Please use side menu if you cannot use dropdown boxes, or the header boxes
CAGE Code: 5DME2
ITAR Registration Code: M28889

- March 26th 2017 -
Call: 1 (800) 377-3244
We Are Open Monday-Friday 8 am - 5 pm PT

AppNotes - Advances in Power Sources


The following is an edited version of a paper first appeared in the May 1987 issue of the
Power Conversion and Intelligent Motion Magazine.

Advances in Power Sources


K. Kit Sum

Classical Problems

According to the Statistical Abstracts of the U.S. Department of Energy Annual Report to Congress in 1981 (D.O.E. Monthly Energy Reviews 1982), two billion kilowatt-hours (kwh) of energy was sourced by hydro electric generators in 1900. By the year 1920, the amount of energy sourced was increased to 20.3 billion kwh; to 51.7 billion kwh in 1940; and to 276 billion kwh in 1980 - this is in addition to 251.1 billion kwh sourced by nuclear means in the same year. The increase of power generation is due primarily to the increase of power consumption and the increase of population. However, it is evident that the increase in population will never keep up with the increase in power consumption. This condition prompted careful consideration in the utilization of electrical energy.

Today, it is inconceivable to think of living in a society without electricity. In today's modern society, almost everything runs on electricity. The need for power processing is obvious. To elaborate further, one must go back a few years.

Ever since there were electron tubes, there were linear regulators whose output voltage is regulated or controlled by a tube functioning as a variable resistor. Back in the early 1900's, vacuum tubes took more than an ampere to energize the filament. They were triodes without cathodes, and were some of the most primitive types of tubes. Electron tubes of that era were used to build regenerative radio receivers. These tubes perform the functions of r.f. amplifier, detection, audio amplifier, and power amplifier. Sometimes all these functions are performed by the same tube type. Many of these receivers use the loudspeaker armature for the core of the filter choke while the pulsating direct current flowing through the choke also served as the energizing current for the electromagnet, which also functions as the magnet for the loudspeaker. Needless to say, the resultant audio quality of the receiver was poor and the sound was modulated by the rectified line ripple frequency. Power processing for this type of receiver was simple and straightforward; even half-wave rectification was used in some of them. Many of these receivers did not have regulators in them. Voltages were allowed to drift up or down with the line voltage. Others with regulators were of the linear type, whose design is based on the classical feedback theory of Black [1] and Bode [2,3], which formed the foundation of the voltage mode control. Consumers also grew accustomed to the line ripple "hum" lurking in the background of the music.

As time went on, electron tube technology improved. Cathodes were inserted between the filament and the grid to isolate the "hum" produced by the line frequency. More electrodes, screen and suppressor screen grids, were inserted to form tetrodes and pentodes, respectively. By the end of World War II, the functions of electron tubes were well-defined. There were beam tetrodes for power amplification, triodes for small signal amplification, pentodes for radio frequency amplification, etc. The acorn tubes (such as the 954) were developed to handle radio frequency signals in portable military equipment. Other types of electron tubes were also designed for use with battery equipment. These tubes were the so-called miniature tubes and subminiature tubes, and were very popular in automobile radio receivers. To energize them, a +90 volts bus is required for the plate circuits, and a 1.5 to 3.0 volts bus is required for the filaments. Sometimes the filaments of all the tubes were connected in series with a combination of resistors, and only one bus was used to run both plate and filament circuits. But, in the automobile, there is only a 12-volt battery. In order to step up the battery voltage to run the car radio, a vibrator was used to chop up the d.c. from the battery, then stepped up with a transformer designed to run at the chopping frequency of the vibrator.

The vibrator, as used in the automobile, was probably one of the first commercial attempts in switch mode power supply (SMPS) implementation. A fundamental problem, in this case, is that the vibrator has moving parts and physical contacts that can wear out in time, rendering a very limited life for the power processor. Another problem is that the frequency of operation is also very much limited by the physical design of the reeds, thus a humming sound is always audible during operation. Power conversion was sometimes performed by ferroresonant regulators, which lack the precision of today's SMPS. At this point, the reader must realize that power processing is a multi-discipline technology involving control theory, filter synthesis, signal processing, thermal control, and magnetic components design. The designer must be conversant with all of the above mentioned disciplines before an optimum design can be conceived. The concept of switch mode power conversion was long in existence, but the technology, up to this point in time, was not quite ready.

Technological Advancement

On July 15, 1948, the birth of the point contact transistor was announced [4]. This transistor was somewhat difficult to manufacture, since the point contact arrangement is, to a great extend, similar to that of the old crystal detector, which relies on a "whisker" to make point contact at a sensitive spot on the crystal to obtain optimum detection.

The point contact transistor was quickly replaced by the junction transistor. The transistor did not reach the consumer market until the mid 1950's. However, in the earlier days, the junction transistors had rather low gain and low speed. Thus, many power processing circuits were designed to operate at well below 20 kHz. The transistor played a key role in facilitating the practical implementation of solid state power processing equipment. Research work was carried out to investigate the behavior of the transistor under large signal conditions [5]. Suffice to say that, similar to many products, the transistor took a while to get established in the consumer market. In those early years, many researchers and engineers toyed with transistors in the area of power oscillators and processors [6 to 63].

About this time, Shockley [64] announced the findings of his investigation - the field effect transistor. This was the early version of a small signal type junction field effect transistor. The first example of a power metal oxide silicon field effect transistor (MOSFET) did not appear until 1959 [65], and in spite of further progress [66,67], did not reach the consumer market until the late 1960's.

The big boost in electronics development came when the Soviet Union launched the Sputnik I, the world's first artificial satellite on October 4 of 1957. In November 1960, John F. Kennedy was elected president of the United States. He told the Americans to overtake the Soviets at all costs. By this time, the U. S. had already launched the first successful earth satellite TIROS I (Jan. 31, 1958). This was the time when the reduction of size and weight of power processors became a top priority in the space programs. Also, efforts were made to increase the switching speed of 'slow' transistors. One of these methods made use of the tunnel diode [68] in a most unconventional manner [69].

Fundamental Development

If the power processor is viewed as a control system, then the lack of understanding is quite evident in reports by Raposa, Seaver and Ponsi [73] and Poulo and Greenblatt [74], especially in the behaviour of the control loop. In an attempt to solve a practical loop stability problem, Froeschle [75 to 78] designed a push-pull power converter using the approach now known as current mode control to avoid loop instability. This is probably one of the first, if not the first, documented application of the current mode control concept. The "two state modulation techniques" reported by Froeschle correspond to the well known continuous conduction mode of operation. Control is achieved by duty ratio adjustment. This converter was regarded as a state-of-the-art pulse-width control design at the time.

