This post by Ain Rehman. Looking through the literature a HEMT model that seems to be understandable and useful for hand calculations seems to be the Chalmers model. It is intuitive and resembles the familiar MOSFET type models that designers have been used to. Signal Processing Group Inc.is in the process of developing some useful relationships from this model that can be used for simple first order calculations for design. These relationships will be tested thoroughly and all assumptions validated. At the conclusion of this exercise a set of equations will be presented that will allow hand calculations to be made, to within a reasonable degree of accuracy as a starting point of a HEMT ( or pHEMT ) circuit design. Please visit our website for other technical information and our service and product offerings.

# Monthly Archives: June 2016

# HEMT/pHEMT absolute basics

HEMTs and pHEMTS are important new semiconductor devices that have a high performance profile. They are also low noise. GAN HEMTs are being used in high power RFMW power amplifiers. pHemts are also being used in high frequency and low noise amplifiers. The operation of the HEMT is not immediately intuitive. This post seeks to provide a basic description of the backbone of the HEMT operation. Following posts will explore more details of the HEMTs/pHemts. Interested parties may access a detailed description from the Signal Processing Group Inc., website under the “complemntary items ” menu.

# Bipolar transistor noise: The equivalent noise resistance

A large number of very detailed noise analyses are available for bipolar noise and some of them are very detailed indeed. However, for a quick design assessment of noise it is more useful to use the bipolar equivalent noise resistance. This is given by : RN = rb + 1/2gm. Here rb is the base series resistance and gm is the transconductance. What this expression is telling us is that to minimize noise use a stripe geometry for the bipolar to reduce rb and increase the current to increase gm. However, in the case of variable gain amplifiers this is a challenge to say the least. Most variable gain bipolar amplifiers use a varying gm. So as the current changes in the device the nose resistance also changes. A compromise has to be made.