V.I. Smirnov – Dr. Sc. (Eng.), Professor, Department «Electronic Instrumentation Design and Technology», Ulyanovsk State Technical University
V.A. Sergeev – Dr. Sc. (Eng.), Associate Professor, Director of Ulyanovsk branch of Kotel'nikov IRE of RAS
A.A. Gavrikov – Ph. D. (Eng.), Senior Research Scientist, Ulyanovsk branch of Kotel'nikov IRE of RAS
A.M. Shorin – Post-graduate Student, Department «Electronic Instrumentation Design and Technology»,
Ulyanovsk State Technical University
High power MOSFETs and IGBT transistors form the core of power electronics. They are able to switch hundreds-ampere currents, and their dissipating power can be up to 1 kW. Such operating modes can cause dramatic overheating of the transistor die with subsequent negative effects. Therefore, quality controlling of heat removal from an active area (p-n junction) to a heat sink and then to ambient is very critical. There are several methods for measuring thermal impedance of high power transistors. The standard JESD24 3 method of-fers to pass a pulse sequence of heating current through a device under test and measure the temperature of a p-n-junction before and after the pulse sequence passing. The method has low accuracy since the measurement results are affected by a number of factors that are difficult to take into account because of their uncertainty. Another method uses transistor self-heating by means of heating pulse se-quence with different duty factor. Compared with the previous method, it allows one to determine more accurately the junction-to-case thermal impedance, but its implementation requires to maintain the case temperature constant since the calculation of the junction-to-case thermal impedance module is based on measuring the difference between current junction temperature, and initial temperature. A method stated in the JESD51-14 standard provides measuring thermal resistance components of semiconductor devices. The principle of the method is as follows. Stepped varying power, P, is dissipated in the device and a response on this influence is measured. That is a change of p-n-junction temperature, while heating the device before it reaches a steady state. The p-n-junction temperature, as in the previous methods, is determined indirectly by measuring a TSP. Forward voltage drop of the p-n-junction at low measuring current is preferably used as a TSP.
The paper proposes a new genuine method for measurement of thermal impedance of powerful MOSFET and IGBT transistors. The ad-vantage of this method is in substantial reduction of case temperature influence during thermal impedance measurement. Heating power is modulated by passing current pulses through the device, duration, τ, of which varies harmonically. Modulation of heating power causes corresponding changes of the p-n junction temperature that are phase-shifted with respect to power. The junction temperature is deter-mined by measuring the TSP between heating pulses, at the time low measuring current is passing through the device under test. With known amplitude values of the variable components of heating power and junction temperature, it is possible to determine the module and phase of the thermal impedance of the device under test. This method of calculating the amplitude of the temperature variable com-ponent allows one to determine T1(f) with high accuracy. That is because of a large sample number at the discrete Fourier transform (N ≈ 1500). As a result, measurements can be taken at relatively small heating currents since the oscillation amplitude of the junction temperature at 1°C is sufficient to determine Zth(f) with adequate accuracy. The study showed that, unlike standard methods of measur-ing thermal impedance in which transistors are heated by a single pulse or a sequence of pulses of heating current, the influence of a case temperature trend on measurement results is significantly reduced in the modulation method. In this case, thermal impedance can be measured with relatively small heating currents since the amplitude of the variable component of the junction temperature deter-mined by the discrete Fourier transform with a large sample size can be less than 1° C to ensure a sufficiently high accuracy of measur-ing thermal impedance.
- Thermal Impedance Measurements for Vertical Power MOSFETs (Delta Source-Drain Voltage Method). JEDEC JESD24 3 standard.
- Power MOSFETs. JEDEC JESD24 standard.
- Thermal Impedance Measurement for Insulated Gate Bipolar Transistors – (Delta VCE(on) Method). JEDEC JESD24-12 standard.
- Application Manual Power Semiconductors // SEMIKRON International GmbH. 2015.
- Current Rating of Power Semiconductors // Vishay Siliconix. Application Note AN-949. 2010.
- Transient Dual Interface Test Method for the Measurement of the Thermal Resistance Junction to Case of Semiconductor Devices with Heat Flow through a Single Path. JEDEC standard JESD51-14.
- Szekely V., Tran Van Bien. Fine structure of heat flow path in semiconductor devices: a measurement and identification method // Solid-State Electronics. 1988. V. 31. P. 1363−1368.
- Rencz M., Szekely V., Morelli A., Villa C. Determining partial thermal resistances with transient measurements and using the method to detect die attach discontinuities // SEMITHERM. 2002. P. 15−20.
- T3Ster - Thermal Transient Tester - Technical information // Mentor Graphics. URL = http://www.flotrend.com.tw/products/st3/t3ster/data/T3Ster_b.pdf.
- Patent RF № 2572794. Sposob izmereniya teplovogo soprotivleniya perexod–korpus moshhny'x MDP-tranzistorov / Smirnov V.I., Sergeev V.A.,Gavrikov A.A. Vy'dan 05.11.2014. Opubl. 20.01.2016. Byul. 2.
- Smirnov V.I., Sergeev V.A., Gavrikov A.A. Spektral'ny'j i vremennoj metody' izmereniya teplovogo soprotivleniya poluprovodnikovy'x priborov // Promy'shlenny'e ASU i kontrollery'. 2014. № 10. S. 58−63.
- Smirnov V., Sergeev V., Gavrikov A. Apparatus for Measurement of Thermal Impedance of High-Power Light-Emitting Diodes and LED Assemblies // IEEE Transactions on Electron Deevices. June 2016. V. 63. № 6. P. 2431−2435.