exploding Bridgewire dets

[telecard]

I thought i would start a new thread on bridge wire dets as i cant seem to find much info on the concept on the web (Check the risi site, it give some tech info). I was wondering if ne one has ever done it. I am looking at giveing it a try once i can make up some more petn. I am planning on using a .97uf cap @ 4kv to power the system.

[Tuatara]

How are you planning to switch 4kV? Thyristor stack? That'll be fun to drive.

[telecard]

i wish, the scr s neccesary would cost in the excces of 150$, no i think i would have 2 find another way. But it think it would be worth it just as well as there would then be no need 4 making ne pimary explosives, i dred loosing another finger. :-(

[Boomer]

Thanks for saving me from starting a new threat as a newbie! Exploding Bridge Wire caps (EBW caps) have been one of my main interests lately. I just add some first results here for comments (Tuatara, Jumala, any more EEs here?).

1. Cap

Years before learning here that such things exist, I tried to make caps without primaries, as most are not storage stable (peroxides) or can only be handled in smallest amounts (Ag2C2, silver fulminate etc.). I had repeatedly been startled by the power of a big cap discharging. Even a small photo flash has approx. 100.000 watts for a microsecond. So I took the biggest cap I had (500V/2200uF) and shorted it, first with wires, but I soon found it was louder using resistors of 1-3 Ohms. The carbon film goes plasma and stays conductive, pulverising the ceramic body and even cracking parts near it! I took 1g of MHN in an aluminium pipe, with a 0.5W / 2.2 Ohm carbon resistor in the middle, soldered to thick wires, then epoxied the ends closed. When I held the cable ends to the charged cap, I got a full det (pipe pulverised).

2. Blasting Machine

The problem was to find a switch for 2500 Amps / 500 Volts (Ri,cap = 0.1 Ohm plus 2 x 20m of 4mm cable). You can use flash tubes as a switch, but they convert much of the energy to light. In the end I built a relay from metal bars and a big electro-magnet, powered by another, smaller cap (220uF/400V). The machine has a step-up circuit similar to a photo flash, and a flash tube to trigger the big 'relay'. I always trigger by hand, because I have nightmares of children walking past the charge while the timer runs / fuse burns, and I cannot disable it from where I sit! 20 meters is sufficient for up to 1kg of ANFO if you have cover and military ear plugs.

3. Small Blasting Machine for simplified caps

To make the thing more portable/concealable, I downsized it using an SMD CCFL driver, four diodes and a single capacitor (100uF/400V). It fits into a cigarette box including a 9V battery and a hand-operated switch (= insulated screws with a spring-loaded shorting bar)!
It will not detonate MHN with a resistor, but can be used for small Christmas bulbs (3x3 mm) with 50 mg HMTD inside a one-gram-cap of MHN. Peak current is 300 Amps, reached after 20 microseconds. Time inconsistency is +/-10 microseconds, thus they can be used to set of charges ‘simultaneously’ in rock etc. but not for opposed charges in cutting steel. Industrial EBW caps are accurately timed to +/- 0.025 … 0.25 microseconds!

4. Improving time consistency and making HMTD redundant

My goal is to make caps without primaries that vary only 1-2 microseconds (equals 1/2 inch at VoD 7 km/s). MHN needs 4cm x 2kg = 0.8 Joules in the drop test. To be reliable I want ten times that. Assuming just one quarter of the energy is dissipated in the arc from the bridge ‘wire’, the cap must hold 32 Joules. This leads to 400uF/400V, or better 4 x 100 uF to reduce inductance. The bridge element must be very low resistance to reach peak current/plasma fast. This eliminates bulbs. It must also concentrate the energy on a small spot. I am currently trying SMD resistors of 0.22 to 3.3 Ohms. Results will follow, best till now is 5 microseconds rise time. TBC

P.S. Funny to be a newbie again after 20 years of ‘blasting around’….

[Tuatara]

I'd say the biggest trick is getting the resistance of the bridge wire to be at least 10 x the resistance of the lead wires.

