[hilbert42]

The screening/Faraday cage will still work with the mains power disconnected, so one should ensure the mains plug is removed before such a test.

Note: I'm a past master at destroying old electronics in my microvave (this way I can't change my mind later and then not throw it out).

About three seconds is all it takes (just time for the magnetron filament to warm up) - any longer and the stuff starts to stink from the heat. There's no second chance if you make a mistake.

[alanbernstein]

This seems like an odd way to destroy electronics. What's wrong with a hammer?

[hilbert42]

It's quick and easy and less dangerous (and more final) than a hammer!

I've been known to change my mind half way through destroying something when I think of another use for it - the microwave method ensures that the part of me that doesn't like throwing out electronic things that still work always gets thwarted.

The idea of microwaving electronics came to me a long while ago for very different reasons. In the early days of microcomputer control systems I used to work with Intel Multibus computer systems (these are so old you better check this Wiki out to know what I'm talking about: https://en.m.wikipedia.org/wiki/Multibus).

In those days Multibus boards were hellishly expensive and we used quite a lot of them. If one went faulty then we either fixed it at board level or if it failed within the warranty period then we returned it to the agent who then had it fixed (boards were always repaired - never thrown away).

Anyway, one board we received had an intermittent fault and I had it returned under warranty and the agent supposedly fixed it and returned it but it was still intermittent (you know, it had one of those annoying intermittents that was rare enough for one to think the problem was fixed when it wasn't). This toing and froing to the agent occurred at least three times and the problem still wasn't solved so I put an end to it by putting the board in the department's lunchroom microwave.

The plan was to damage a sufficient number of ICs to render the board beyond economic repair (BER) but not make the matter of microwaving it obvious to the supplier. This was very successful, a new board with a different serial number was supplied and the intermittent problem disappead.

To accomplish a truly defective board without obvious signs of user 'abuse' requires some care. The aim is to provide just sufficient microwave energy to puncture or short-circuit the junctions in the ICs but to not make it obvious (in practice, about one second of RF is needed to accomplish this).

Excessive RF will heat the silicon to the point where it boils the ICs' encapsulating compound and one ends up with little craters in the middle of the ICs which make it obvious the board's been zapped.

Most microwave ovens take about 3 seconds before any RF appears (no RF will appear until the magnetron cathode is hot enough to become emissive), so it's important to calibrate your oven before embarking on such an exercise.

I've used this technique very effectivly a number of times over the years. I must mention another incident long after the first, however this one had nothing to do with Multibus.

It involved a setup jig for programming certain electronic equipment and its motherboard was about as complex as a Multibus one. When motherboards failed I either fixed them or returned them to the manufacturer's repairer who was in a third country (he was renowned for fixing dead boards that people like me considered BER).

Again, we had an intermittent board that had crossed the oceans a number of times and it was still intermittent so I finally gave it the microwave treatment and returned it to him. Right, this silly bugger actually wasn't going to let it defeat him, he returned the board to me with a message to the effect 'this was a very problematic board, I had to replace about half the ICs [about 50 total per board], seems it was hit by a lightening surge'. Well, the board worked - sort of - but it had new and different intermittents (presumably, some of the ICs he'd not replaced were suffering breakdown puncture in their SiO2 insulation and thus were experiencing partial failures).

I returned the board again without any extra zaping and this time it was replaced with a different one (presumably, he'd either given up on it or wasn't prepared for my wrath when or if it failed again).

BTW, one can get pretty good at this if you calibrate your microwave carefully. Such zapping should never exceed nuking more than about 25% of the ICs. This is usually sufficient to deter most diehard repairers whilst not drawing their attention to the actual cause of the failure.

[alanbernstein]

Fair enough! Personally, when I reuse broken electronics, it's often because I'm scavenging more mechanical parts like ports and switches. I think those would tend to survive either method, as long as you're not utterly destroying the device.

[hilbert42]

Right, scavenging for mechanical parts is also my main aim (LSIs etc are now so specific they're useless as spares). For example, only a week or so ago I had to replace a PCB-mounted USB connector on one of my PVRs and I scavenged its replacement from an old mobo. :-)