How to simulate the Gm value of a transconductor

[abcyin]

it's about the transconductor simulation method.

when I have to acquire the Gm of a transconductance, first, I think the current variation should be simulated versus the input voltage variation, and then, maybe the derivation can applied to the plotted current curve, however, the current variation is on the PMOS loads of the transconductor, or on the resistor loads attached to the output node?

in a word, how can I get the Gm value of the transconductor by spectre simulation?

Thanks very much!

 

[elbadry]

I think Gm is simply i/v. So you apply an input signal and measure output current

 

[nxing]

Beware, there are two gm, one is for large signal and another one is for small signal. FOr the first one, you can run DC analysis to get it and for the second one, you can get it from the derivative method.

 

[abcyin]

THx!

but what I cannot be certain is which current should I detect, the current flowing through the PMOS load(for the simple CMOS differential pair), or do I have to attach some resistor to the output nodes and to get the current through the resistor?

 

[manish12]

ids Vs vgs

 

[anglee3015]

Beware, there are two gm, one is for large signal and another one is for small signal. FOr the first one, you can run DC analysis to get it and for the second one, you can get it from the derivative method.

i wonder how to get the gm for large signal, i try to run dc analysis ,but the result is not satisfied, i don't know how to measure the output current(if full-differential), should i connect a resistor or a voltage source for the load in my output ports?

the gm for the small signal, how to get the output current?