In the paper selected methods of measuring the thermal resistance of an IGBT (Insulated Gate Bipolar Transistor) are presented and the accuracy of these methods is analysed. The analysis of the measurement error is performed and operating conditions of the considered device, at which each measurement method assures the least measuring error, are pointed out. Theoretical considerations are illustrated with some results of measurements and calculations.

In the paper the problem of modelling thermal properties of semiconductor devices with the use of compact models is presented. This class of models is defined and their development over the past dozens of years is described. Possibilities of modelling thermal phenomena both in discrete semiconductor devices, monolithic integrated circuits, power modules and selected electronic circuits are presented. The problem of the usefulness range of compact thermal models in the analysis of electronic elements and circuits is discussed on the basis of investigations performed in Gdynia Maritime University.

In the paper a new method of Random Telegraph Signal (RTS) noise identification is presented. The method is based on a standardized histogram of instantaneous noise values and processing by Gram-Charlier series. To find a device generating RTS noise by the presented method one should count the number of significant coefficients of the Gram-Charlier series. This would allow to recognize the type of noise. There is always one (first) significant coefficient (c0) representing Gaussian noise. If additional coefficients cr (where r > 0) appear it means that RTS noise (two-level as well as multiple-level) is detected. The coefficient representing the Gaussian component always has the highest value of all. The application of this method will be presented on the example of four devices, each with different noise (pure Gaussian noise signal, noise signal with two-level RTS noise, noise signal with three-level RTS noise and noise signal with not precisely visible occurrence of RTS noise).