The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> Power Amplifier Compensation
https://designers-guide.org/forum/YaBB.pl?num=1312797439

Message started by aaron_do on Aug 8^th, 2011, 2:57am

Title: Power Amplifier Compensation
Post by aaron_do on Aug 8^th, 2011, 2:57am

Hi all,

I have heard that it is common in Power Amplifier design for the nonlinearity of one stage to cancel the non-idealities of another stage. For example, one stage may show gain expansion, while the next might show gain compression, and the two effects nullify resulting in linear gain.

Can I get anybody else's perspective on whether this is a good design practice? I'm thinking that although better efficiency may be achieved, the circuit's robustness to process variation may degrade.

thanks,
Aaron

Title: Re: Power Amplifier Compensation
Post by RFICDUDE on Aug 8^th, 2011, 5:29pm

Hi Arron,

It is feasible and possibly practical to cascade two stages where the IM3 distortion partially cancels out.

I was involved with some academic work looking at a two-stage class AB amplifier with cancellation for CMOS. The outcome of this was that both gm and Cgs nonlinearity matter in the cancellation approach, so both must be dealt with. After a lot of analysis, it was found that both can be dealt with and track over process, but the IM3 improvement over PVT was a minimum of 10dBc. This may not sound like much, but it can make a difference in efficiency when trying to meet a ACPR/ACLR spec.

An important point here is that you really must "need" the incremental improvement in order to put forth the effort it takes to insure the benefit (i.e. it is not a trivial slam dunk in linearity improvement). There is a R&D cost!

Title: Re: Power Amplifier Compensation
Post by aaron_do on Aug 11^th, 2011, 6:21pm

Thanks for sharing. 10 dB is quite respectable but I guess its quite borderline in terms of if its worth the time investment. It also can improve efficiency since you can design a more nonlinear amplifier.

Aaron

Title: Re: Power Amplifier Compensation
Post by RFICDUDE on Aug 12^th, 2011, 3:31am

Yes, the objective is better linearity and better power efficiency. If you are not achieving better power efficiency at a given level of distortion (dBc) then you really didn't linearize the amplifier.