Electronic switched dets

[ProdigyChild]

I want to poll, what good ideas are around for electronically triggered dets.
Possible applications that fly to my mind are: FAE triggering, multiple fireball firing for a desired shape.

I've posted in the 'exloding bridgewire dets' thread and finally succeeded with that. However, electronic switching poses a problem because of the need of a device capable of switching kAmps. The biggest affordable SCR/thyristor/triac I've found is BTA41 (<10$), capable of switching continuous 40A, and 400A surges.
The main benefit is the avoidance of dangerous primaries in blasting caps. An average equipped electronics freak can't add precision timing to EBW caps. Pity.

The low end are simple bridgewire devices - lightbulbs filled with some primary explosive. Easy to make, cheap, reliable, can be driven by a single transistor from a average battery. The only downside may be a more or less unprecise timing, depending on how much the bridgewires are overdriven. If you want it to go fast, you need high currents again, you're approaching EBW technology, yet still having to use primary explosives.

I've just tried a very simple detonator: take a piece of paper, glue some metal foil stripe on each side. Now cut it with a scissors. The result is 2 conductive layers that end next to each other with only the distance of a paper's thickness - roughly 0.1mm.
Now apply 1000V to these and the arc starts immediately.
The good thing is: you don't need high current for initiating the current immediately. I've tried a series resistor of 330 Ohms, i.e. peak current of 3A (only). There are quite a few transistors around, that can switch such easily: BU205, BU208, BUX47, .... even high voltages MOSFETS will do and IGB-FETS would just be the perfect choice :)
This easily sets of AP and probably any other primary, too.
The main delay and jitter of such a det would be the primaries deflagration-to-detonation delay. Does anybody know numbers?

I've also experimented with piezos a bit. Not those found in lighters, but the ones used as mini loudspeakers. They have the size of a coin, at most 1mm thick and they convert electric tension to mechanic force. They cost almost nothing. At low voltages this is fine for creating sounds. Nothing spectacular. But if you drive them with a ignition coil and back them up with steel, you can let these devices hammer against a steel plate! Rather impressive to me. I failed to detonate AP in a first experiment, but still can imagine, it is possible to set it of like that. The good news would be the lacking DDT delay, because you directly generate a shock wave!

Also I wondered, if it would be possible to create a (destructive) flash out of a LED to set off a sensitive (to light/heat) primary explosive. Maybe too far fetched!?

Any other ideas of effects, that could be used to trigger dets by common currents and voltages at good timing?

[Hang-Man]

However, electronic switching poses a problem because of the need of a device capable of switching kAmps. The biggest affordable SCR/thyristor/triac I've found is BTA41 (<10$), capable of switching continuous 40A, and 400A surges.

Thats a shame for you. Try ebay. I got a new SCR rated to 6.5 kv at 350 amps, surge rated at 4500 amps, for $30.

I'm sure you could get a low V high I one for the same price.

I don't think blowing up LEDs would work, but what about diodes? When diodes fail they really go bang.

[Boomer]

The question here is whether you need millisecond or microsecond precision.

For FAEs, multiple fireballs etc milliseconds are more then enough. So a small electrolytic capacitor at 350V, switched trough a BTA41 into a 6V bulb full of AP should do. This set-up is not true EBW, as it would NOT trigger a secondary. But the filament explodes within less than 1ms, and violently enough to either detonate the AP, or overdrive it so much it makes DDT nearly instantly. The ESR of a 100µF cap, plus the impedance of 20m cable limit the current to a level the BTA41 can handle.

BTW SCRs are A: too slow for a real EBW (you need <500µs switching time!), and B: are destroyed by the current rise speed, not the peak current itself! So you might try adding 1mH to the circuit to let the current rise more slowly (still within 1ms). If you overdo this, it possibly makes the filament burn through before the current is high enough to give an explosive arc effect. I sometimes had the cap near-full, and the (empty) bulb only flashed without exploding.

EDIT Answer to ProdigyChild (below):

The reason SCRs are destroyed by fast current rises is that the current starts to flow near the gate, and it takes time before the whole square section of the silicone conducts. Thus if the current rises too fast, it is carried by a small section of the crystal only. It is the current density resulting in destruction, not the value.

This would be another form of Hot Electron Death :p

[ProdigyChild]

Thanks for pointing that out, Boomer.

Why does a rapid rise of current destroy the SCR??
I can understand destruction by over-tension (avalanche effects), by over-current (overheating).
But why should the rise of the current destroy the SCR? Maybe because it's taking too long to switch to low impedance thus dissipating most energy inside the SCR instead of the bridgewire?
Or is it because of induction of a electric field due to rapid change of magnetic flow. Gut feel tells me, that this would be less in voltage than the applied switching voltage though.
I vote for the first explanation.

In fact, BTA41 has a maximum allowed 50A/us dI/dt (@125&deg; ) which is quite far from what we would like to have (while a surge peak of 2000A/200us is allowed @ 25&deg; ).
However, closer inspection of the data sheet revealed a strong influence of junction temperature and maximum allowed dI/dt. The cooler the junction, the more is allowed. In case of BTA41: 300A/us @ 10&deg; 200A/us @ 37&deg;

Semikrom Semipack modules SKKX 92.... (X=L,H or T), continuous 95A SCRS permit 150A/us @ 125&deg; and probably in analogy to BTA41's properties over 4 times as much at 25&deg; i.e. over 600A/us.
So as a rule of thumb, a 200A+ continuous current thyristor will stand the needed 1 kA/us rise along with the allowed surges.

What I mean is: proper selection of SCR and using right operation temperature should avoid self-destruction.

The poor timing can't be improved by over-driving the gate A BIT (<1kA ;) ), can it?

[ProdigyChild]

Can't edit my post above so I have to do a bit post-whoring to report my latest experiences :p
I'm simply fascinated by the performance. Progress is so fast, that my posting is more or less a monologue. Or it's just so boring for most of you :rolleyes:

I finally found a method of precision timing initiation with all good properties I wish to have:

1) No need for primaries
2) No need for special industrial switching device - SCR's don't even work!
3) Most probably ignition delay of a few microseconds
4) Probably arbitrary scaling up of power - parallel initiation of multiple charges; exploding foils in reach???

You need both EBW high voltage/high current circuitry and a precision timed even higher voltage (ignition coil - about 10kV-20kV). Ignition coil timing is easy, because we don't need high peak power. A standard car ignition will do - I use a car ignition darlington transistor S637. Probably a high voltage MOSFET like BUZ325 would be the prefect choice for easy interface. I should be able to handle 400V and 10A.

A first implementation of the switch - a voltage triggered spark gap made of foil and tape mainly - is shown in the EBW thread. I improved the switch within a few hours, so that it's reusable umpteen times and can even detonate PETN. More or less the same performance as I already reported about my manually swiched EBW experiments based on Boomers excellent work.

As my improvised EBW detonators are not much more than a tape with a foil stripe inside, I would like to know, if it is possible to detonate plastic explosives directly from such a tape. Imagine to place a charge that is smaller than your ordinary detonator...