Simulating Op Amp in Eldo Spice

[Nurahmad]

Hi,

I was simulating the conventional Op Amp (OTA) in Eldo Spice, in the AC analysis, when the amplitude of AC input signal getting higher the Op Amp open-loop gain also getting higher ( Gain and Bandwidth is increasing, Phase Margin is decreasing), why?

The open-loop gain should be independent from the amplitude of AC input voltage right?

 

[FvM]

A question without a full problem specification (simulation circuit and setup) is always prone to misunderstandings. AC analysis is small signal and doesn't care for magnitudes by nature. Your gain measurement expression would refer to voltage ratios of course.

But how about DC bias point setup in your simulation?

 

[Nurahmad]

Hi FvM:
Thank you for kindly replying.

I am simulating the op amp in open loop configuration. For the DC bias for my circuit I just simply implemented a DC current and for sure all CMOS in saturation region. I know AC analysis is small signal and doesn't care for magnitudes, but when I tried give some values(magnitude) for AC, different value had different open loop gain...
I don't understand which part got wrong...

 

[nitishn5]

I have seen this issue too.
From my understanding the ac analysis just measures the output voltage in dB and is not the 20log(Vout/Vin).

So when you change the magnitude of the input, the output also changes. Just measure the difference in dB of the output and input and it should be the same for different input ac magnitudes. This would be the gain of the op-amp.

 

[FvM]

From my understanding the ac analysis just measures the output voltage in dB and is not the 20log(Vout/Vin).

More exactly, it measures what you specify. In the SPICE variants I'm using, you can also plot voltage ratios respectively gains.

I am simulating the op amp in open loop configuration. For the DC bias for my circuit I just simply implemented a DC current and for sure all CMOS in saturation region.

That's why I asked for a full problem specification. To achieve a reasonable DC bias point, you'll usually need a DC feedback path in the simulation setup. Either if the DC bias point in your simulation is good or bad, it won't be affected by signal amplitude changes in AC analysis, but it would by any other circuit change,