Switching 2500 A can be done with a 100A thyristor. Most thyristor ratings are based on continuous load conditions, not the once a day kind of pulse you want for det. It does work, I've done it in a commercial product. I had an 85 A 1200V thyristor ($NZ65) switching 2000A, 100us pulses, at a rate of once every 1.5seconds . I've still got one of those thyristor beasties kicking around somewhere - sort of an unofficial bonus from when I left the company:D

The next tricky bit is getting really low impedance caps. That means low ESR and low ESL. Lots of little caps can be better that one big one, if connected up the right way. These parameters are somewhat beyond control if you are relying on scavenging for parts - but photoflash caps are a good bet.

After playing with the arc out of the electric fence energiser I designed I wonder if the arc alone would be sufficient to use as a det. No bridge wire - just the two lead wires with a small gap filled with explosive. A high voltage arc fired through the explosive would heat and vapourise the compound in microseconds, especially if it was backed up by a 50-100 Joule cap bank. Certainly firing the arc underwater was always impressive. Now I think about it, thats how lithotripters generate shockwaves for smashing kidney stones.
You'd do it the same way a fence energiser works - caps charged to 600v or so. Thyristor switches caps into a low impedance step-up transformer to give at least 5kV on the output for the arc. It could be an autotransformer as we don't need safety isolation from the power source.

Damn - so many ideas and so little time ...

[Boomer]

Tuatara, after some oscilloscope tests I think the bridge wire resistance in not so important: The short circuit current pulse of the cap looks roughly like the surge pulse of IEC 400x, 1-5 microseconds rise time and 30 microseconds duration (instead of 1,2/50). If you use a bulb of some Ohms with a 300V cap, you see a short pulse of 30A (5µs) followed by a nearly perfect pulse (300A/30µs) directly after. The first pulse is the evaporation of the wire, its current is Vcap/Rwire, it takes little total charge. The second pulse is much higher and has most of the energy. It seems the arc where the wire evaporated/exploded is much lower in impedance than the wire was. You can prove this by taking 1206 resistors of 2R2, 4R7, and 15R. The first peaks are then 150A, 60A and 20A, while the second pulse is close to the cap’s short circuit pulse!

Your fence energiser idea is like the stun gun people here used to set of AP. I tested that too, but even with electrolytics of some 100uF discharged into the coil, the arc only ignited the surrounding explosive, that’s probably why it only works with AP. Maybe with a bank of caps and a bigger coil it would work? But then you need a wire for 10kV or more, perhaps on wet ground…but try it and tell me about the results.

P.S. Funny to be a newbie again after 20 years of ‘blasting around’….(Maybe I can forget this sentence now after 10 posts)

[Tuatara]

A wire for 10kV is easy. Most people don't realise that plain old RG59 coax cable will handle 20kV with ease. Thats the kind of cable used for TV aerials (the round stuff, not the flat ribbon).

What you've said about the vapourising wire makes a great deal of sense. That means it is even more imperative that the lead wires be a really low resistance. Can you measrue the voltage across the arc at the same time? Then we could get a rough idea of the arc impedance and try for a better match.

[Boomer]

Hi, after some more oscilloscope tests I can give you some waveforms. The problem is I have not found out how to put graphics or pictures into the text here (like Axt’s SC thread), so I will have to describe it (component values from my last 2 post, with 10 m cable 2x1.5mm):

The current rises to Vcap/Rbridgewire instantly (scope resolution), stays there for a time t1, then jumps to near the short circuit current of the cap and lead wires (300A), and drops exponentially with the fall time tf=50µs). The voltage is 300V for the first period, drops within 3-5µs to 50V at the instant the current jumps up, and then stays nearly constant until the arc stops. The only variable is t1. It is <3µs for 1R/1206 and grows linear with the resistance. It is ca. 5µs for 3R3, 20µs for 10R, 50µs for 22R and also 50µs for a Christmas bulb. (With the bulb the current stops before the cap is fully discharged, so I will stay with resistors.)

