Sciencemadness Discussion Board

Exploding HV consumer capacitors to initiate secondary explosives?

IndependentBoffin - 16-6-2011 at 02:47

Are exploding consumer high voltage capacitors sufficiently powerful to initiate secondary explosives without further modification?

http://www.youtube.com/watch?v=B8gFgIQl2HI
http://www.youtube.com/watch?v=ToWsF3HcQUU&NR=1

The capacitor basically acts as a very fast switch when it undergoes dieletric breakdown; it's resistance suddenly falls from a large value to a low one.

I really don't fancy filling my own metal tubes with primaries.

The problem I seem to discern from the Youtube videos is that not all dielectric breakdown events results in an explosion of the capacitor. Sometimes it just fizzles. I'm not sure if this is because the pulse isn't large enough or it is an inherent issue of where the dielectric breakdown happens (e.g. if it is near the outermost layer, it will just bleed out. If it is in the innermost layers, then the pressure will build up and an explosion is more likely).

[Edited on 16-6-2011 by IndependentBoffin]

bbartlog - 16-6-2011 at 04:14

Seems to me that if you can generate the sort of pulse required to blow up a high voltage capacitor, you'd be better off trying to rig some simpler mechanism that would explode in a more consistent and reliable fashion (like a thin wire). Anyway, to me it still sounds more like a way to detonate your primary, not your secondary.

IndependentBoffin - 16-6-2011 at 04:47

Surely if enthusiasts without access to professionally manufactured explosives and detonators, can initiate secondary explosives directly without a need for primary explosives, the number of accidents and lost limbs/eyes will be much lower.

That is why I am thinking hard about how to conveniently initiate secondaries without primaries with commonly available equipment.

Of course, it is only useful in a lab setting and not a car or belt for "you know who", because of the bulkiness of the HV + electric pulse equipment.

Lord Emrone - 16-6-2011 at 13:57

You could shoot tannerite

IndependentBoffin - 16-6-2011 at 14:53

Quote: Originally posted by Lord Emrone  
You could shoot tannerite


Nice idea, unfortunately not every enthusiast has access to guns. In any case electrical initiation is very convenient especially for situations where timing is crucial.

I was thinking....what about if:
1) An aluminum foil electrolytic capacitor is stripped of its end caps on both ends
2) The electrolyte is displaced with an oxidising agent, e.g. a nitrate salt by soaking it in a concentrated bath of it.
3) Solder is put on both ends. This effectively turns a high resistance parallel plate capacitor (where voltage is perpendicular to the plates) into a low resistance conductor (voltage is in plane to the coiled up plates and flows axially) .
4) Send an electric pulse through that.

Comments:
1) You probably won't be able to get stoichiometric ratios of oxidiser - fuel (aluminum foil electrodes) but it is just meant to help provide energy to vaporise the aluminum. Aluminum has a very high enthalpy of vaporisation (10.9 kJ/g).

2) Two effects are responsible for generating a shockwave, which then initiates your secondary. The first effect is pressure rise due to heating. This is governed by the energy input of your firing pulse, oxidation of aluminum and the enthalpy of vaporisation. The second is the actual generation of gas (more moles of gas on RHS of equation than LHS). In normal capacitors undergoing dielectric breakdown the only energy source is from the firing pulse causing a pocket of superheated gas. Having an oxidiser there permits an exothermic reaction to help fuel expansion of the gas. As we all know the heat of oxidation of aluminum is very high.

3) The reason why it may be worthwhile to get the current flowing down the axis of the capacitor, rather than radially, is to produce a parallel circuit arrangement. In the case of a normal parallel plate capacitor undergoing dielectric breakdown, the current density on the remaining aluminum foil electrodes drops to zero once breakdown occurs. A modified capacitor whose current is now flowing axially (no longer serves as a capacitor but a parallel resistance array), the current density actually increases as bits of aluminum foil vaporise and the cross-sectional area decreases. Therefore the probability of vaporising more aluminum increases.

4) Confinement. Physical rupture of a container provides a shock. The strength of the shock is what determines whether it can initiate secondary explosives. In theory if you have a container that is strong enough and slowly pump in a high gas pressure (no primary explosives or electric pulse heating needed) and rupture the walls suddenly, it is only a matter of raising the initial pressure high enough, to get successful shock initiation of your secondary.

