Skip to main content
  • Home
  • Happenings
  • Events
  • Thermal modeling of SiGe multifinger HBTs
Thermal modeling of SiGe multifinger HBTs

Thermal modeling of SiGe multifinger HBTs

Date10th Jun 2021

Time11:00 AM

Venue Google meet

PAST EVENT

Details

Silicon germanium heterojunction bipolar transistors (SiGe HBTs) are popularly used as power amplifiers in the RF frontend modules. In order to reduce the current crowding effect and improve the maximum oscillation frequency, it has been a common practice to partition the large emitter of HBTs into smaller fingers leading to multifinger devices. Although in such a structure, each emitter finger is electrically isolated by shallow trenches, they are thermally coupled through the common silicon substrate. Self-heating is a severe problem in modern SiGe HBTs where lateral dimensions are significantly scaled-down and additional trench isolations are present. In the case of multifinger transistors, thermal coupling from nearby fingers further increases the device temperature. Since the increase in temperature can affect the electrical behavior of the device, a thermal model is included in the compact model to accurately calculate the operating temperature.

In this seminar, an efficient thermal model for multifinger SiGe HBTs will be presented. First, the nonlinear nature of the heat flow problem in SiGe HBTs will be discussed. Then, the implementation of the existing models, which assume the validity of superposition of self-heating and thermal coupling effects, will be discussed. The proposed model, which avoids the use of superposition in this nonlinear problem, will be elaborated subsequently. The developed electrothermal model allows one to consider each finger in the multifinger transistor as an isolated decoupled transistor. The decoupling is done by properly partitioning the total heat flow volume between the fingers of the transistor. The model, when implemented in a circuit simulator, reduces the number of nodes required and hence improves the speed of simulation by 40%. Finally, the modeling results that are verified with TCAD simulations and measured data will be presented.

Speakers

Mr. Nidhin K (EE15D030)

Electrical Engineering