The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> K' value from Mosis test runs
https://designers-guide.org/forum/YaBB.pl?num=1176262820

Message started by eng on Apr 10^th, 2007, 8:40pm

Title: K' value from Mosis test runs
Post by eng on Apr 10^th, 2007, 8:40pm

Hi,
The K' value of IBM .18 um RF process is given at Mosis test run results in the following link
http://www.mosis.org/cgi-bin/cgiwrap/umosis/swp/params/ibm-018/t6cc_7rf_6lm_ml-params.txt

It is written over there that K'=Cox*u0/2 = 155.9 for NMOS transistor.
However when we calculate K' by using u0 (=270.6 & Cox~7.67 fF ) from spice model in same page it gives a different value.

1) Which K' should be used when doing hand calculations?

2) For a basic inverter the mobility difference could be compasated by making (W/L)p=2.3(W/L)n.
However here (if it's correct) K'n (=156) is almost 5 times larger than K'p(=32.5). Is it a normal ratio for .18um process?

Thanks in advance
eng

Title: Re: K' value from Mosis test runs
Post by ywguo on Apr 10^th, 2007, 8:58pm

Hi, eng,

BSIM3 model is very complex and composed of many empirical equations. So don't calculate K' value based on the BSIM3 model parameters directly.

I think that you will get almost exact K' value of IBM .18 um RF process is given at Mosis test run results if you plot a I-V curve and extract the K' value using SPICE or Spectre.

Best regards,
Yawei