The first major breakthrough in understanding the behavior of this type of nonlinear circuits came in 1972 when Wester [79] announced his research results in three separate models with a linearization technique, which allowed the use of linear circuit theory to analyze the control loop of the basic power converters. The linearized model succeeded in representing the small-signal transfer properties of the basic power converters. However, this does not mean that there were no ways to analyze the power converter [80]. There has always been the large mainframe computer with numerical approximation techniques or time domain analysis techniques which when executed, produce quite a few yards of paper in the printout, from which the designer is expected to gather some meaningful information or indication of design performance. Those who have tried these techniques will understand the tedious nature of the interpretation process. Wester's results provided a simple means of arriving at meaningful results with physical significance without complicated programs or costly computing equipment.

An important generalization in modeling, which essentially solved the problem of modeling the transfer properties of switch mode modulators and converters, was made by Middlebrook and Cuk in 1976 [81], the birth of the canonical model. The 1976 paper provided complete small signal modeling information on the three basic converters operating in the constant frequency continuous conduction mode. One of the most interesting results was the discovery of the existence of the right-half-plane zero in the control-to-output transfer functions of the boost and the buck-boost converters. This discovery has brought about significant understanding of the nature of the non-linearities of the two basic converters. However, the behavior of these two converters are rather different when operating in the constant frequency discontinuous conduction mode. The details of converters operating in the discontinuous conduction mode were given in a later paper [82]. A simplified explanation with explicit design equations for constant frequency discontinuous conduction mode converters can be found in [83].

While doing investigations in power converter modeling, Cuk also discovered that, by applying the principles of duality, new converters are obtainable. Thus, the boost converter was found to be a dual of the buck converter. Delving into the dual of the flyback (buck-boost) converter, Cuk invented the optimum topology (Cuk) converter (patented) [84], a converter utilizing the principle of capacitive energy transfer. Analysis of this converter in its basic form have been performed with the programmable calculator [85].

Another by-product of Cuk's investigation was that the windings of the input and output inductors of the new optimum topology converter could be wound on the same magnetic core, producing the coupled-inductor Cuk converter. A natural evolution from the coupled-inductor converter was the integrated magnetic converter [86 to 91]. The reader would be glad to know that, after many years of design and experimentation, a workable approach to the design of coupled inductor power converters is now available, thanks to Cuk and Zhang [92] who made a thorough investigation of the problems surrounding the coupled inductor and emerge with some elegant solutions. Zhang's findings also served to provide an interesting solution to the integrated magnetic converter problem.
Since most of the popular power conversion topologies (the bridge converter, the half-bridge converter, the forward converter, and the push-pull converter) are buck-derived, they can be similarly analyzed with no difficulties [93,94].

State of Power Conversion Technology

SMPS - This is, by far, the most encompassing phase in power conversion. Broadly speaking, any power processor that uses switch(es) for the purpose of power conversion can be regarded as a switch mode power supply (SMPS). Even a switching power amplifier is a form of SMPS. For this reason, most SMPS fall into one or more of the following categories:

DC to DC Converters
DC to AC Inverters
Off-Line AC to DC Converters
High Voltage Power Supplies
Switching Power Amplifiers
High Frequency Power Converters

DC to DC Converters - As the name implies, this class of power converters accept d.c. input of one voltage range to produce one or more d.c. output voltage levels. The d.c. input requirement is usually imposed by the mobility of the equipment, such as a military communication transreceiver or an airborne data acquisition system.

Other applications include battery backup of other power systems. The current technology, with wide availability in the consumer sector, is the constant frequency pulse-width modulated control power supplies at the higher power range, with some variable frequency power supplies at the lower power end.

By definition, the d.c. to d.c. converter does not require larger input capacitors, since there is usually no hold up time requirement. It is, therefore, imprudent to compare the power density of this class of converters with the off-line type converters.

D.c. to d.c. converters are also making adjustments to accommodate lower output voltage requirements (down to 1.2 volts) by modularization. Power converters are made in modules of a given output wattage. More modules are plugged in with the outputs connected in parallel to achieve higher output power levels. This means that a certain degree of precision must be designed into each module to provide current sharing capabilities. The latest technology in this class of converter is reflected in a following section on High Frequency Power Conversion.

D.C. to A.C. Inverters - This is another class of power processor that command a great deal of attention in earlier years [95] but much less so in recent developments. Much of the design methodology is well known and serves to extend to quite high power levels. These power processors are usually intended for single frequency output with the more specialized ones designed for variable frequency output. They are very popular in military mobile applications. For the single frequency type inverters, a few methods of power processing are widely practiced. One method is to 'chop' the input d.c. at a constant frequency corresponding to a multiple of the fundamental output frequency (in the case of the push-pull converter, this frequency will be the second harmonic of the fundamental for single phase design). The duty ratio is set by constraining the operating point with the transformer turns ratio to approximately 67% for the rejection of the third harmonic. To ensure a low distortion waveform at the output, a series LC tuned circuit is usually followed by a parallel LC tuned circuit to filter out unwanted harmonics. An optional third harmonic trap is sometimes included for better waveform quality. Another method uses pulse-weighted modulation technique [96] to obtain a sinusoidal output waveform. This method is highly suited for MOSFET implementation due to the promising switching characteristics of the device. By design, this type of power inverter requires somewhat less stringent output filtering than the previous method. The step-approximated synthesis method [97] promised less harmonic distortion and thus, required less filtering.

Due to the portability of this class of equipment, surface mounting devices will dominate the future packaging of this type of power processors for less bulk and weight.

Off-Line A.C. to D.C. Converters - This class of power converter is the most in demand by both the consumer and the industrial sectors. These converters serve a wide variety of requirements and applications. One of the most popular application is found in the computer industry where these converters are used to run computers and peripherals. Thus, the requirement for hold-up time is imposed on these power converters for data preservation in the case of a power transient surge or brown out. For a 47 to 63 Hz a.c. input, the hold-up time is provided by input capacitors with effective capacitance approximating to 3 µF per watt of output power, but with current state of technology it is still difficult to find suitable capacitors to meet current ripple requirement at the higher power range. These capacitors are necessarily of the high voltage type and occupy a substantial amount of volume within the power converter package. Parallelability is important for this class of power processor, since customer requirements tend to increase rather than staying within a prescribed power limit.

