for compensating power supplies that is automatic. Intersils Zilker Labs has recently released several parts with auto compensation. All of those use an advanced digital algorithm to characterize the plant and to determine appropriate compensation settings for stable operation.
During auto-compensation, the microcontroller adjusts the state machine to stabilize the power conversion process by adjusting the compensation coefficients in a systematic way while observing the response of the system. While this does produce a slight perturbation on the output it is almost imperceptible and well within the allowed transient envelop. In practice, auto compensation is easy to use. Simply enable the power supply and the controller does all of the work. Fig4 shows the transient response of a typical power supply before (upper) the supply is adequately compensated. The second trace (lower) shows the transient response after auto compensation. An additional benefit of automatic compensation is that the plant is characterized by the compensation algorithm. The values of Gain, Q, and Natural Frequency can be monitored over the life of the power supply and significant changes in the plant can be observed, many times, before failure of the system. This allows the user to incorporate predictive diagnostics of the system health for improved reliability. Auto-compensation saves the design engineer a considerable
Visit www.eeweb.com
14
1.8V to 3.3V, Micro-Power, 15kV ESD, +125C, Slew Rate Limited, RS-485/RS-422 Transceivers
ISL32600E, ISL32601E, ISL32602E, ISL32603E The Intersil ISL32600E, ISL32601E, ISL32602E and Features ISL32603E are 15kV IEC61000 ESD protected, micro power,
wide supply range transceivers for differential communication. The ISL32600E and ISL32601E operate with VCC 2.7V and have maximum supply currents as low as 100A with both the transmitter (Tx) and receiver (Rx) enabled. The ISL32602E and ISL32603E operate with supply voltages as low as 1.8V. These transceivers have very low bus currents, so they present less than a 1/8 unit load to the bus. This allows more than 256 transmitters on the network, without violating the RS-485 specifications 32 unit load maximum. Rx inputs feature symmetrical switching thresholds, and up to 65mV of hysteresis, to improve noise immunity and to reduce duty cycle distortion in the presence of slow moving input signals. The Rx input common mode range is the full -7V to +12V RS-485 range for supply voltages 3V. Hot Plug circuitry ensures that the Tx and Rx outputs remain in a high impedance state while the power supply stabilizes. This transceiver family utilizes slew rate limited drivers, which reduce EMI, and minimize reflections from improperly terminated transmission lines, or unterminated stubs in multidrop and multipoint applications. The ISL32600E and ISL32602E are configured for full duplex (separate Rx input and Tx output pins) applications. The half duplex versions multiplex the Rx inputs and Tx outputs to allow transceivers with output disable functions in 8 Ld packages. Single 1.8V, 3V, or 3.3V Supply Low Supply Currents . . . . . . . ISL32601E, 100A (Max) @ 3V . . . . . . ISL32603E, 150A (Max) @ 1.8V - Ultra Low Shutdown Supply Current . . . . . . . . . . . . . . 10nA IEC61000 ESD Protection on RS-485 I/O Pins . . . . . . 15kV - Class 3 ESD Levels on all Other Pins . . . . . . . . . >8kV HBM Symmetrical Switching Thresholds for Less Duty Cycle Distortion Up to 65mV Hysteresis for Improved Noise Immunity Data Rates from 128kbps to 460kbps Specified for +125C Operation 1/8 Unit Load Allows up to 256 Devices on the Bus -7V to +12V Common Mode Input/Output Voltage Range (VCC 3V) Half and Full Duplex Pinouts; Three State Rx and Tx Outputs 5V Tolerant Logic Inputs Tiny MSOP Packages Consume 50% Less Board Space
Applications
Differential Sensor Interfaces Process Control Networks Security Camera Networks Building Environmental Control/Lighting Systems
10m
DYNAMIC (256kbps)
ICC (A)
ICC (A)
DYNAMIC (128kbps) 1m
STATIC 10 2.7 DE = VCC, RE = GND 2.8 2.9 3 3.1 25C, RD = , CD = 50pF 3.2 3.3 3.4 3.5 3.6 100 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6
FIGURE 1. ISL32600E AND ISL32601E HAVE A 9.6kbps OPERATING ICC LOWER THAN THE STATIC ICC OF MANY EXISTING 3V TRANSCEIVERS
FIGURE 2. ISL32602E AND ISL32603E WITH VCC = 1.8V REDUCE OPERATING ICC BY A FACTOR OF 25 TO 40, COMPARED WITH ICC AT VCC = 3.3V
Intersil (and design) is a registered trademark of Intersil Americas Inc. Copyright Intersil Americas Inc. 2012 All Rights Reserved. All other trademarks mentioned are the property of their respective owners.
photodiodes, solar power is increasingly prevalent in autonomously powered systems. With the worldwide evolution toward lower power operation using more green energy sources, emphasis on deploying solar power in a greater variety of environments has been on the rise. In low-power solar systems, it is critical to assess whether there is sufficient sunlight at a given time to power the system. In some cases, this involves determining whether there is sufficient power to enable the microcontroller. In many ultra-low-power systems, the simple act of waking the microcontroller to make a voltage measurement might collapse the solar source or waste precious power from a reservoir capacitor.
