Question about class AB opamp

[andover]

class-ab op-amp

Hi guys,
I have a question about two transistor class AB opmap. For small signal analysis, it is linear around Q point so that vout=vin*gain. However, for Class AB opamp, the input ranges from rail to rail, and due to square law, Id is proportional to (vin-vth)^2, which introduces distortion in the output. How do the opamp suppose to work?

Thanks

Andover

 

[jianjing526]

So where is your puzzle?
AB is just one bias for output stages. Vout=Vin*gain exits in all networks. Id proportional to (vin-vth)^2 for saturation mosfet. Is there any conflicts?

 

[andover]

To make myself clear, I believe in small signal analysis, it assumes the input signal should be small enough around Q point, so that linear relationship between input and output voltage exists. However, the input of AB opamp could be rail to rail, which violates the presumption for the small signal, and results expanding curve and distortion at the output.

my question is how the class AB opamp has linear relationship between input and output?

Thanks

 

[FvM]

However, the input of AB opamp could be rail to rail, which violates the presumption for the small signal.

No.
Small signal analysis doesn't require linearity, only a continuous transfer function. The small-signal gain is generally different for each operation point, that's neither specific for class AB output stages nor rail-to-rail op. Of course, you can expect larger changes near the rails. Cause too many parameters are involved (supply, voltage, load impedance) you most likely won't find exact specifications in datasheets. But most OPs have a distortion specification for a typical configuration, that allows to estimate the closed loop behaviour.

 

[LvW]

No.
Small signal analysis doesn't require linearity, only a continuous transfer function.

Hi FvM - I like to jump into the discussion because I need some clarification about your statement above. I think the following was completely correct, isn´t it ?

quote andover:
the input signal should be small enough around Q point, so that linear relationship between input and output voltage exists.

With other words: small signal analysis assumes linearity around the actual bias point.
That´s what simulation programs are doing when they perform ac analyses for opamps with input voltages up to kilovolt ranges leading to the correct transfer function.

 

[refugee]

However, the input of AB opamp could be rail to rail, which violates the presumption for the small signal.

No.
Small signal analysis doesn't require linearity, only a continuous transfer function. The small-signal gain is generally different for each operation point, that's neither specific for class AB output stages nor rail-to-rail op. Of course, you can expect larger changes near the rails. Cause too many parameters are involved (supply, voltage, load impedance) you most likely won't find exact specifications in datasheets. But most OPs have a distortion specification for a typical configuration, that allows to estimate the closed loop behaviour.

I messed up about "Small signal analysis doesn't require linearity, only a continuous transfer function." Could you please give me some details about it?

 

[pdxicdesigner]

This is a late answer (well I joined not too long ago).
In class AB you are right that the transistor exhibits lots of non-linearity over its operation range. Since gm is related to current you can deduct that as the output signal of a class AB stage goes from small to big the gm changes alot also. Think of how much the current through say the PMOS device (NMOS off) would change with swing over a fixed load. Hence the small signal analysis would change alot as well over the range.

So with that said, many audio amps require 3 stages to help reduce the distortion(and of course the feedback reduces the distortion as well).