One of the major reasons for this increase in sophistication in power converter design is due to the system designer/manager, who failed to include some of the features into the 'system' as an integral design; but left them all to the power design engineer. The power processor should, at all times, be regarded as part of a system and be designed as such. It should never be regarded as an add-on item to run a particular system. It is foreseeable that the optimum system will have the power supply designed for and with the system, such that the power supply is not required to operate as a stand-alone item. By designing the power supply with the system, it is possible to optimize packaging, heat-sinking, power-up sequencing, cooling, etc. to a new level of performance, reliability and power density. In future, off-the-shelf power supplies will have to make adjustments to match an optimum system design philosophy.

One of the latest awareness, as power output level reaches 2kw and beyond, is the power factor improvement problem. This is due to the peak-charging of the input capacitors within a switching period which caused a high peak current for a short period of time, resulting in inefficient usage of the present form of input power.

Surface mounting components are also beginning to take the place of the conventional components in this class of power converters.

For device utilization, the senseFET (announced by Motorola in Electronic Design, Feb.20, 1986), which has the capability of sensing the current through the device facilitated by an extra electrode, is expected to be found in future designs of this class of power converters. International Rectifier is expected to second source this device.

Frequency synchronization will be important for large systems requiring more than one power supply. Parallelling and current sharing will be common place in such a system.

High Voltage Power Supplies - As far back as 1891, Nikola Tesla performed high frequency high voltage experiments in the United States [99]. The development of high voltage d.c. was due to the pioneering work of Cockcroft and Walton [100,101]. In 1939 a high voltage photography technique was perfected by Kirlian [102] who claimed to be able to capture the human aura on film. Another area of photography requiring the use of high voltage is X-ray photography for which a high voltage ranging from 30 to 300 kV is required [108]. Other high speed photography includes the generation of spectrograms in which the spectrum of a faintly radiating source in time periods of a few microseconds is observed through an image intensifier sourced by a high voltage power supply. In this case, a four-stage image intensifier with a radiant gain of 105 to 106 is placed at the spectrograph exit for this purpose [103].

The high voltage power supply is also used in more advanced image intensifier tubes with microchannel plates [104 to 106]. Some of these image tubes were installed with suitable optics for night vision enhancement, known as 'green eyes' to military personnels. The most popular ones were those of the 18mm second generation type, which were installed in pairs for bifocal low light vision applications [107]. The design of the second generation 18mm image tube power supply was known for its difficulties due to stringent size, weight, low noise and low power consumption specification. Each power supply typically accommodates an input voltage range of 2 to 2.7 V.d.c., not exceeding 14 mA of nominal input current for the delivery of one +6 kV output, one -800 V output, and one +200 V output, with automatic brightness control and bright source protection shut-down features. The third generation is even more stringent on requirements. The tube remains at the same size, but with much higher resolution, consumes more power, but maintains the same input power budget for the new power supply. Current development on night vision technology has been more concentrated in the forward looking infra red (FLIR) systems.

Note that for a 20 kV power supply, 0.1% output voltage ripple represents 20 V peak of ripple amplitude, or a swing of ±40 V peak-to-peak. For a tightly regulated output, this ripple appears on top of the voltage divider which samples the output voltage for feedback control. This voltage divider is usually made up of one high value (as well as high voltage) resistor in the range of 10 M" to a few gigaohms, and a low value resistor. The potential drop across the high value resistor is, by design, very high. In many cases, this resistor acted as a radiator and affect the control circuit components in the proximity. Sometimes loop stability is not achievable until the high voltage resistor is completely shielded.

Low noise high voltage power supplies are also required for applications to devices such as the traveling wave tube (TWT) and the laser (Light Amplification by the Stimulated Emission of Radiation).

The TWT was particularly sensitive to phase noise, which is one of the major reasons that variable frequency type power supplies are rarely found in a TWT amplifier system [109].

Other applications of the high voltage power supply include the radar modulator [110], xenon or mercury arc lamp power supply circuits, high resolution computer monitors, and television receivers.

The state of high voltage electronic technology was nicely summarized by Williams [111].
Switching Power Amplifiers - One of the first known switching power amplifiers is invented by Bedford in 1931 [114], who later also took out another patent [115] on switching amplifier. This type of amplifier is later known as the class D amplifier. Other pioneering work in this area was reported by Œhmichen [116] and later by Page, et al. [117], Robertson [118], and Chudobiak and Page [119]. The work by Stark [120] follows along a similar line of description of [116].

Many Class D power amplifier techniques form the basis for switch mode regulated power supplies. Class D amplifiers use pulse modulation techniques to convert input signals to pulse trains, which are then used to recover input signal information from amplified pulse trains. A high theoretical efficiency of 100% is obtainable with 'ideal' components.
The switching power amplifier is expected to find a good future in public address audio systems because of its high efficiency and light weight. Except for the critical audiophile, the switching amplifier is gaining acceptance in products such as guitar amplifiers and some electronic organs.
Recent high efficiency switching amplifiers have been reported by Koenig [121] and Lee & Troiani [122].

High Frequency Power Conversion - In recent years, the interest in high frequency power conversion has been increased substantially; partly due to the demand from equipment manufacturers for more compact and lighter weight power processors and, partly due to the advancement in circuit integration technology, reflecting that most of the bulk of the system would soon be the power supply (unless it is also proportionately reduced in size).

Higher conversion frequency is a solution, at least in principle [123]. But maturity in many areas of research is required to bring high frequency high power density converter realization into perspective. These research areas are: (a) Power Topology, (b) Magnetic Materials, (c) Capacitors, (d) Semiconductor Devices, (e) Circuit Fabrication Techniques, and (f) Synchronous Rectifiers.

(a) Power Topology: This is a multi-dependency choice influenced by power level, method of conversion and control, nature of the input power, and the application intended.

As far as the method of conversion is concerned, the resonant type of converter is currently attracting a lot of attention. A good description of different classes of resonant converters and their modes of operation can be found in a series of articles by Todd and Lutz [124]. Other approaches worthy of attention are given in [125,126,127].

For low power d.c. to d.c. applications, the forward converter appears to have taken a foothold in the military market [128]. This converter operates in the semi-resonant mode. The resetting of the forward converter power transformer is performed by a novel circuit [129]. This power converter claims a power density of 24 watts/in.3 but requires external add-on heat-sinking arrangement. This is one off-the-shelf item that is not designed to function as a stand-alone part. It is, therefore, prudent to assume that the state-of-the-art power density is still under 24 watts/in.3 In terms of power density, it is evident that this converter should not be compared with off-line power converters, which are being penalized by the inclusion of large input capacitors for meeting the hold-up time requirement.