Can anyone try this out and report back? My lab is all packed up due to an upcoming house move :(

Zelot - 16-6-2011 at 22:02

Here is a fun example of dielectric breakdown leading to a shockwave:
http://powerlabs.org/waterarc.htm

Of course, some sort of fuel/oxidizer mix will probably have a lower dielectric breakdown than distilled water. I would definitely like to do some testing of this.

Do you think that simply soaking a cap in a concentrated nitrate solution will allow effective displacement of the electrolyte?


IndependentBoffin - 16-6-2011 at 23:00

Quote: Originally posted by Zelot  
Here is a fun example of dielectric breakdown leading to a shockwave:
http://powerlabs.org/waterarc.htm

Of course, some sort of fuel/oxidizer mix will probably have a lower dielectric breakdown than distilled water. I would definitely like to do some testing of this.

Do you think that simply soaking a cap in a concentrated nitrate solution will allow effective displacement of the electrolyte?



Thanks for the link.

Once you cut off the ends of the capacitor and solder on terminals on either end so that current can flow in the axial direction, it isn't serving as a capacitor anymore but rather like a fault-tolerant parallel conductor. As each "branch" of the parallel conductor burns off by vaporisation/reaction of the aluminum with the oxidiser, the current density in the remaining branches increases making the destruction of further branches more likely (as long as you can sustain the power output).

Because it no longer is functioning as a capacitor, your fast switch must be somewhere else in the circuit. A Marx generator has arc gaps that can serve this purpose.

Soaking in concentrated nitrate solution has to displace the electrolyte to some degree, it is an entropy-favoured process. What might be more interesting is to dissolve some ETN in methanol and soak the capacitor in that. And of course methanol has a low enthalpy of vaporisation so you don't have to "dry" it off. ETN is great for this role because not only is it an explosive, it has a positive oxygen balance. Meanwhile liquid explosives are also a good choice because no drying and worry about volumetric change/distorting the capacitors is necessary.

You could also try any of the following:

Quote:
Examples of materials with positive oxygen balance are eg. ammonium nitrate (+20%), ammonium perchlorate (+34%), potassium chlorate (+39.2%), sodium chlorate (+45%), sodium nitrate (+47%), tetranitromethane (+49%), lithium perchlorate (+60%), or nitroglycerine (+3.5%).


http://en.wikipedia.org/wiki/Oxygen_balance

Chlorates look OK but are less compatible with the aluminum metal electrodes than nitrates. Check chemical compatibilities here:

http://www.coleparmer.co.uk/techinfo/chemcomp.asp

If you use an aqueous solvent, try drying it out and report back if the capacitor significantly distorts due to the salt crystals. Water is an energy sink that will just absorb the heat from your exothermic reaction as it turns into steam :(. Better to have as much energy as possible available to vaporise more aluminum and cause it to react with your oxidising agent.

enhzflep - 18-6-2011 at 22:43

Never tried to det secondaries with it, though I was amazed at the effect when 350v is dumped into some graphite powder.

I pulled 2 wires of the 40 wire IDE cable, striped ~2mm from the end and dipped in a slurry of graphite powder and NC laquer. When I hit this with a pair of fully charged photo-flash caps, it threw most of the water out of a 250ml plastic cup and threw the cup some 3 inches high.

IIRC they had a resistance of around 5 ohms, meaning that at _best_ one could expect to find an instantaneous power input of:
V = IR
350 = I*5
.: I = 70

P = VI
P = 350 * 70
P = 24,500 Watts

Though somehow the figure sounds higher than I recall calculating. Assuming a resistance of 10 ohms, the figure becomes

P = 12,250 Watts

In either case, not something I'd consider worth sneezing at. Should the charge voltage reach 500v, and the bridge remain at 10 ohms, the power figure becomes a more impressive

P = 25,000 Watts

Of course, wire length and inductance will have an effect on the rise/fall times of this pulse. Perhaps several dips into plain NC laquer after the initial dip would create a strong coating that would help contain the energy for just that little bit longer. It would also offer some secondary as food for that intense ball of plasma to do some work on.