information needed on the effect of backbias on mismatch

[rubink]

mismatch with backbias?

Anyone got any information on the effect of backbias on mismatch? I have a circuit which is varying more than expected. The foundry model doesn't take the effect of backbias on the mismatch, which could be part of the problem. I found a few papers which showed mismatch was worse with backbias, particularly in weak inversion but only limited data so far.

Any data or links gratefully received.

Thanks,
R

 

[rubink]

Re: mismatch with backbias?

Example paper

 

[rubink]

Some data from a different for 1.8V nmos from a different 0.18u process.
This confirms that increasing backbias increases backbias.

 

[erikl]

This confirms that increasing backbias increases backbias.

I guess nobody would ever question this! SCNR!
But jokes aside: you got your confirmation. Thanks for sharing the picture!