The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> difference between the calculation and simulation
https://designers-guide.org/forum/YaBB.pl?num=1180105951

Message started by e_fever_l on May 25^th, 2007, 8:12am

Title: difference between the calculation and simulation
Post by e_fever_l on May 25^th, 2007, 8:12am

hi,every one.I'm a beginner .I have a question,how big difference is acceptable between the calculation by hand and simulation by computer? 20%,50%,or more? such as calculate the current of a simple current mirror,use Id=1/2CoxUnW/L(Vgs-Vth)(Vgs-Vth)*(1+lambda*Vds).

I use tsmc BSIM3v3 0.6um model parameters to calculate and simulate,
and Un, Vth, TOX can be found from model
Cox can be get by Cox=εox/Tox
lambda from simulation.
but difference between results are very large than 50% or 100 %.
how to improve calculate???
thanks!

Title: Re: difference between the calculation and simulat
Post by carlgrace on May 30^th, 2007, 9:58am

50 or 100% is probably too big a difference. It is worth digging into this issue to see if there is a problem. Are your devices minimum size? Then short channel effects will play a big role here. How did you estimate lambda? Also, be careful that Vt and mu from the model is really what you think it is. There may be some adjustments to these parameters made by the simulator.

As a sanity check, you should look at the dc operating point of your current mirror and make sure everything is in deep saturation. Also, I recommend you start with long channel devices. Finally, make sure the vt that you get from the dc operating point is the one you expect.

Good luck!
Carl