This means two things: Firstly we need a bridge ‘wire’ resistance as low as possible (1R) to have the arc start fast and with little delay variation. Secondly the arc voltage is 45V for 20A and 50V for 300A. It will not get much higher even with a much bigger cap, only the current increases. Thus the power is approx. 50V * cap current or 50V * 300V/ (Ri,cap+Rcable), dropping with the current. The initial power density is 50V*300A / (area of resistor) = 750kW for 1*2mm (1206).

In the Fibre Optics Detonator threat a value of 12MW is quoted for PETN initiation. With MHN having 1/5 the drop test value, I guess it will require ca. 2MW. Therefore a 470µF cap will do, especially if loaded to 450V. This method might even need less energy per area, as we already have heat and a mechanical pulse from the exploding resistance film, while the laser pulse first has to be converted to heat, and this to a pressure pulse.
I will try this soon and let you know!

[BLASTER]

The key for working EBW - 1kA current rise in less than 1us in EBW circuit

few hints how reach this:
- low inductance capacitors - ideal are "snubber" caps used in power SCR circuits, best two in parallel to divide parasitic inductances by half (2x 0.47uF rated at 2500V)
- use of coaxial cable
- superfast switching circuit, use of SCR's impossible - too slow switching
best (and used) is triggered sparkgap called trigatron, look in google for some references of this device
- also good power source - good compromise is DC to AC converter working at some 10kHz at least, also with voltage multiplier 250V to 2000V, of course use of fast diodes in multiplier

[Tuatara]

One can build a triggered spark gap readily enough. And the switching speed can be improved by filling the gap with hydrogen.

Another option if you want to stay solid-state is to use a saturating inductor pulse compressor. It is possible to sharpen a pulse rise time from 100us to 10ns with one of these things. The you could use a thyristor switch.

One could also use an IGBT. A GA200SA60U handles 400A pulse with a rise time of just 75ns. $225 from Farnell, so not exactly a cheap option! Maybe multiple parallel lower current ones?

Edit: Just found this fantastic page explaining EBWs and EFIs (http://www.risi-usa.com/0products/8td/page01.html)

[telecard]

These may not be very usefull but i got some shots(http://bridgewire.8m.com) captured @ 30 fps of 2 instances of an exploding resistor. As can be seen the resistor remains intact on resistor**bang test and only a small burnt out hole forms. The other test just overexposes the frame and i get a blank frame. The tests i carried out on a 400v 330uf charged to 320volts and contact made via a heavy duty relay i aquired from a microwave. I am working on an array, parralel, all charged to 320v of a mixture of caps exceeding 150uf@ 330volts+ and i found an src i can use will try it out sometime in the near future.

EDIT
Dont take notice of the basktball crap, jus free hosting

[Dunkelmann]

A very interesting text on Detonators and EBWS can be found at
http://www.risi-usa.com/0products/8td/page01.html

have fun!

[Boomer]

I conducted some more oscilloscope tests to see current waveforms. With electrolytic caps at rectified mains you don’t get under 5µs rise time, even at 1000µF with peak currents approaching 1kA. This means you get 200-300A after 2µs. After adding some foil caps (MKP, 3x3.3µF/X2) I got the waveforms described in the risi website mentioned above. The current jumps to 500A in 0.5µs, then makes a short drop and rises again more slowly to 800A. BTW I found that 3.3R/1206 need 10µs to evaporate the resistance material at 100A before any further rise is possible, therefore I switched to 0R1/1206 which do not limit the current (it is limited by Rcap + Rleadwire).

For a fast switch, why not parallel some MOSFETS? We use IRF types with 130A peak current in TO247, they cost only 3 $. For 30 $ you can get a switch handling 1300 Amps, at switching times in the order of 100ns. Driving 20A into the gates can be done with a simple P-MOSFET for a buck. Spark gaps aren’t always that fast, I once build one for 8kV, it needed 2-3µs to switch 1kA.

