The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> question on simulating gm of amp
https://designers-guide.org/forum/YaBB.pl?num=1162905814

Message started by glacieryy on Nov 7^th, 2006, 5:23am

Title: question on simulating gm of amp
Post by glacieryy on Nov 7^th, 2006, 5:23am

Hi all.
Could anyone tell me how to simulate the gm of amplifier ?

Thanks a lot.

Glacier

Title: Re: question on simulating gm of amp
Post by egibson873 on Nov 30^th, 2006, 4:12pm

gm, or the transconductance is defined as the change in the output current with respect to the change in the input voltage. So, if you run an AC simulation you can get the desired results. Place the AC source at the input. Measure the change in output current divided by the change of input voltage and you will get your gm.

Title: Re: question on simulating gm of amp
Post by Croaker on Nov 30^th, 2006, 8:15pm

To measure the Gm of the amp, you need to see how the current changes when the output is AC shorted. Think of the way an amplifier may be represented (a controlled current source in parallel with an output resistance). AC shorting the output allows you to see the current the controlled source is generating, while eliminating the output resistance, Ro, from the picture. The DC bias conditions must remain intact though.

Let's say you have a common-source amplifier. At the output, put a really big cap to ground. The big cap makes the output node shorted to ground at higher frequencies, but it's open at DC, so it doesn't change the DC operating point.

Run your AC analysis and Gm=iout/vin, where iout is the current through the big cap.

Title: Re: question on simulating gm of amp
Post by avlsi on Dec 29^th, 2006, 12:53am

In Cadence calculator, you have the SKILL COMMAND OP(). You can use this to plot the gm of the MOSFET.

Title: Re: question on simulating gm of amp
Post by Croaker on Jan 8^th, 2007, 8:50pm

Op will work for the operating point gm of the MOSFET. As discussed in another posting, if you want to plot gm vs. something you need to do your dc sweep and use a command like 'save m1:all'

The gm of a MOSFET is only going to be the same as the gm of the amplifier in certain cases.

Title: Re: question on simulating gm of amp
Post by topquark on Jan 9^th, 2007, 1:09am

Croaker,

What exactly do we do when you say "..use a command like 'save m1:all' .." ??

Could you please clarify is it in the results browser or icfb command window?

Thanks,
Gau

Title: Re: question on simulating gm of amp
Post by Ken Kundert on Jan 9^th, 2007, 4:11pm

save m1:all goes into the spectre netlist.

Title: Re: question on simulating gm of amp
Post by Croaker on Jan 11^th, 2007, 11:31am

Ken Kundert wrote on Jan 9^th, 2007, 4:11pm:

save m1:all goes into the spectre netlist.

Right. For other simulators like HSPICE, you can plot parameters such as gm by using a template element, as described in the manual (Ch.4).

eg.
.probe LX7(m1)