Visit www.eeweb.com
16
TECHNICAL ARTICLE
1V2.5V T1 2N3906 C2 0.1F
TECHNICAL ARTICLE
+
D1 photodiode
R2 2.2M U1 TS1001
Power Status
good wait
R1 100k
Measured Light
V lux
C1 0.01F Figure 1: An ultra-low power op amp can be a continuous, always-on measurement mode.
One solution is to incorporate a simple analog op amp into the system. An ultra-low-power analog op amp can support always-on circuitry around the microcontroller and may be the simplest and best solution.
Ultra-Low-Power Op Amps
Figures 1 to 3 illustrate a few simple circuits using an ultra-low-power op amp in a continuous always-on measurement mode assessing the state of the solar cell. The technique hangs on using an op amp whose total power is as low as practical driven primarily by ultra-low supply voltage operation.
The circuit in Fig. 1 shows a photodiode in photovoltaic zero-bias mode. The short-circuit current is measured and converted to a voltage across resistor R1 while the feedback action of the op amp forces 0 V across D1. Zero-bias photovoltaic current is a convenient parameter that is generally well characterized and can be directly referenced to most photodiode manufacturer datasheets. Note that with the rail-to-rail input range of the analog op amp, in this case Touchstone Semiconductors TS1001 op amp, the photodiode may be connected directly to the positive supply voltage rail. The circuit in Fig. 2 generates a simple positive-polarity output for a similar zero-bias-mode measurement. In
1.8V2.5V
+
D1 photodiode
C2 0.1F
U1 TS1001
Measured Light
V
R1 100k
C1 0.01F
lux
Figure 2: An ultra-low-power op amp offers simple, positive polarity output to support a zero bias condition.
Visit www.eeweb.com
17
TECHNICAL ARTICLE
T1 BSH205 P-ch R10 5.1M C1 47F R1 test load R8 1.4M 1% R2 220K T1 BSH105 N-ch
+
TECHNICAL ARTICLE
Power Source
(solar cell)
(uC, etc)
Load
R7 365K 1%
+
Power Status
wait good
R5 5.1M
R6 510K
R9 220K
T3 BSH105 N-ch
U1 TS1001
R3 1M
C2 10nF
T4 BSH105 N-ch
Power Check
disable enable
this case, the op amp servos its output to sink sufficient current to support the zero bias condition for D1. This creates a voltage across resistor R1 at the negative supply voltage pin of the TS1001. Since this ultra-lowpower op amp contributes less than 1 A to this current, the current measured from D1 has minimal error. For a more comprehensive assessment, the circuit in Fig. 3 tests the solar source to see if it can handle the load. The circuit shown makes this assessment without burdening the microcontroller and risking collapse of the supply during measurement. Approximately once per 100 ms, the circuit disconnects the solar-cell power source from its reservoir capacitor and load, applying a test load (R1) and assessing the resulting voltage drop. If the voltage drops 25% or more, the result is latched into U2 and the power status is provided to the microcontroller. This ultra-low power circuit draws less than 3 A at 1V. Op amp U1 provides the timer function and controls transistor switches T1 and T2 to apply the test load while simultaneously disconnecting the load. Capacitor C1
temporarily holds the voltage to keep this circuitry and any standby loads powered. Op amp U2 serves as a comparator, tripping when the power source drops more than 25% (with 5% hysteresis). Transistor T3 latches the result, while transistor T4 resets the latch during each assessment period to ensure a fresh reading. Such test loading is useful for determining the available power from a solar cell, since merely measuring the open circuit voltage generally does not provide an accurate assessment.
Visit www.eeweb.com
18
TECHNICAL ARTICLE
low-power solar systems to assess the available power from solar cells before a load is applied, and generally to support standby, always-on circuitry, while drawing negligible currents.
TECHNICAL ARTICLE
Dead Zone!
(mV)
-3
0
SENSE
(mV)
19
RETURN TO ZERO
RETURN TO ZERO
EEWeb
Electrical Engineering Community
Join Today
www.eeweb.com/register EEWeb | Electrical Engineering Community
Visit www.eeweb.com
20