The 2MW I estimated were also mentioned on that page, but for PETN instead of MHN. Thus I am positive MHN caps will work with 0R1/1206 and two 220µ/450V caps with three 3.3µ foil caps to support the first peak. The explosion of the resistor is very loud and you feel a shockwave and some painful particle spray on your hand 4 inches away.

I will make such caps this evening with only 100mg of MHN in brass caps. This way they can be tested in my room, inside a big steel pipe, wrapped in cloth and some old blankets, with a wooden box around all. It is easy to see if the MHN detonated, it pulverises the metal cap body, and if you take more than a quarter gram, you get dents in the steel pipe, and it gets too loud for indoors if you’ve got neighbours next floor.

Results coming soon…

[Dunkelmann]

Maybe you could use a Transistor or a Diode in a metal TO39-Case, ie 2N2219, if you remove the top.
Inside you have the Die, connected with very thin bonding wires made out of gold, maybe this could be used as an OTC Bridgewire. Just fill in your explosive, and there you go..

[Tuatara]

Better still, take the top off a TR5 or TR3 style fuse (http://www.wickmann.com/html/ptr5.html) and fill that with PETN or whatever. At least fuses will be consistant from one to the next.

Incidently Dunkelmann, that link you posted is the same as the one I posted - check next time, eh?

The thing that really interested me on that site was that they were suggesting that EBWs could be detonated with only 2 joules. Thats really not very much, considering your average throw-away photoflash stores about 12J. Obviously the trick is to get that energy into the bridgewire at a colossal rate. The RSI website describes it in terms of getting the wire vapourised, while still confined by its own inertia. This suggests to me that the bridgewire metal should be of low specific heat, and low boiling point. Aluminium?

[Dunkelmann]

Sorry, must have overlooked your link.

In the Risi-Document they say that they use Gold for Bridgewires, which made me think of the Bonding Wires.

[Tuatara]

That might be for corrosion resistance, since the dets they make are likely to be stored for some time before use. I can't think of any other justification for using something as pricey as gold. Or maybe the inertial confinement works better because gold atoms are heavy.

[Boomer]

IT WORKS !!!
3 x 470µF charged by rectified mains (325V), discharged into 0R1/1206. With the resistor in the end of a 5mm i.d. stainless steel pipe (0.5mm walls), followed by 100mg MHN, other end open with only some tissue, I got a partial det (pipe blown to twice its diameter). With the resistor in the middle of 150mg MHN, and the cap crimped, it was a full det (pipe was completely fragmented). The tests were conducted inside of water pipes to keep the fragments of the cap for inspection. When the last test was repeated with 400mg, the ½” water pipe around the cap was split and fractured! Thank God this had been in still another 1” pipe, capped and wrapped in lots of cloth, inside a wooden box full of pillows, in my bedroom. The interesting thing is, that though it wrecked the ½” steel pipe, there was hardly any sound outside the box, just a muffled ‘plop’! This method saves a lot of time, you don’t need a hole in the ground, no 12-240 Volts converter, no digging and no running home and returning every time. Just stay below ¼ gram. BTW a 1” pipe from a bicycle is completely severed by a 1-gram cap, even if both ends are open. Nice way to demonstrate the power of BCs.

In further tests it must be investigated whether the current rise time is really so critical. It was around 5µs with only the electrolytic caps. Maybe foil types at a higher voltage need much less capacitance. At 1.5kV, 2 joules equal only 2 µF, while at 300V it is 50µF. And this is NOT enough with electrolytics, I only got deflagration with so small capacitors. But to speed up Christmas bulb detonators, 47µF/400V per BC is good to fire simultaneous charges. The delay between the shots is much smaller than at 9 Volts, only for e.g. cutting steel with opposed charges it will be insufficient.

[Tuatara]

Excellent! What did you use for a switch? Did you still have the foil caps supporting the electros?

For those buying foil caps for this job, be aware the not all foil caps are suitable. You must buy pulse rated caps, otherwise they will spit the dummy under these high discharge conditions. What seems to happen is the connection on the end of the foil metalisation gets burned away (i've dissected a few in my time as an electric fence designer) and the cap becomes useless.

