George Pau @ on
not converging to the default tolerance
I simulate a case where Vg = 0.4V and Vd = 0.0V. The rest of the parameters are the defaults value. I am using uncoupled mode with average onto a single slice. The solution did not converge to the default tolerance. Is there a reason why it is not converging? What should I do to improve the convergence?
1 Like 0 Dislike
Saumitra Raj Mehrotra @ on
Hi, The problem is inherent to ballistic simulations. At high gate bias the barrier becomes very small and there are not enough carriers near the source. To satisfy poisson it has to push the potential lower which leads to wrong results and non-convergence.
Please refer :nanohub.org/resources/6400/download/Supporting_Doc.ppt for further explanation for this. I hope it help.
Convergence can be improved by having a lowly doped region between source-channel and drain-channel.
thanks Saumitra
Reply Report abuse
Please login to answer the question.