help! monte carlo problem

[ning_jing]

I have a problem with monte carlo problem. when i made monte carlo simulation, the variation is really huge. I don't think it is reasonable.
So I tried to use a very simple circuit to see how big is the varition. But the result seems unreasonable.

This is my circuit.


This is the result of monte carlo simulation.

the variation is over 100%

Does anyone can help me?

 

[keith1200rs]

Your test circuit seems a little strange. Why not pick a standard MOS characterisation curve that you can compare with the process data for which you have a graph?

Keith.

 

[ning_jing]

thanks. you are right. I should pick a standard MOS characterisation curve to compare with.

Here is the new test circuit :


the standard MOS characterisation curve:


and the monte carlo simulation result


the variation is really huge. I think something must be wrong....