Semiconductor Rectifier Research Articles

A high-current metal—Semiconductor rectifier

Schottky-barrier diodes have been in use for some time, but, in the main, only for small-signal applications. Now a large area (17K mils <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ) Schottky-barrier diode has been successfully fabricated for use as a high-current rectifier. The device has a rated Io of 50 amperes. Measured values of forward voltage drop at 50 amperes (dc) are typically less than 500 mV. The inherent speed characteristics of this majority carrier device were indicated in the measurement of rectification efficiency versus frequency. The efficiency of the device remained unchanged through 500 kHz. A brief theoretical review of metal-semiconductor junctions is presented and applied to the empirical characteristics of the rectifier. Chip geometry, fabrication techniques, and complete device characteristics are discussed. Specific device applications are covered with emphasis on performance and limitations.

Discussion on “Transductor-assisted tap changing on a 50 c/s locomotive with semiconductor rectifier”

Discussion on “Transductor-assisted tap changing on a 50 c/s locomotive with semiconductor rectifier”

The physical principles of semiconductor power rectifier operation

Abstract The efficient conversion of electrical power from AC to DC is one of the most important problems in electrical technology. The ideal rectifier will present a short circuit path to current flowing in one direction, and an open circuit in the opposite direction. All practical rectifiers have a finite resistance in the conducting or forward direction and measurable leakage in the blocking or reverse direction; further, they all cease to block at all beyond a certain voltage. There is always a certain temperature for each type of rectifier above which its efficiency drops rapidly. The most important features of any rectifier are therefore:

Transductor-assisted tap changing on a 50 c/s locomotive with semiconductor rectifier

The principles and advantages of transductor-assisted tap changing are described and the design of a particular a.c. locomotive incorporating this feature is explained.The particular type of transductor circuit employed is examined in detail, together with the reasons for its selection, its method of operation and its performance in practice.The advantages to locomotive performance of controlling the tap changer so as to automatically regulate the traction current are discussed and the performance of a prototype system is described.

Proposed method of loss measurement for semiconductor rectifier equipments

IT HAS long been recognized that the efficiency of a large rectifier unit cannot be accurately determined by input-output power measurements. Errors of measurement may be as great or greater than the losses in high efficiency equipments. These errors are aggravated to an unknown degree by the distortion of current and voltage waveshapes inherent in rectifier operation.

Loss measurement in semiconductor rectifier equipment

IT HAS long been recognized that the efficiency of a large rectifier unit cannot be determined accurately by input-output power measurements. The “segregated loss” method has been written into rectifier standards because it reduces greatly the errors inherent in input — output measurements of high-efficiency equipments. However, when determining the losses of the semiconductor rectifier itself, we must take into account the difference of wave shape of the cell currents in normal operation and in the test circuit. The high incremental resistance of the cells precludes the direct equation of losses for the same average currents at different form factors.

Transient current decay in paralleled rectifiers

THE ADVENT of silicon semiconductor rectifier applications to large electrochemical lines has developed application problems heretofore not significant. The majority of existing potlines have mercury-arc rectifiers as a source of d-c power. In general a single a-c breaker will feed the entire potline rectifier group. Load tests in the field have been made which show that the most satisfactory method of removing a rectifier line from service is to trip the master a-c breaker, removing a-c power simultaneously from all units. The potline current will then decay through the mercury-arc tubes and the rectifier transformers, dividing between paralleled units.

Transient decay of current through paralleled mercury-arc and silicon rectifiers

THE ADVENT of silicon semiconductor rectifier applications to large potlines has brought about application problems heretofore not significant. Many of the users of large amounts of d-c power wish to try a large power silicon rectifier in actual service before the purchase of a complete potline. Since the greater majority of the existing potlines have mercury-arc rectifiers providing the d-c power, it is only natural to put a silicon rectifier on a potline in parallel with the mercury-arc rectifiers. Fig. 1 is a single-line diagram of a modern electrochemical potline. The normal method of de-energizing the line is to trip the master a-c circuit breaker feeding all units. The problem of magnitude of potline decay current which must be handled by a single silicon rectifier in parallel with several mercury-arc rectifiers under these conditions requires analysis.

Frequency characteristics of a semiconductor rectifier at voltages greater than kT/q

In this paper, the frequency characteristics of a half-wave p-n junction rectifier with a resistance load are analyzed for applied voltages greater than kT/q. Beginning with the derivation of an implicit integral equation which describes the large-signal operation of this circuit, the effect of frequency on the rectification efficiency is then obtained from an iterative numerical solution of this equation using an IBM 650 computer. Contrary to the small-signal theory developed by Kalashnikov and Penin, it is found that at higher applied voltages the half-power frequency is very dependent on voltage as well as load resistance, carrier lifetime and reverse saturation current. The relationships between the half-power frequency and these parameters are shown graphically and the results are compared to the small-signal theory for small applied voltages.

A method for testing and establishing the rating of semi-conductor rectifiers under dynamic conditions

Owing to the combination of high current density and low forward voltage drop in the rectifying junction of a semi-conductor rectifier, the physical size and thermal capacity for a given rating are low, so that the current on overload can be much greater than for other types of rectifier. Consequently, the junction-temperature rise on overload is extremely rapid, and in some cases considerable temperature fluctuations occur at mains frequency. It is therefore necessary to be able to monitor the junction temperature dynamically (i.e. under working conditions at mains frequency) and to observe the rise of temperature with time. A further necessity for dynamic testing occurs since the measurement of the forward characteristic at high current by d.c. methods can result in excessive heating and possible failure. In the apparatus to be described the separation of forward and reverse half-cycles is secured by means of a synchronous commutator, permitting independent measurement of forward and reverse current and voltage at mains frequency, and hence continuous indication of junction temperature. Operation of the rectifier at full rating without expenditure of the full power necessary to accomplish this by direct methods is also possible. This is particularly useful in life testing large numbers of semi-conductor rectifiers. Limited experience indicates that this method of testing provides operating conditions which are in all ways identical to those which a device experiences under fullload conditions

On the Transient Behavior of Semiconductor Rectifiers

If the forward current through a p-n junction suddenly ceases, an emf appears due to nonequilibrium carrier concentrations at the junction barrier. Measurements of the barrier emf have been made with a wide range of injection currents and yield current vs barrier voltage characteristic curves. The open circuit case of voltage decay across a p-n junction in the range V≫kT/q is discussed. The voltage decay is observed to be linear with time with a slope kT/qτ. It follows from an analysis based on a simple model that the excess minority carrier concentrations decrease exponentially with the time constant τ. The floating emitter and collector voltages, VE(t) and VC(t), of p-n-p transistors are observed to decay like the open circuit voltage across a p-n junction, and it is noted that VE(t) ≈ VC(t) in the range V≫kT/q. The transient response of a semiconductor rectifier connected in parallel with a large external capacitance is analyzed. The response calculated for large negative excursions of voltage gives a simple relation by which the saturation current may be determined. Measurements are presented and discussed.