Power conversion researchers have also looked into different standard topologies for high frequency power conversion. One of these researchers is Schlecht [130], who suggested that the single-ended forward converter is among the best, if not the best (for component count, simplicity, cost, etc.).
(b) Magnetic Materials: A limitation in increasing the power density of a converter is that the magnetic components are limited to a minimum of one turn on a winding. Note that the implementation of fractional turns are neither simple nor economical. For off-line designs, the turns ratio becomes approximately 14:1 for a 5-volt output power converter. The remaining problem is to find a core with a wire window large enough to accommodate the number of turns (14 in this case) and the current capability of the winding. With this thought in mind, it is conceivable that operation above a pre-determined frequency will not reduce the core size significantly.

As far as magnetic materials are concerned, Schlecht [130] also indicated that for a 50-watt d.c. to d.c. converter operating at 10 MHz with 4C4 material, the flux swing had to be reduced to .05 Tesla to avoid excessive core loss. This is an indication of the limitation in size reduction.

It is noteworthy that as the frequency goes up, the required permeability of the core material decreases. At frequencies approaching 50 MHz, the permeability of the core material could practically be reduced to that of air. Under this condition, the air-core inductor becomes a radiator similar to the radiation output coil of a radio transmitter. The normal interference filtering problem due to conducted emission will be transformed to one of shielding due to radiation.

Some results of advancement in magnetic core material and magnetic component implementation can be found in Ueda, et al. [131], Matsuki, et al. [132], and Pollack and Smith [133].

(c) Capacitors: There are very few types of capacitors in the market to-day that are suitable for high current ripple, high frequency and high voltage applications [138]. Many capacitor manufacturers do not specify the capacitor in a manner required by the power converter design engineer. Sometimes, the factory test conditions are insufficient for the characterization of the capacitor for power conversion purposes. As a result of the inadequacy of the manufacturers' specification, failure modes have to be studied separately by the user.

One of the recent capacitors from Nippon Electric Corporation is the crushed-ceramic capacitor with working voltages of 25 and 50 volts. It has low equivalent series resistance (esr) and is characteristically a promising candidate for high frequency power conversion. Unfortunately, the price of this unit is far from being competitive.

A more interesting line of capacitor called the "SupraCap" is being marketed by AVX Corporation [139]. This capacitor is the multi-layer ceramic type exhibiting very low esr (a few milliohms) in the frequency range of 40 kHz to 100 kHz. This line of capacitors has voltage ratings ranging from 50 to 500 volts.
The current state-of-the-art is aiming at high dielectric constant for reduction in size, low esr for high current ripple capability and high voltage for industry requirement.

(d) Semiconductor Devices: At present, it appears that the MOSFET is the prime candidate for high frequency power conversion. But experimental circuits have indicated that the output capacitance (Coss) of the existing devices are too high (a few hundred pico farads) for efficient high frequency power processing. An obvious area of improvement is, therefore, the reduction of the output capacitance of these devices. Another important area of improvement is the reduction of the device ON resistance (RDS(on)). For high frequency power processing, the drain-gate capacitance also plays an important role. However, this capacitance can be 'neutralized' by classical circuit techniques used in tube circuits.

In regard to characterization of rectifiers, Carsten [140] suggested that the quantity 'reverse recovered charge per ampere of forward current' be used as the basis for switching performance evaluation.

A glimpse into the advances in device technology can be found in a report by Korman, et al. [141]. (e) Circuit Fabrication Techniques: The power density of a power converter is very much dependent on the packaging technique used by the manufacturer. Approximately ten years or so ago, Zenith Corporation was advertising their television sets as having the 'real guts' for real point-to-point hard wiring. That was not a problem for a floor model console set.

Today, possibly one of the most enlightening experience is to open the chassis of a video cassett recorder to observe the packaging techniques used in the manufacture of such a sophisticated piece of equipment. The indication is that the circuit fabrication technique plays a very important role in the final appearance and performance of the product.

Power conversion is a continually evolving technology. As the conversion frequency increases, the packaging and fabrication techniques are expected to change accordingly. Some of these efforts are reflected in works by Jones and Vergez [142] and Estrov [143].

For the more conservative power supply manufacturers, the logical evolvement would be to go from 'surface mounted pcb' to a combination of 'surface mounted pcb with hybrids'; and eventually to 'high voltage power integrated circuits with any of the previous combinations' for optimum packaging fabrication.

The surface mounted pcb is attractive for present development because the printed circuit board (pcb) has no size limitation, whereas the size of the ceramic substrate of the hybrid circuit is limited by its brittleness.
Due to its anticipated application at high frequencies, it would appear that printed resistors on hybrid circuits are at a disadvantage because of conductor (resistive tract) path length tend to occupy a substantial area and is susceptible to noise pick-up. A simple solution would be to substitute printed resistors with chip resistors.

(f) Synchronous Rectifiers: Presently, all MOSFET's are manufactured with a body diode which is slower than the MOS device itself. When connected into a synchronous rectifier circuit, the direction of the polarity of this body diode directly affects the performance of the circuit, depending on which direction the diode is facing. Perhaps a more specialized device could be designed for this particular application in the near future.

Noting that the inefficiency in low voltage output power supplies is always due to the inefficient rectifier. Improvement in this area would definitely constitute great improvement in temperature rise as well as power density, since there would be less heat sinking requirement.

A more suitable device for synchronous rectification has yet to be found.
On April 4, 1987, President Reagan [150] announced that NASA is to go ahead with a scaled-down version of the space station. The finishing date has been targeted for mid-1990 at a cost of $10.9 billion. One of the most controversial issues in the power conversion circle is the method of power distribution in a space station. A quick insight into the complexity of this task can be obtained from a few recent articles [151 to 154].

A crude and non-exhaustive overview of the advances of power sources for approximately half a century have been provided. To enhance the interest of the reader, various applications of power converters to specific devices were briefly outlined. References have been included for further reading and follow-up investigation purposes. The author would appreciate the opportunity of receiving comments and corrections of errors from helpful readers.

