ISSN 1068-3712, Russian Electrical Engineering, 2017, Vol. 88, No. 6, pp. 351–354. © Allerton Press, Inc., 2017.
Original Russian Text © S.S. Kapitonov, N.N. Bespalov, M.V. Il’in, I.V. Gulyaev, 2017, published in Elektrotekhnika, 2017, No. 6, pp. 31–34.
Selection of the Heating Method of Semiconductor Devices
during Their Testing in a High-Conductivity State
S. S. Kapitonov*, N. N. Bespalov, M. V. Il’in, and I. V. Gulyaev
Ogarev Mordovia State University, Saransk, Republic of Mordovia, 430005 Russia
Received May 16, 2016
Abstract—Possible heating methods of power semiconductor devices during their testing in a high-conduc-
tivity state are discussed. It is shown that the diffusion capacitance of the p–n junction has a significant effect
on measurements of device parameters. The effect of the diffusion capacitance on the results of testing of
power semiconductor devices at various shapes of heating-current pulses was investigated. Conclusions on
the possibility of using current pulses of various shapes for testing power semiconductor devices in the high-
conductivity state are drawn.
Keywords: power semiconductor devices, diodes, thyristors, power converter, high-conductivity state, test-
pulse shape, diffusion capacitance
It is necessary to ensure close to nominal operating
conditions of particular diodes and thyristors for the
improvement of the technical, economical, and oper-
ating characteristics of electrical-energy converters
based on power semiconductor devices (PSDs) used
in railway and subway transport, nuclear-power
plants, and the oil and gas industry. For this purpose,
it is necessary to know the parameters of every PSD.
This may be achieved by continuous control of the
parameters and culling of potentially unreliable
devices during their production and operation.
The most important parameters of PSDs in a high-
conductivity state (HCS) are current–voltage curves
(CVCs), “junction–case” steady-state thermal resis-
, and “junction–case” Z
thermal resistance. A technique for determination of
CVCs and thermal parameters was developed in [1, 2].
With the use of this technique and new computing
technologies from National Instruments, the whole
testing cycle did not exceed 3–5 min for every PSD.
This technique is based on heating a testing device
(TD) with current pulses to a certain temperature of its
semiconductor structure (SS) and subsequent rapid
cooling. Then, the parameters and temperature
dependences for the calculation of R
measured. The measurement accuracy and complexity
of a circuit-design realization of the device for the
determination of the thermal parameters of the PSD
are substantially dependent on the shape and duration
of the heating-current pulses. Hence, the optimum
method of heating of the TD should be selected. The
article is devoted to an investigation of the effect of the
heating-current shape on the results of thermal
parameters’ measurement of the PSD, obtained with
the use of the technique [1, 2].
In accordance with this technique, thermal resis-
is defined as follows:
is the average power loss, Т
is the tem-
perature of the SS, and T
is the temperature of the
case of the PSD.
There are two basic approaches to the formation of
heating current: heating by direct current and heating
by impulse current [3, 4].
The first method makes it easier to calculate the
heating power and, accordingly, to simplify the mea-
suring part of the test equipment. Power loss P
calculated by determining current I
the PSD and voltage U
in the high-conductivity state
when thermal equilibrium between the TD and the
cooler is reached:
Equation (2) is valid for a direct heating current
only. The main difficulties in the construction of a
high-power direct current source are associated with
high requirements for the level of heating current pul-
sations and the possibility of adjusting its level.
Unstabilized pulsed-current sources are much sim-
pler in circuit design. When using pulsed-current
sources, it is impossible to determine P
by Eq. (2),
totAV hc hc