Hi all, I dug out a copy of an article by Nathan Sokal (Pres. of Design
Automation) that was published in Design Electronics on 20 Sept 1977 titled
"Class E can boost efficiency".
This turns out to be a very practical article with simple design formulae,
and a 'build it and play' circuit with example scope traces of current and
voltage waveforms. The test circuit is to run at 10 MHz but the detail
looks
eminently suited to 136kHz. Efficiencies of 82 to 95% are quoted for some
very ordinary transistors and FETs. There is no formula given for the RF
Choke but I think that can be scaled from the 10MHz value.
I am using a 'G3YXM' PA, slightly modified from the original. Main changes
are a pair of IRFP350's on each side of the push-pull (running at 56V) and
an extra LC section in the LPF.
With 9.2A draincurrent (= 515W input) I get 2.95A RF current into my 52 Ohm
dummyload (= 452W output). That means 87.8% efficiency.
Seems to be within the range of this famous 'Class E'.
The nice thing on Dave's PA (besides good efficiency) is that is a
straighforward and 'no nonsense' design, if you stick to scematics and
component list it cannot go wrong.
It is absolutely stable (I know an OM who blew more than a dozen FETs on an
oscillating PA) and survives open circuit (no problem, current goes down)
and even shortcircuit if you have a proper current-trip in the PSU.
73, Rik
Rik Strobbe ON7YD
[email protected]
Villadreef 14 B-3128 Baal BELGIUM (JO20IX)
|