[Boomer]

I had a look into the box again, it were in fact electrolytics of together 2000µF without foil caps, giving 105J at rectified 230V mains. Even this does sometimes cause only a partial det, and adding some µF foil caps did not help. I wonder why with 50 times the required energy it is still not reliable? The ‘switch’ was two copper rods moved towards each other manually. This sounds crude, but the current waveforms are very consistent. I will try a parallel MOSFET configuration soon, as SCRs are too slow and do not withstand the rate of current rise.

In case it really is the current rise time, I prepared a second box with metallized foil types, 18 x 3.3µF/275V/X2/MKP = ca. 60µF total. I know these are not perfect for pulse applications, but film/foil caps like FKP1 mean 30 x the size and money! The best results are achieved by putting them in two series groups of 9 each, giving 15µF/550VAC = min. 1000 VDC peak (X2 types hold more then SQR(2) times the AC value, they are tested with 1kV DC. In stun guns they are even used up to this limit; I assume 500V for each to stay safe). The are charged via a variac with voltage doubler. The discharge sound is much sharper, and the current rises to 1000A in 500ns, this is 2kA/µs through 2x10m of loudspeaker cable! I will test this box tonight at home with real BCs – poor pipes and pillow box…..

Another thought is that the 1206-resistors are not as good as gold wires, though the current waveform is now very similar to that given at the risi website, rising to 1kA in 500ns and falling to 800A in 500ns, then rising again more slowly to 1,2kA. If the current drop signals the time the bridge material is vaporized but still hold in place by inertia, it has taken only 700µC. This equals 160mJ at approx. 230 Volts dropped above the resistor. Ergo only 8% of the required 2 Joules are put into the exploding bridge, then 7.34 Joule are put into the arc. On the other hand they claim their EBWs to work from a minimum of 500V and 180A, which puts only 90mJ into the bridge wire in 1µs before it explodes and the arc starts. It is easily imagined that a 0.002” gold wire takes less energy than the 1mm x 2mm active surface of a 1206 resistor.
So with our total energy being 7.5J instead of 2J, and putting 160mJ instead of 90mJ into the bridge material, we should expect this set-up to work 100%, especially with MHN being 5 times more (drop-)sensitive than PETN. Otherwise it is indeed the shape of the bridge (i.e. gold wires instead of conductive surface) which makes risi’s caps work better! We will know tomorrow…. By, Till then.

[Tuatara]

I think it may simply be that the 1206 resistor is spread into a thin film, where the solid wire is a narrow cylinder. So the whole thing comes down to energy density. Bear in mind the fact that resistors are designed to dissipate heat, not blow up, and I think it is obvious why resistors seem less reliable. Shame - they're so convenient!

Can you try the same trick with a bit of steel wool, or fine nichrome?

As for the switch - I see nothing wrong with slamming two chunks of copper together! Its probably a better switch than a spark gap or MOSFETs;)

[Boomer]

Did some more testing again, but this time after some thinking: It was wrong to think that the total energy of the cap counts much! The discharge of electrolytic caps takes 50µs, the bridge explodes after 2-4µs at the current rise times reached with these caps, and 99% of the energy comes later. One microsecond equals 5-9mm propagation of detonation, therefore the detonation front is an average 2cm away from the bridge wire when the arc starts (!) to converts the other 99% of energy, and it is through the whole main charge when the capacitor is discharged!

Keeping this in mind, plus the data from the risi website, I concentrated on the foil cap solution giving 1kA current, rising at 2kA/µs. Measuring the voltage drop over the bridge, I found it is only a fraction of the available cap voltage, much more is dropped over the lead wire’s and the cap’s impedance. Increasing the resistance back to 3R3/1206 most of the voltage was dropped over the bridge (1R would have been ideal though).
With these resistors I did three tests in the same manner as before (5mm stainless steel tube, 200mg MHN etc.) and had three full dets!

