An aside re: IRF510s and so called fragility.
I have four amps that are not QRPL design and all have many hours in
linear service from 10W to 55W and one using parallel device that runs a
very comfortable 220W. To date and two of those amps are over a decade
old have not blown a device. The 220W monster (running on 6M) is even
running the poor irf510s at 30V so its in the range of tickling the dragons tail
for explosive failure. So far 3 contests and its performing well.
I also have 5 QRPL amps 4 of the 10W over a year old and running great
and now the 50W. I see no bad behavior from them. Should I I'll say what
Yet I see people that blast them apart at 5W all the time. So due to poor design,
long leads and other faults some of those are legendary for being unstable
so no real surprise. Others its a wonder on my part why or how.
Remember in most cases the IRF510 is wired in a circuit where if any reason
they go to full conduction they receive the full wrath of the power supply
(or battery). Typically that is not milliamps its many amps if not 10s of amps.
Any who the QRPL amp is class C so the bias should be set so there is NO
drain current (less than 10ma per device). It should not change any amount
(IE from 10 ma to 20ma from 10V to 20V).
ALSO the devices are called MOSFETs and the gate is indeed static
sensitive. I have killed a few in winter that way not even installed in
the project. it is possible to partially fail the gate and have it go the
rest of the way [sudden dead device] when running under power.
This also applies to BS170s, 2n7000 and other MOSFETs.
Amplifiers must always have a load, even if there is no RF input.
When testing an amp for the first time I usually put a big dummy load
at the output and a smaller on on the input as then the amp should be
stable. I bring up the voltage and bias very carefully. Maybe because
I treat them as if they are expensive and should not be abused.
Please reply on list so we can share.
No direct email, it goes to bit bucket due address harvesting in groups.IO