The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> Figure of merit for a technology for PA design
https://designers-guide.org/forum/YaBB.pl?num=1346978381

Message started by aaron_do on Sep 6^th, 2012, 5:39pm

Title: Figure of merit for a technology for PA design
Post by aaron_do on Sep 6^th, 2012, 5:39pm

Hi all,

I was wondering if there is a good figure of merit to determine how good a technology is for PA design. I am interested in both linear PAs and PAs for constant envelope modulation schemes. Any input is welcome.

thanks,
Aaron

Title: Re: Figure of merit for a technology for PA design
Post by RFICDUDE on Sep 11^th, 2012, 3:42am

I am surprised no one has responded to this, so here's my 2-cents.

It is really difficult to come up with a meaningful FOM to compare different PA technologies because device performance is taken to the extreme and PA application requirements (gain, linearity, efficiency, cost, reliability, packaging, matching, supply voltage, etc.) vary wildly.

In general, a good PA technology for power efficiency will permit very high currents with a very small voltage drop at the output (near short) and very high open circuit voltages (no or little current at high supply). Of course the device has to switch back and forth between high current and high supply at the highest frequency of interest, so ft for the device size necessary to deliver the required output power is useful to know. Next might come passives (caps/inductors/tlines), but that depends a lot on whether external matching is going to be used or not.

Output power and linearity swept over power might be some indication of the technology performance, but these tend to be optimized for specific applications using load pull characterization.

Reliability and ruggedness are also important. PA for mobile applications have to tolerate fairly extreme changes in VSWR.

It would be interesting to hear what others have to say about picking PA technologies (in general).

Title: Re: Figure of merit for a technology for PA design
Post by aaron_do on Sep 11^th, 2012, 11:26pm

Hi RFICDUDE,

thanks for the input. I think if we talk about a couple of very specific examples, there should be good FOMs.

For example, suppose we have constant envelope modulation, and we aim to get the best possible efficiency while disregarding linearity. There must be a simple relationship between efficiency, fT, VDD, breakdown voltage etc which can be used to get a rough idea of how good a process is.

Or for a linear PA, you could maybe use the IIP3 or something like that.

Aaron