Because the metallisation of the SMDs kept breaking at the solder joint either while pressing or even while bending the external cable, I looked for something else and remembered the comment about the TR5 fuses. Being lazy, I first drilled a small hole and filled them with NG. Had one success and two failures (plastic fuse body only split), proving what is said in ‘Josef Koehler, Rudolf Meyer Explosives 4th ed.’, i.e. about NG not detonating from a No1 in the lead block, while other nitro-esters will. Using MHN again, I had success two consecutive times. I prepared 8 more of the caps, using from 250mA to 1.5A fuses. If all these work, a medium value will be tested further, as ‘10 out of 10’ can very well still mean only 90% reliability.

When all goes well I will upload photos and schematics somehow.

[Boomer]

Sorry it took so long, but I was busy. The pics will have to wait, it seems not that easy. But one thing I DID find out: The fuses with a wire wound around an isolator never work: five duds out of five tests. The fuses with only a thin wire inside are more promising: eight full dets out of eight tests. I got noticeable dents in the water pipe around the 250mg cap each time.

I ordered 100 TR5 fuses to see which value is best, and if it is consistent. They cost 25 Cent instead of 5 Cent for the 5x20mm – how unfair!

(could not edit)

[Tuatara]

Probably the fuses with the wire wound on a support are slow-blow or surge resistant types - not what you want for an EBW!

[inverse]

150nF , 3KV, flat, ultralow impedance. Self made with polyethilene and aluminium foil and very large flat leads.
It was originally built for resonant charging of a Blumlein switch, but I thought it might be worth doing some nice play with it.
The discharge peak power is around 100-300 MW. A thin aluminium wire 2cm long explodes with a bright flash and a sharp crack when directly closing the circuit.
The same wire, when bound to paper sheet, cuts it in half with its shockwave.

The discharge itself, when happening inside a glass pipe filled with water, destroys the tube.
Longer leads are used in this case.

Please note: no explosives of any kind were used.

Risetime is essential, total energy being only a secondary factor.
For maximum output, a current risetime near or below 10ns is better.
Depending on the wire parameters, at such speeds the wire is subject to magnetic collapse:
the magnetic from high currents inside the wire generates an inward pressure which a few nanoseconds makes the wire collapse on itself and heat much more than by simple inertial confinement.

A properly designed foil slapper won't work with 1us discharges. It needs properly designed
sparkgap switches, capacitors and impedance matching of discharge paths.

Now, the foil slapper concept is fascinating, to say the least.
Currently my capacitor won't help, tho. I think I'll have to scale it 10x.

[Boomer]

Hi, sorry for my long 'holiday' but I was in hospital and am now typing with 7 fingers. NOT due to an EBW cap, an AP cracker etc ... I cannot say if I stop completely, but I will probably limit my work to things without fuses/primaries. Maybe EBWs are my field: insensitive plus me sitting on the other end of a long wire!

Inverse: foil slappers are the best there is, but for us 'normal' EBWs are the first goal IMO, I already got a design working with standard electronic parts fore a few bucks.

P.S When my hand hurts less I will start a thread in the WC about what went wrong for me, and whether others had similar (near-) accidents.

[Bert]

There's a thread in the watercooler called What has been your worst accident (http://www.roguesci.org/theforum/showthread.php?s=&threadid=3297&perpage=50&pagenumber=2) Sorry to hear you got bit-

[Blackhawk]

I have started my EBW detonator today, using the power supply for a cold cathode flurecent light (CCFL) and two 400V 330uF electrolytic caps in series (as the CCFL inverter outputs 487V so the combined power handling of the caps would have to be 800V instead of 400V). I used a simple diode bridge to rectify the CCFL output, strangely enough on my multimeter (a crap one btw) it appears the CCFL sends out pulses of energy every half second or so. I tested a 15sec charge period (to the caps) today over a hair thin 2cm long length of copper wire, which vapourised instantly with a loud crack, I will try resistors tomorrow. For switching I plan on simple two metal plate contact for now. Would Picric acid or Potassium Picrate be sensetive enough to use in the EBW cap?