1. H. S. Black, Stabilized Feedback Amplifiers, Bell Systems Technical Journal, January 1934.
2. H. W. Bode, Relations Between Attenuation and Phase in Feedback Amplifier Design, Bell Systems Technical Journal, 1940.
3. H. W. Bode, Network Analysis and Feedback Amplifier Design, Van Nostrand Reinhold, New York, 1945.
4. J. Bardeen and W. H. Brattain, The Transistor, a Semiconductor Triode, Physics Review, Vol.74, No.2, p.230, July 15, 1948.
5. J. J. Ebers and J. L. Moll, Large Signal Behaviour of Junction Transistors, Proceedings of the Institute of Radio Engineers, Vol.42, p.1761, 1954.
6. R. L. Bright, et al., Transistors as On-Off Switches in Saturable-Core Circuits, Electrical Manufacturing, Dec. 1954, p.79.
7. R. L. Bright and G. H. Royer, Electrical Inverter Circuit, U.S. Patent No. 2783384, February 26, 1957.
8. G. Bruck, R. Harter, I. M. Wilbur, Power Supply Circuit Using Controllable Electron Solid State Devices, U. S. Patent No. 2798160, July 2, 1957.
9. G. W. Bryan, Application of Transistors to High Voltage Low Current Supplies, Proc. I.R.E., Vol. 40, 1952, p.1521.
10. J. W. Caldwell & T.C.G. Wagner, Boosting Power Transistor Efficiency, Electronics, Nov.21, 1958, p.86.
11. E. L. Campbell, Combination Power Supply and Modulator Using Transistors, Q.S.T., Sept. 1958, p.18.
12. C. H. R. Campling, Differential Magnetic Multivibrators, A.I.E.E. paper number 57-769.
13. C. H. R. Campling, A Transistor D.C. Converter, Electronic Engineering, Vol.30, 1958, p.508.
14. C. H. R. Campling, Magnetic Inverter Uses Tubes or Transistors, Electronics, March 14, 1958, p.158.
15. W. H. Card, Four Transistor Inverter Drives Induction Motor, Electronics, Vol. 32, Feb.20, 1959, p.60.
16. K. Chen and A. J. Schiewe, A Single Transistor Magnetic Coupled Oscillator, Trans. A.I.E.E. Vol. 75, Part 1, 1956, p.396.
17. M. Cohen and D. Arany, Subminiature Beacon for Guided Missiles, Electronics, April 1957, p.144.
18. E. H. Cooke-Yarborough, et al., D. C. Converter Circuit, British Patent Application No. 20668, 1953.
19. R. O. Decker and F. Gourash, Operational Magnetic Amplifier with Audio Frequency Transistor Power Supply, Trans.A.I.E.E., Vol.74, Part 1, 1955, p.490.
20. G. M. Ford, 400-Cycle Inverter is Transient-Proof, Aviation Age, Vol.29, May 1958, p.150.
21. E. Franklin and J. B. James, The Application of Transistors to the Trigger, Ratemeter and Power Supply Circuits of Radiation Monitors, Proc. I.E.E., Paper No. 2049 M, Vol. 103B, March 1956, p. 497.
22. G. Grimsdell, The Economics of the Transistor D.C. Transformer, Electronic Engineering, Vol.27, 1955, p.268.
23. R. M. Hubbard, A Transistorized D.C. to A.C. Inverter with Good Frequency Regulation, Nat. Conf. Aeronautical Electronics, Dayton, Ohio, May 14, 1956.
24. P. J. H. Janssen, Circuit Arrangement for Converting a Low Voltage into a High Direct Voltage, U.S. Pat. No.2780767, February 5, 1957.
25. P. J. H. Janssen and C. De Vijver, Ringing Choke D. C. Converter Circuit, Brit.Pat.App. No.21263/54.
26. J. L. Jensen, An Improved Square Wave Oscillator, Trans.I.R.E., Vol. CT-4, 1957, p.276.
27. J. L. Jensen, Saturable Core Transformer Oscillators, U.S. Patent No. 2774878, December 18, 1956.
28. R. P. Johnson, High Power Transistorized Mobile Power Supply, Q.S.T., April 1958, p.11.
29. D. L. Johnston, Transistor H.T. Generator: Replacement Unit for Hearing Aid H.T. Batteris, Wireless World, Vol.60, 1954, p.518.
30. R. L. Karl, 100 Watt Transistor Mobile Power Unit, Q.S.T., June, 1958, p.36.
31. A. Kruper, Switching Transistor Substitute for Mechanical Choppers, Trans.A.I.E.E., Vol.74, Part 1, p.141.
32. L. H. Light, Transistor Power Supplies: Circuits for Obtaining H.T. and E.H.T. from Low-Voltage Sources, Wireless World, Vol. 61, 1955, p.582.
33. L. H. Light, Circuit Arrangement for Converting a Lower D.C. Voltage into a Higher D.C. Voltage, U.S. Patent No.2791739, May 7, 1957.
34. L. H. Light and P. M. Hooker, Transistor D.C. Convertors, Proc.I.E.E., Vol. 102B, Paper No. 1862 R, April 1955, p.775.
35. J. F. Lohr, Transistorised Static Inverter Design, Electronic Design, April 16, 1958.
36. H. A. Manoogian, Transistor Photoflash Power Converter, Electronics, August 29, 1958, p.29.
37. W. A. Martin, Development of Transistor Inverter at 20 kc/s Using Power Transistors, Trans.I.R.E., Vol. I-6, 1957, p.118.
38. A. J. Myerhoff and R. M. Tillman, A High-Speed Two-Winding Transistor-Magnetic-Core Oscillator, Trans. I.R.E., Vol. CT-4, p.228.
39. A. G. Milnes, Phase Locking of Switching Transistor Converters for Polyphase Power Supplies, Trans.A.I.E.E., Vol.74, Part 1, 1955, p.587.
40. A. G. Milnes, Transistor Circuits and Applications, Proc. I.E.E., Vol. 104B, Paper No. 2368 R, May 1957.
41. D. C. Mogen, Operation of Saturable-Core Square-Wave Oscillator, Nat.Conf. Aeronautical Electronics, Dayton, Ohio, 1956.
42. H. T. Mortimer, D.C. to A.C. Square-Wave Static Converter and Amplifier, Naval Research Laboratory, Washington, D.C., N.R.L. Report No. 4582, July 27, 1956.
43. J. Noordann, The Balanced Transistor Converter, Philips Telecom. Review, Vol. 18, No.3, Sept. 1957.
44. F. Oakes, Design Considerations of Junction Transistor for the Conversion of Power from Direct to Alternating Currents, Proc. I.E.E., Vol. 104B, Paper No. 2299 R, January 1957, p.307.
45. D. A. Paynter, An Unsymmetrical Square Wave Power Oscillator, Trans.I.R.E., Vol. CT-3, 1956, p. 64.
46. D. A. Paynter, A Single Power Transistor D.C.- D.C. Converter, Proc.A.I.E.E. and I.R.E. Conf. on Transistor Circuits, Pennsylvania University, February 17, 1956.
47. D. A. Paynter, Transistor A.C.- D.C. Converters, Nat. Conf. on Airborne Electronics, May 9, 1955.
48. A. R. Pearlman, Transistor Power Supplies for Geiger Counters, Electronics, Vol.27, August 1954, p.144.
49. B. Reich, Report on Power Transistors for Converters, Electronic Design, Vol.5, March 15, 1957, p.22.
