linearity of true class AB and pseudo class AB opamps - which one is better and why?

[Debdut]

As the title suggests, please help me understand this. I am not able to find suitable material over the internet on this topic.
In design, I am finding that the IIP3 of a true class AB opamp is worser than a pseudo class AB stage for similar power dissipation.

 

[Dominik Przyborowski]

In design, I am finding that the IIP3 of a true class AB opamp is worser than a pseudo class AB stage for similar power dissipation.

Linearity is quite a topic itself, while your statement is tricky to judge without investigation of your results.
What do you mean by true and pseudo class-AB stage? Is pseudo one, the class A bias structure with AC coupling for signal?

In general, input differential pair has non-linear transconductance, current mirror has limited linear range and conversion to output voltage is done on nonlinear current source. What makes OPAMP linear, is a negative feedback configuration, suppressing input signal swing by gain of amplifier, so changes of operating points of each device inside opamp are so small that we considering their higher-order terms as negligible.

And now is the first point. If we consider class-AB biasing (lets look on Monticelli's as is very common), for quiescent conditions opamp achieves minimum gain (first stage load is a sources of floating current source structure, so OL gain is like single stage only), which is boosted at high output load case (floating current source become high impedance for signal and gain can be doubled in dB).
So, if we suppose that the main source of nonlinearity is input differential pair, class-AB opamp has an extra nonlinear term related to gain variability with signal.
However, it can be different story for different class-AB solutions.

 

[FvM]

IIP3 isn't specifically a linearity figure, rather specifying large signal behaviour, e.g. undistorted voltage swing.

I also doubt that "pseudo class AB" is a commonly understood technical term or even specifying an unique topology, there are only a few papers dealing with it.

 

[D.A.(Tony)Stewart]

CMOS drivers seem to be given unique names and compared with BJT drivers may result in some confusion will need references to relate to some of us "old dogs" to learn new tricks.

Both enhanced FET and bipolar currents have 2nd behaviors yet we think of 3rd order dominant distortion in transistors which is only due to clipping. IIP3 which mainly applied to RF amplifiers for linear regression from the 3rd order intercept with 1f at high output in order to limit high power distortion. The degree of Class A is mainly due to the rise in hFE with rising current such that inverting the signal and comparing the peaks clearly shows the amplitude difference in % which closely matches %THD.

In Class B for PNP and NPN in addition to crossover has distortion is due to the mismatch of hFE in these drivers and thus in AB drivers Rout=Rin/hFE differences are often reduced by adding a fixed power Re to each emitter to more normalize the differences. This also improves risks of thermal runaway when many same devices are put in parallel with NTC thermal effects. Of course Class AB crossover distortion is minimized with a minimal constant current to prevent the starving of device current which results in load regulated distortion and it's effect on output impedance during polarity transitions which is most noticeable at low signal levels.

I have yet to see IIP3 applied to Op Amps but if it is linear with output current near max power then it would be useful. Some Op Amps have UHF BW and others have 0.5A drive capability.

 

[D.A.(Tony)Stewart]

Is this a valid comparison using an IIP3 figure of merit?

https://static.aminer.org/pdf/PDF/000/424/267/a_pseudo_class_ab_telescopic_cascode_operational_amplifier.pdf

"If you know the IIP3 vs. OIP3 point on an output vs. input graph for your power amplifier, you can determine the appropriate input signal level to use."

Can you back up your assertion with assumptions and details?

 

[dick_freebird]

To me the main thing is, you can Vbe match NPN to PNP, both vertical get within 2:1 fT and a true AB is pretty plain work to tweak in. I used such for ICs and old timey transistor databooks used to have matched pair power devices at various f, I, V

You can't get MOS in a CMOS flow to match P to N VT like that, and power device mV/decade is worse for MOS than BJT for more hold-the-bias fun. So "pseudo" will have to clean up those issues I expect.

 

[D.A.(Tony)Stewart]

pseudo-class-AB = telescopic-cascode op-amp
This was proposed to increase GBW and capacitance current driving capability with higher output poles in IEEE circa 2004 with later citations.
e.g. https://www.researchgate.net/public...S-AB_TELESCOPIC_OPERATIONAL_AMPLIFIER/figures

Push-pull "Source Follower" Class AB cannot be Rail-Rail (R2R) output since Vg just be |greater| than Vd.
It will rely heavily on current mirrors for CC sink.
Ref: https://aicdesign.org/wp-content/uploads/2018/08/lecture21-140624.pdf

B vs AB Transfer function​

But push-pull "Common Souce" can be R2R output but open loop will have higher output Zo.

If you are behind an IEEE paywall search for titles in Researchgate.com

e.g. https://www.researchgate.net/public...rs_The_Development_Route_and_State-of-the-Art