[Blackhawk]

Just a quick update, I have noticed that using two pieces of metal touching is definately not the best way of switching as when attempting to fry objects with a resistance of around 1ohm or so the contacts spark and errode significantly without the resistor being touched. I thought that I might use a MOSFET to drive the caps, I have one currently that is rated to 500V at 14A continous, the device code is IRFP450, it is a HEXFET if that means anything (appart from cost :(). The data sheet states that it's maximum pulsed drain current is 56A, I am wondering if this is somehow internally fused or safeguarded in any way so that the max current would never pass this, also could this MOSFET stand discharging two non pulse rated electrolytic caps with a charge of 400V 330uF through a load of around 3 ohms (so around 133A)?

Sorry about asking all these questions but when 'playing' with this kind of power I would like to be sure of what I'm doing.

EDIT: I have an SCR rated at 160A pulse so I will try this to trigger my circuit, I know that their triggering time is not optimal, but at the moment using the two metal chunk method all of my energy with low resistance bridge wires is being eaten by sparking, and high resistance bridges (such as lightbulbs) restrict current too much to accheive any decent explosion from the wire.

[powdermunkey]

Blackhawk- there is no internal overcurrent protection (on cheap FET's). It'll fry pretty dam quick. One thing that is overlooked often in EBW and slapper driver design is that the pulse impedance of the connecting wires is mismatched- i.e, 50 or 75 ohm coax will not transmit a short pulse efficiently into a 10 ohm bridgewire. In my own experiments, I made a low impedance transmission line by using two 1/8 inch wide conductors on the opposite sides of a piece of scotch tape. It makes a transmission line with about 2.5 ohms impedance, and about 5 kV standoff. Fast risetimes are very important.

[Boomer]

Yes, but if you do not need picoseconds, a loudspeaker cable will do. I managed to get 2000A/µs i.e. approx. 1kA after 500ns through 10m of 2 x 1.5mm cable. But forget about electrolytics (see my posts above). 15µF MKP at 1kV gave reliable dets, while several 470µF electrolytic caps in parallel needed 2-5µs to reach peak current. And forget about the mosfet, unless you put 10 in parallel and build a driver that puts 10 Amps into the gates.
A last tip: go down with the bridge impedance. At 3 Ohms the current will start at 133A and stay there for a few µs until the bridge goes plasma. Then it jumps up to the short circuit rating of the caps + cable. I can mail you scope pictures if needed. Thin wires below 1 Ohm are better (TR5 fuses even come with their own cap housing). Avoid those where the fuse wire is wound around a glass fibre core, the naked wire type is better!

BTW, the idea behind an EBW is to avoid primaries completely. So PETN or MHN etc is the thing to use, not picrates or even peroxides!

[powdermunkey]

Boomer- what kind of scope did you say you have? I was playing at the time with slapper type caps, using a .01 uF at 2kV. The literature suggests a sub microsecond risetime, and I was about 10 meters down the cable, so a flat stripline was cheap and easy. I was trying to directly detonate sensitized AN, with ambiguous results. Maybe I should make some RDX....

[Boomer]

I used a digital storage scope from Fluke for some thousand bucks (company of course). And AN is impossible to initiate with a slapper or EBW, unless with ‘sensitised’ you mean 50% NG!
About the choice of explosive: HNS and TNT are the hardest to get going, followed by RDX, PETN and ETN (in this order). The easiest – also concerning manufacture and precursors – is probably MHN. It is 3 times more sensitive than PETN, but still considerably safer than primaries like fulminates or peroxides. It cannot explode from sparks or static, and takes a much harder blow and more friction (probably because the crystals kind of ‘smeer’). And it only deflagrates if heated, unless at >180C with confinement, or if boiled.
Another thing is that IMO slappers are even more sophisticated than EBWs, and harder to improvise. It is hard enough to home-make reliable EBWs with a small, portable, 9V-battery operated B-box. The sub-microsecond rise time is also true for ‘normal’ EBWs, mine were unreliable and needed thousands of µF until I went beyond 1 kV, reaching at least 1kA in 500ns. This calls for foil caps or even better MICA pulse caps – used in military exploding foil initiators for smart bombs.

