The Proposals

  • Riko Radojcic


Another few weeks go by. With Dave’s help, I have managed to extract a model for the device I was testing in the lab. The “model” is really a set of complex equations that describe the behavior of a given type of a transistor, with something like 100 different coefficients that have to be tuned so that the model predictions match the measured data. The format of the model, i.e., the basic equations without these coefficients, is standardized – fortunately. Otherwise this would have been really hard. We are using BSIM models (*TBB 1.7). Then I had to run a set of SPICE simulations – these are circuit-level computer simulations – to ensure that a selected set of the so-called Figure-of-Merit circuits behave well in simulations using the models I extracted. We found some problems – I am told that this is normal and not because my models were bad – and I had to go back and tune some of the coefficients. I must say, I enjoyed all this very much. Somehow very elegant – this describing the complex physics of a transistor that I characterized in the lab through a set of compact mathematical equations. And, I can do it! Dave likes to say that there are maybe 100 people on the planet who do this model extraction and fitting – and I have just become one of them. Yaaas!

Copyright information

© Springer International Publishing AG, part of Springer Nature 2019

Authors and Affiliations

  • Riko Radojcic
    • 1
  1. 1.San DiegoUSA

Personalised recommendations