In article ,
Newbie Ham wrote:

1) Radiated signal from the coax leaking into the inverter.
2) Radiated signal passing into the inverter via the shared positive or
negative feeds.
3) Some weird ground loop issue.

I have no ideas as to how to diagnose this and trouble shoot it without
having to replace fets everytime. And that's a big job.

Any thoughts would be greatly appreciated.

Could be any of the three paths, or more than one at once... conducted
vs. radiated is an interesting question, but perhaps a bit academic.
All of your equipment is going to be in the near field of the
antenna/ground system, I think, and so you're going to be ending up
with RF pretty much everywhere.

If I had to guess, I'd guess that your inverter may have some sort of
half-bridge / push-pull switching circuitry. The RF getting into the
system is switching both sides of the push-pull drive into conduction
at the same time (a few volts of RF on the gates of the FETs could do
it) and shorting the supply through the FETs.

The fact that the RF was able to blow the FETs with the inverter
switched off suggests to me that it may be a fairly direct, local
pickup of RF onto the gates which did the damage. With the inverter
switched off, its controller wouldn't have been driving the FET(s) at
all.

As to how to prevent it from happening again... my guess is that your
best bet is going to be to add heavy-duty RF filtering to both the DC
inputs, and the AC load, and make the invert case as RF-tight as
possible.

I'm not sure whether it'd be possible to add snubbers, ferrites, etc.
on the FET gate leads themselves to keep the RF from triggering
them... the inverter design may require driving the FETs quite fast to
achieve proper control of the voltage. Figuring this out would
require a schematic and some study of the design.

It might be worth sniffing around inside the inverter with a grid dip
meter, with the power completely disconnected and everything
unplugged. You might get lucky and find a portion of the circuitry
that's actually resonant at around the frequency at which you're
having trouble. If so, spoiling the Q of this resonance might reduce
or eliminate the problem.

--
Dave Platt AE6EO
Hosting the Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!