50. G. H. Royer, Switching Transistor D.C. to A.C. Converter Having an Output Frequency Proportional to the D.C. Input Voltage, Trans. A.I.E.E., Vol.74, Part 1, 1955, p.322.
51. S. Schenkerman, Designing Transistor D.C. to A.C. Converters, Electronics, September 26, 1958.
52. P. L. Schmidt, Voltage Conversion with Transistor Switches, Bell Labs. Record, Vol. 36, 1958, p.60.
53. J. S. Smith, D. C. Transformers, Trans. I.R.E., Vol. VC-6, 1956, p.53.
54. R. R. Smyth, Transistors as Power Conversion Devices, Proc. A.I.E.E. and I.R.E. Conf. on Transistor Circuits, Pennsylvania University, February 18, 1955.
55. R. R. Smyth and M. G. Schorr, Transistorised Power Sources for D.C. to A.C. and D.C. to D.C. Conversion, Electronic Design, November 15, 1956.
56. E. E. Thompson, A Transistorised D.C. Power Supply Employing a Novel Method of Voltage Regulation, Program of the 1958 Canadian Convention of the I. R. E.
57. P. M. Toscano and J. B. Heffner, C.R.T. Power Supply uses Transistor Oscillator, Electronics, Vol.29, September, 1956.
58. G. S. Tsykin, Semiconductor D.C. Converters, Radio Engineering, Vol. 12, 1957, p.70.
59. G. C. Uchrin, Transistor Power Converter Capable of 250 watts D.C. Output, Proc. I.R.E., Vol. 44, 1956, p. 261.
60. G. C. Uchrin and W. O. Taylor, A New Self-Excited Square Wave Transistor Power Oscillator, Proc. I.R.E., Vol. 43, 1955, p. 99.
61. H. H. Van Abbe and J. J. Rongen, The Design of Transistor D.C. Converters, Electronic Applications, Vol. 16, Autumn 1955, p. 59.
62. R. L. Van Allen, A Variable Frequency Magnetic-Coupled Multivibrator, Trans. A.I.E.E., Vol. 74, Part 1, 1955, p. 356.
63. F. Weitzsch, The Application of PNP Junction Transistors as Electronic Switches, Electronic Applications, Vol. 16, Spring 1956, p.154.
64. W. Shockley, The Unipolar Field Effect Transistor, Proc. I.R.E., Vol. 40, November 1952.
65. H. A. R. Wegener, The Cylindrical Field Effect Transistor, IEEE Trans. Electron Devices, Vol.6, 1959.
66. S. Teszner and R. Giquel, Gridistor - A New Field Effect Device, Proc. IEEE, Vol. 52, 1964, pp. 1502-1513.
67. R. Zuleeg, Multi-Channel Field Effect Transistor Theory and Experiment, Solid State Electronics, Vol.10, 1967, pp. 559-576.
68. L. Esaki, Physical Review, Vol. 109, No.2, p.603, January 1958.
69. R. E. Morgan, High Frequency Time-Ratio Control with Insulated and Isolated Inputs, IEEE Trans. on Magnetics, March 1965.
70. R. E. Morgan, A New Control Amplifier Using a Saturable Current Transformer and a Switching Transistor, A.I.E.E. Trans. (Comm. and Electronics), Vol. 77, November 1958.
71. R. E. Morgan, A New Magnetic-Controlled Rectifier Power Amplifier with a Saturable Reactor Controlling On Time, Paper T-121, A.I.E.E. Special Technical Conf. on Nonlinear Magnetics and Mag. Amps., Philadelphia, Pa., October 26-28, 1960.1
72. R. E. Morgan, Conversion and Control with High Voltage Transistors with Insulated Inputs, 1968 INTERMAG Conf. Digest.
73. F. L. Raposa, R. K. Seaver and R. A. Ponsi, Non-Dissipative DC to DC Regulator-Converter Study, Final Technical Report, June 15, 1964 to March 31 1967, Contract No. NAS 5-3921, Goddard Space Flight Center, Greenbelt, Maryland, Report Access Number N67 31460.
74. L. R. Poulo and S. Greenblatt, Research Investigations on Feedback Techniques and Methods of Automatic Control, Bose Corporation, Natick, Massachusetts, Contract Number DAAB07 -67-C-0520, Report Access Number AD 687311, April 1969.
75. T. A. Froeschle, Two State Modulation Techniques for Power Systems, Semi-Annual Report, Bose Corporation, Natick, Massachusetts, Contract Number DA28-043-AMC-02282(E), Report Access Number AD 815603, June 1967.
76. T. A. Froeschle, Two State Modulation Techniques for Power Systems, Semi-Annual Report, Bose Corporation, Natick, Massachusetts, Contract Number DA28-043-AMC-02282(E), Report Access Number AD 823239, November 1967.
77. T. A. Froeschle, Two State Modulation Techniques for Power Systems, Semi-Annual Report, Bose Corporation, Natick, Massachusetts, Contract Number DA28-043-AMC-02282(E), Report Access Number AD 831768, April 1968.
78. T. A. Froeschle and J. J. Wawzonek, Two State Modulation Techniques for Power Systems, Semi-Annual Report, Bose Corporation, Natick, Massachusetts, Contract Number DA28-043-AMC-02282(E), Report Access Number AD 844635, November 1968.
79. G. W. Wester and R. D. Middlebrook, Low Frequency Characterization of Switched Dc-to-Dc Converters, Proc. IEEE Power Electronics Specialists Conference, 1972 Record.
80. B. A. Wells, et al., Analog Computer Simulation of a dc to dc Flyback Converter, Suppl. to IEEE Trans. Aerospace, Vol. AES-3, November 1967, pp. 399-409.
81. R. D. Middlebrook and S. M. C~uk, A General Unified Approach to Modelling Switching Converter Power Stages, Proc. IEEE Power Electronics Specialists Conference, 1976 Record.
82. S. M. C~uk and R. D. Middlebrook, A General Unified Approach to Modelling Switching Dc-to-Dc Converters in Discontinuous Conduction Mode, Proc. IEEE Power Electronics Specialists Conference, 1977 Record.
83. K. K. Sum, On the Design of Constant Frequency Discontinuous Conduction Power Converters, Proc. Power Electronics Design Conference, San Jose, California, October 1986.
84. S. M. C~uk and R. D. Middlebrook, A New Optimum Topology Switching Dc-to-Dc Converter, Proc. IEEE Power Electronics Specialists Conference, 1977 Record.
85. B. P. Erisman and K. K. Sum, Analysis of Optimum Topology Converter Using HP-41 Calculator, Powerconversion and Intelli- gent Motion, November 1985.
86. G. E. Bloom and A. Eris, Practical Design Considerations of a Multi-Output C~uk Converter, IEEE Power Electronics Specialists Conference 1979 Record.
87. S. M. C~uk, DC-to-DC Switching Converter with Zero Input and Output Current Ripple and Integrated Magnetics, U.S. Patent No.4257087, March 17, 1981.
