Dear Scott, LF Group,
To save long-winded explanations, I have drawn a
circuit diagram of my modified version of the G0MRF PA output circuit, with some
details of the output transformer. It is too big to send via the reflector, but
if anyone wants it, let me know and I will e-mail it direct.
My unit gives 400W out with 37V supply - for 300W
out with a 48V supply, I reckon the primary windings of the output transformer
should be changed from 4 to 6 turns. The value of the choke in the DC supply is
not critical - the iron dust cores can often be found in old PC SMPSUs. The most
significant change to the original circuit was to remove the decoupling
capacitor from the centre tap of the transformer primary, and increase the
inductance of the choke, converting the circuit into a "current feed" class D
configuration. This means the DC supply has a high impedance at 137kHz,
which prevents high transient currents into the low pass filter capacitors when
the MOSFETs switch. The ideal drain voltage waveform would be a half sine wave,
allthough in practice there is still considerable ringing at the switching
transitions, allthough much reduced. The value of the damping resistors and
capacitors is a compromise; reducing R and increasing C reduces the amount of HF
ringing, but increases dissipation in the resistors.
Cheers, Jim Moritz
73 de M0BMU
|