I have seen quite a number of postings here with questions about the
performance, design, etc. of pi networks, both in front of and behind those
big shack warmers that we call PA's. O ver the years I have developed my own
methods of analyzing these networks, all based on standard circuit theory.
Its really a no-brainer - just a lot of calculations with vector voltages,
currents and impedances. You can really learn everything you want or need to
know about resonant pi-net transformers by just looking at the current and
power flows.
I have recently put it all together a nice little piece of software called
"PINET" (very original, wouldn't you say?) which I will be glad to email to
anyone. It allows you to input all of the relavent values to the modelled
pi-network on a simple, interactive screen, along with a starting frequency.
It will then sweep upwards in freq. and at every frequency step it reads out
all of the network currents, as well as all of the related specs. such as the
power output, efficiency, Q's, power loss in the coil, and attenuation of
all harmonics up to 8. It will locate the resonance automatically. You can
then watch the output and instantly analyze the effects of Q on harmonic
levels, instead of just relying on nebulous numbers that cicrculate. Its very
enlightening to just sit and run software "lab experiments" on this important
network, so if you are interested in this subject, I am sure you will enjoy.
Just send me your email address and I will shoot you a copy. (It is a
self-contained DOS .EXE application of about 250K which will run in
stand-alone DOS or as a Windows application.)
Eric von Valtier K8LV
It works equallty well as the input circuit or the output circuit, so it can
also analyze those input networks.
|