How do I simulate the fully differential OTA with HSPICE

[aziz2010]

Hi,

I want to simulate with Hspice, the fully differential OTA such as, AC gain and phase margin. Generall speaking, fully differential amplifier has two input vin+/vin-, two outputs vout+/vout- .

I know the differential mode gain is Av=(vout+ - vout-)/(vin+ - vin-).

But How I do this in the Hspice.

The circuit that i simulate is depicted in the attached file.

Regards

 

[keith1200rs]

To get the DC operating point you can fix one input & do a DC sweep to find the DC input voltage which will give you a DC output voltage in the correct range. This will need to be a precise sweep if the gain is high.

Then you can fix the second input at that DC voltage so you are working in the linear range of the amplifier.

Then you can use either one or two AC sources (or AC parameters set on your DC sources). If you want to use two AC sources you need the phase to be opposite otherwise you will end up measuring the common mode gain, not differential gain. So, set one with a phase of zero degress the other with 180 degrees. An alternative may be to make one AC source a magnitude of -1.

The plot differential gain. You may find it easier to use an E source to turn the differential output to single ended for probing.

Keith.

 

[aziz2010]

Hi,
Think you Mr.keith for these Informations. But I can not understand some info.

So, can you please tell me can i use Hspice to measure the Diferential Gain of the Ota.

Thinks...

 

[keith1200rs]

I think you need to fix the DC voltage on M9 gate somewhere in the input operating range. The sweep M10 gate voltage [DC sweep] to find the point where the outputs are in their working range (i.e. linear and with high gain).

The fix the DC voltage of the source on the gate of M10 at the input voltage that gave the correct operating point.

You already have the voltage source on the gate of M9 with AC=1 so you now need to change the voltage source on the gate of M10 to AC=-1 or AC=1 with phase=180. The no an AC analysis.

Probe the voltage between your points h&g - that will be your differential gain.

Keith.