The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Measurements >> Other Measurements >> mismatch analysis
https://designers-guide.org/forum/YaBB.pl?num=1178089042

Message started by Pravinkumar on May 1^st, 2007, 11:57pm

Title: mismatch analysis
Post by Pravinkumar on May 1^st, 2007, 11:57pm

Hi all,

Am trying to find mismatch in current mirrors through DC analysis. I try to measure the current output by varying the three parameters w, l and vtho of the key transistorsm in the current mirror. I am able to sweep the two component parametrs w and l. Since vtho is a model parameter am not able to introduce mismatch to it.

Is there any way to declare it as a variable so that i can use it as another sweep parameter? Is it like the pdk should support the variations of the vtho? any suggestions??????

Am using spectre simulator.

Thanks in advance.

PS: No model files available to run monte Carlo or dcmatch analysis.

Title: Re: mismatch analysis
Post by ACWWong on May 2^nd, 2007, 8:04am

A simple (but crude) method would be to introduce a voltage source in the gate of the transistor to sweep the impact of vgs mismatch, which is correlated to threshold mismatch.

Title: Re: mismatch analysis
Post by John O Donovan on May 2^nd, 2007, 9:34am

Hi,

If you are using BSIM4 or BSIM3v3 models, then there are instance parameters for doing this in Spectre. You can use

delvto shift in zer0-bias threshold voltage vth0.
delk1 shift in body bias coefficient k1.
delnfct shift in subthreshold swing factor nfactor.

You can find descriptions in 'spectre -h'

Regards,
John

Title: Re: mismatch analysis
Post by Andrew Beckett on May 8^th, 2007, 1:41pm

There's also spectre's dcmatch analysis (see spectre -h dcmatch for more details). There are parameters in bsim4 and bsim3v3 (and some other models too), for example:

Code:

DC-mismatch dependent parameters:
816 mvtwl=0.0 V*m Threshold mismatch area dependence.
817 mvtwl2=0.0 V*m^1.5
Threshold mismatch area square dependence.
818 mvt0=0.0 V Threshold mismatch intercept.
819 mbewl=0.0 m Beta mismatch area dependence.
820 mbe0=0.0 Beta mismatch intercept.

Regards,

Andrew.