88. G. E. Bloom and A. Eris, DC-to-DC Converter, U.S. Patent No.4262328, April 14, 1981.
89. S. M. C~uk and R. D. Middlebrook, DC-to-DC Converter Having Reduced Ripple Without Need for Adjustments, U.S. Patent No. 4274133, June 16, 1981.
90. G. E. Bloom and R. Severns, Magnetic Integration Methods for Transformer Isolated Buck and Boost Converters, Proc. of POWERCON 11, April 1984.
91. S. M. C~uk, A New Zero Ripple Switching Dc-to-Dc Converter and Integrated Magnetics, IEEE Power Electronics Specialists Conference 1980 Record.
92. Z. Zhang and S. M. C~uk, Coupled Inductor Analysis and Design, Proc. IEEE Power Electronics Specialists Conference, 1986 Record.
R. L. Bright, Junction Transistors Used as Switches, Trans.A.I.E.E., Vol. 74, Part 1,1955, p.111.
93. R. D. Middlebrook, Design Techniques for Preventing Input-Filter Oscillations in Switched-Mode Regulators, Proc. POWERCON 5, the Fifth National Solid-State Power Conversion Conference held in May 4-6, 1978, in San Francisco, California.
94. K. K. Sum, Switch Mode Power Conversion - Basic Theory and Design, Marcel Dekker, Inc., 270 Madison Avenue, New York, N.Y. 10016, September 1984.
95. B. D. Bedford and R. G. Hoft, Principles of Inverter Circuits, John Wiley & Sons, Inc., New York 1964.
96. R. Ruble and W. Treitel, A New Technique for Simplifying Sinewave Synthesis Inverter Design, Proc. POWERCON 6, 1979.
97. J. P. Vergez, Jr. and V. Glover, Low Power Solid State Inverters for Space Applications, Paper No. 20/4, WESCON 1966 Record.
98. R. W. Lye, ed. Power Converter Handbook - Theory, Design, Application, Power Delivery Dept., Canadian General Electric Co. Ltd., 107 Park Street North, Peterborough, Ontario, Publication No. PGEI-10355A, 2nd printing 1979.
99. N. Tesla, Lectures, Patents, Articles, Beograd, Yugoslavia (1956). Available from W. S. Heinman, Imported Books, 1966 Broadway, New York, N.Y. 10023; I. Hunt and W. Draper, Lightning in His Hand, the Life of Nikola Tesla, Sage Books, c/o Swallow Publications, Chicago, IL (1964).
100. J. D. Cockcroft and E. T. S. Walton, Experimental with High Velocity Positive Ions - (I) Further Developments in the Method of Obtaining High Velocity Positive Ions, Proc. Royal Society, A, Vol.136, pp.619-630, 1932.
101. E. Everhart and P. Lorrain, The Cockcroft-Walton Voltage Multiplying Circuit, The Review of Scientific Instruments, Vol.24, No.3, pp.221-226, March 1953.
102. S. Krippner and D. Rubin, ed., Galaxies of Life, the Human Aura in Acupuncture and Kirlian Photography, Gordon and Breach, New York, N.Y. (1973); Republished with minor revisions under the title: The Kirlian Aura, Photographing the Galaxies of Life, Doubleday, New York, N.Y. (1974).
103. P. Kafalas, An Image Intensifier System for High-Speed Spectrography, Proc. 8th International Congress on High-Speed Photography, Stockholm, June 23-29, 1968, pp.182-185.
104. D. J. Ruggieri, Microchannel Plate Imaging Detectors, IEEE Trans. Nuclear Science, Vol. NS-19, No.3, June 1972.
105. Paul Henkel, et al., High Gain Microchannel Plates, IEEE Trans. Nuclear Science, Vol. NS-25, No.1, February 1978.
106. A. Lundy, et al., Avalanche Transistor Pulser for Fast-Gated Operation of Microchannel Plate Image Intensifiers, IEEE Trans. Nuclear Science, Vol. NS-25, No.1, February 1978.
107. A Review, Battlefield Night Vision, Miltronics, Vol.3, No.2, June 1982, pp.73-76 and p.99.
108. Y. Cheron, et al., Study of a Resonant Converter Using Power Transistors in a 25 kW X-Rays Tube Power Supply, Proc. ESA Sessions at 16th Annual IEEE PESC, Univ. Paul Sabatier, Toulouse, 24-28 June, 1985, pp.295-306.
109. W. Muller, The Influence of Mode Switching on Voltage Stability and Current Ripple Control of a Pulsed TWT, IEEE Power Electronics Specialists Conference 1985 Record.
110. F. C. Schwarz and J. B. Klaassens, A Radar Power Supply Without a Voltage Droop, IEEE Power Electronics Specialists Conference 1983 Record.
111. J. W. Williams, High Voltage Electronics - An Overview, IEEE Power Electronics Specialists Conference 1980 Record.
112. I. Arens and F. Forattini, Failure Analysis of a High-Voltage System by Probing on the Low-Voltage Side, IEEE Power Electronics Specialists Conference 1985 Record.
113. G. R. Sundberg, A New Very High Voltage Semiconductor Switch, IEEE Power Electronics Specialists Conference 1985 Record.
114. B. D. Bedford, U. S. Patent No.389855, September 25, 1931.
115. B. D. Bedford, U. S. Patent No.1874159, August 30, 1932.
116. J. P. Œhmichen, L'amplificateur "Classe D", Electronique Industrielle, March-April 1955.
117. D. F. Page, et al., On Solid State Class D Systems, Proc.IEEE (Correspondence), Vol. 53, April 1965.
118. J. Robertson, Solid-State Amplifier Gives 1.2 kW at 95% Efficiency, Electronic News, sec.1, p.5, Monday, May 10, 1965.
119. W. J. Chudobiak and D. F. Page, Frequency and Power Limitations of Class-D Transistor Amplifiers, IEEE Journal of Solid-State Circuits, February 1969.
120. P. A. Stark, Class-D For Efficiency, in three parts in Audio, June, July and August 1964.
121. L. W. Koenig, Very High Efficiency, High Power VLF/LF Amplifier, IEEE Power Electronics Specialists Conference 1983 Record.
122. J. Lee and C. Troiani, Applications of Power MOSFET Devices in an Ultra High Efficiency Servo Amplifier, Ninth International PCI Conf., October 1984.
123. B. Carsten, Radio Frequency Power Conversion - Fad or Future? Powerconversion & Intelligent Motion, January 1986.
124. P. C. Todd and R. W. Lutz, Practical Resonant Power Converters - Theory and Application, Parts I, II, and III, Powertechnics Magazine, Vol.2, Nos. 4, 5, and 6, April, May, and June, 1986.
125. P. J. Carlson, A Duty-Cycle Controlled Series Resonant Converter, Powertechnics Magazine, Vol. 2, No.3, March 1986.
