This article describes the thermal resistance “junction-to-case” measurement results for GaN high-electron mobility transistors (HEMTs). The measurements were taken by using the apparatus which includes two methods of thermal resistance… Click to show full abstract
This article describes the thermal resistance “junction-to-case” measurement results for GaN high-electron mobility transistors (HEMTs). The measurements were taken by using the apparatus which includes two methods of thermal resistance measurements. For the first one that uses standard MIL-STD-750-3, a sequence of heating current pulses passes through the transistor channel and then the junction temperature is measured. The second method uses modulated heating power and measures the response which is a variable component of the junction temperature. To exclude the influence of the heating pulses duration, which is typical for the standard method, preliminary measurements of the transient response of thermal impedance were taken. The analysis of this characteristic allows determining the optimal values of the pulse duration. To reduce the influence of the delay time caused by transient electrical processes in the transistor when it switches from heating mode to the measurement of the thermal sensitive parameter (TSP) mode, the TSP signal was extrapolated to the end of the heating pulse. A comparative analysis shows that the measurement results obtained by the standard and modulation methods differ by less than 2%. The results of a study of the influence of the heating current pulses magnitude on the thermal resistance measurements results are presented. It was found that, as heating current increased, the thermal resistance measured values increased, which indicates the nonlinear nature of the dependence of the temperature in the transistor channel on the power dissipated in it.
               
Click one of the above tabs to view related content.