Jim,
Although Kopp's paper lacks some critical design details there are other online papers that reveal the missing elements in sufficent detail. Key components not shown in the linked article are a Q-switch, a pulse compression device most often used in lasers, that decreases pulse width and increases pulse amplitude (the explosive's detoniation isn't brief enough) and the microwave generator's design.
Heck, according to Kopp you really don't even need an explosive pump. A lightening discharge (triggered by laser or a small rocket) will do the trick. An advantage of the lightening method is the device is not destroyed (if you can safely retrieve it). Disadvantage, you must hide the device near a target (with high lightening activity) for extended periods.
WW
-------- Original Message --------
From: jim bell <jdb10987@yahoo.com>
Apparently from: cypherpunks-bounces@cpunks.org
To: coderman <coderman@gmail.com>
Cc: Georgi Guninski <guninski@guninski.com>, "cypherpunks@cpunks.org" <cypherpunks@cpunks.org>
Subject: Re: USB killer v2.0
Date: Fri, 16 Oct 2015 05:25:36 +0000 (UTC)
From: coderman <coderman@gmail.com>
>> However, for REAL damage, it would be useful to pulse the kilovolts into the
>> magnetron: Generally if you increase the voltage by 10, the power increases
>> by 100. Feed that microwave magnetron with brief (microseconds) pulses of
>> 40,000 volts, rather than 4000, and it will probably generate close to 100
>> kilowatts. It would be necessary to establish that such a system will not
>> arc with over-voltage, either within the waveguide or the associated
>> high-voltage wiring.
>high voltage solid state built into shielding around the
>waveguide/magnetron cavity presents smallest external hazard and
>compact too.
>of course, another order-of-mag level up in explosively pumped EW weapons :)