<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
<body bgcolor="#FFFFFF">
<p>Dave wrote:
<blockquote TYPE=CITE><style></style>
<font face="Arial"><font size=-1>Wil
/ Dick / Tom and all.</font></font>
<blockquote
style="BORDER-LEFT: #000000 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px"> <b>As
many amateurs seem to suffer from FET blow-ups due to fatal voltage and/or
current peaks caused by proper TX loading only at the fundamental frequency
I wonder whether it would be wise to insert a diplexer between TX output
and tuning/loading system; a combination of a lowpass filter feeding the
aerial circuit and a highpass filter feeding a resistor that absorbs the
power in the harmonics. That might save FET's. It won't help
of course when the loading at the</b><b>fundamental frequency is wrong,
causing FET overload.</b><b>What do you think?</b> </blockquote>
<font face="Arial"><font size=-1>I tried the diplexer idea and it didn't
seem to have any advantage. As one of the country's greatest destroyers
of fets I can make a few suggestions. Recently I have not had any blow-ups
atall by following these rules.... (I'm touching wood!)</font></font> <font face="Arial"><font size=-1>.1.
Always match the antenna to 50 Ohms and operate the TX via an SWR bridge
so that you know the match is good before increasing power. The output
filter is designed for 50 Ohms and will transform other impedances into
possibly dangerous loads on the TX output stage.</font></font> <font face="Arial"><font size=-1>.2.
Check the drain waveforms on the fets with a 'scope. They should be clean,
usually looking like a square wave with a sine shaped peak. Lots of spikes
or "fuzz" is bad news and can be cured by fitting a Zobel network between
the two drains. i.e. across the primary of the output transformer. This
network consists of a 10 to 27 Ohm large wirewound resistor in series with
a 10nF pulse-rated C like the ones used in the LPF. Use 27Ohms with 75V
rail or above and 10 Ohms with 30 - 40V rails.</font></font> <font face="Arial"><font size=-1>.3.
Add a fast-acting protection circuit operating on overcurrent and bad SWR
which cuts the drive, relays are too slow! See my circuit or G0MRF's for
ideas. I am convinced that the trip has saved my 4 x IRFP450 rig on numerous
occasions.</font></font><font face="Arial"><font size=-1>If you use a 4013
as the phase-splitter / final divider stage, this can be forced to send
both outputs high by putting a high on the set and reset pins. After an
inverting driver stage this means both fets are turned off within a half-cycle.</font></font> <font face="Arial"><font size=-1>Good
luck and may your fets last a long time!</font></font> <font face="Arial"><font size=-1>Dave
G3YXM.</font></font></blockquote>
</body>
</html>