126. B. Carsten, A Hybrid Series-Parallel Resonant Converter for High Frequencies and Power Levels, Second International High Frequency Power Conversion Conference 1987.
127. B. Carsten, Design Tricks, Techniques and Tribulations at High Conversion Frequencies, Second International High Frequency Power Conversion Conference 1987.
128. P. Vinciarelli, Forward Converter Switching at Zero Current, U. S. Patent No.4415959, Nov. 15, 1983.
129. P. Vinciarelli, Optimal Resetting of the Transformer's Core in Single Ended Forward Converters, U. S. Patent No.4441146, April 3, 1984.
130. M. Schlecht, 10 MHz Switcher Seminar, held at Jet Propulsion Laboratory, January 15, 1986.
131. R. Ueda, et al., Fundamental Analysis for Design of a Divided Open-Core Type Inductor, IEEE Trans. Magnetics, Vol. MAG-22, No.5, September 1986.
132. H. Matsuki, et al., Performance of Miniaturized Magnetic Devices in Cloth Structure, IEEE Trans. Magnetics, Vol. MAG-22, No.5, September 1986.
133. H. C. Pollack and A. L. Smith, Slim Profile Power Inductors Made by Powder Metallurgy, IEEE Trans. Magnetics, Vol. MAG-22, No.5, September 1986.
134. N. Kato, et al., Ferrite Substrates for High-Frequency Switching DC-to-DC Converter Applications, IEEE Trans. Magnetics, Vol. MAG-21, No.5, September 1985.
135. A. Kamada, et al., Optimum Ferrite Core Characteristics for a 500 kHz Switching Mode Converter Transformer, IEEE Trans. Magnetics, Vol. MAG-21, No.5, September 1985.
136. V. J. Thottuvelil, et al., High-Frequency Characteristics of Amorphous Metallic-Alloy Tape-Wound Cores, IEEE Power Electronics Specialists Conference 1983 Record.
137. B. Carsten, High Frequency Conductor Losses in Switchmode Magnetics, Eleventh International PCI Conference 1986.
138. R. A. Frantz, Reliability Considerations for Metallized Plastic Film Capacitors Under High-Stress AC Waveforms, IEEE Power Electronics Specialists Conference 1981 Record.
139. B. Travis, Capacitors, EDN Magazine, January 22, 1987.
140. B. Carsten, Reverse Recovery Characteristics of High Speed Rectifiers, Powerconversion and Intelligent Motion, February 1986.
141. C. S. Korman, et al., High Efficiency, High Frequency Power Supply Development - New Advances in Device Technology, First International High Frequency Power Conversion Conference 1986.
142. C. B. Jones and J. P. Vergez, Application of PWM Techniques to Realize 1 MHz Off-Line Switching Regulator with Hybrid Implementation, Proc.IEEE Applied Power Electronics Conference 1986.
143. A. Estrov, Packaging of 1 MHz Power Supplies, 2nd Int'l. High Frequency Power Conversion Confe. 1987.
144. C. Mullett and R. Hiramatsu, An Improved Parallel Control Circuit for Saturable Reactor Output Regulators in High Frequency Switched Mode Converters, Proc.IEEE Applied Power Electronics Conference 1986 .
145. E. B. G. Nijhof, The Series Resonant Power Supply and Its Control Circuitry, Seventh Internat.PCI Conf. and Fifth Internat. MOTORCON Conf. on Electronic Motion Control 1983.
146. B. Carsten, Distributed Power Systems of the Future Utilizing High Frequency Converters, Second International High Frequency Power Conversion Conference 1987.
147. C. S. Kerfoot, et al., RF Techniques Applied to High Frequency Converters, Second International High Frequency Power Conversion Conference 1987.
148. R. B. Ridley, et al., Multi-Loop Control for Quasi-Resonant Converters, Second International High Frequency Powe Conversion Conference 1987.
149. M. K. Kazimierczuk and K. Puczko, Feedback Control of Zero-Voltage-Switching Resonant DC/DC Converters, Second International High Frequency Power Conversion Conference 1987.
150. Los Angeles Times, April 4, 1987.
151. R. R. Rice, Space Station Power System Selection, 21st Intersociety Energy Conversion Engineering Conference 1986 Record.
152. T. Natarajan and S. R. Yadavalli, An Evaluation of Inverter Topologies for High Power Spacecraft, 21st Intersociety Energy Conversion Engineering Conference 1986 Record.
153. J. Hayden, Multi-Megawatt Power Distribution Considerations, 21st Intersociety Energy Conversion Engineering Conference 1986 Record.
154. W. G. Dunbar, et al., Design and Development Issues for Multi-Megawatt Spacecraft Power Systems, 21st Intersociety Energy Conversion Engineering Conference 1986 Record.
155. R. Charbonnier, Rochar-Electronique, January 1954.
156. C. W. T. McLyman, Flyback Converter Magnetics Design Software, KG Magnetics Inc., P. O. Box 3095, South Pasadena, California 91030-6095, 1987.


Disclaimer: Coil Winding Specialist, Inc., its affiliates, agents, and employees acting on behalf of Coil Winding Specialist, Inc. (collectively, CWS) disclaim any and all liability for any errors, inaccuracies or incompleteness contained herein or in any other disclosure relating to any product. CWS reserves the right at any time and from time to time to make changes to design, technical specification, or construction of any product listed on this website without notice. CWS also reserves the right to discontinue or limit sale of any product or service available throughout this website on temporary or permanently basis without notice. CWS highly recommends customers not to use its products offered on this website for inappropriate use and applications. Products should be tested for its suitability by prospective customers in all possible applications and conditions. Customers using or selling CWS products for use in other than recommended applications do so entirely at their own risk and agree to fully indemnify CWS from any damages arising or resulting from such use or sale.

Any Comments, Suggestions or Questions? Contact us! Email:
CWS 353 West Grove Ave. Orange, California 92865
Call us Toll Free: 1-800-679-3184 or 1 (800) 377-3244

Tel: (714) 279-9010, Fax: (714) 279-9482

Copyright 2008, Coil Winding Specialist, Inc
Updated July, 2013

CAGE Code: 5DME2(Registered with