The reason I speak so much about EBWs is that after my accident I cannot touch AP, HMTD or anything having a fuse without pissing my pants. Maybe this will change again, but it is a sane attitude, which hopefully lets me grow old.

[powdermunkey]

I have a healthy fear of primary explosives, too. I am fond of aluminum sensitized AN because I can shoot it with my FAL and get it to reliably detonate. It would be fun to have an option to detonate it directly. I guess I'll have to make some nitric acid and try MHN. I have access to pulse rated caps of various sorts, so I think I'll play around a bit. I smacked a bit of AlAN with a hammer on my anvil, and got it to explode, so I know it's at least theoretically possible to get it to go off with a slapper/EBW device. Lots of room to experiment, for sure.

[Microtek]

Has anyone considered spark-gap detonators ? I read a patent ( US 3955505 ) where the inventors used a discharge of a 0.1 uF capacitor charged to 5600 V across a gap of about 0.5 mm to detonate PETN or RDX.
The voltage could be reduced somewhat according to the patent, but does need to be quite high.

[Microtek]

Has anyone considered spark-gap detonators ? I read a patent ( US 3955505 ) where the inventors used a discharge of a 0.1 uF capacitor charged to 5600 V across a gap of about 0.5 mm to detonate PETN or RDX.
The voltage could be reduced somewhat according to the patent, but does need to be quite high.

[Boomer]

Powdermunkey: Interesting you made ANAL detonate with a hammer! It can hardly me THAT MUCH more sensitive than ANFO, and IIRC this takes 400 cm*kg. What is FAL, Fulminate/ALuminium?

You don’t have to destill NA to make MHN. It can be made in an hour (distributed over two days) from SA + AN, even without an icebath and much stirring, with – considering the little work - decent yields (>50%): Just look at my method in the ‘MHN-yield’ thread in the HE section. But if your freezer does not reach –25C in the coldest setting you better add the mannitol in two portions with two hours between. I forgot to set my fridge to maximum power last time and had a nice runaway in a one-litre (!) batch. Harmless but a smelly mess: I had to discard all my food. There was so much condensed nitric on everything, even outside the freezer compartment, that I had to neutralise the cans to save at least the beer!

And Microtek: Yes, I tried this. I stepped the discharge of an electrolytic cap up from 300V/500A to 30 kV/5A using a transformer. The spark was 1 cm long, going through the HE all the way. But probably due to the low current/slow rise time it only worked with primaries.

[powdermunkey]

Heh- a FAL is the FN FAL (Fabrique Nationale Fusil Automatique Legier) 7,62X51 assault rifle. It fires a 149 grain projectile at about 2700 feet per second. I put the sensitized AN in a plastic bottle and fire the rifle into it. It does take a pretty sharp blow with the hammer to get it to detonate on the anvil. The hammer is about a half kilo, and I swing it from about two feet away, with a sharp snap of the wrist. I'll look up the mhn-yield thread, thanks.

[Microtek]

I finally obtained a disposable camera with flash so I could use the capacitor and charging cirquit for a blasting box. I only had one camera so just one capacitor, slightly smaller then an AA battery. The capacitor charged to 300 V, but I don't know any other specs for it.
I experimented some with various types of NPE detonators ( exploding wire and foil, slapper, different configurations ), but all of them failed. I then tried with more containment: A brass tube, 10 mm long, 6 mm OD, 5 mm ID with one end closed by a thin layer of melt cast MHN/PETN/EGDN/NC. A little MHN was added ( ca 0.05 g ) and the bridge-foil assembly was inserted.
The bridge-foil consisted of a piece of cardboard about 5 mm in width and 20 mm long which had had a single piece of Al-foil glued on in such a manner as to cover both sides. Before the glue had dried completely, some of the foil was cut away to make sure that there would be no contact between foil and the brass tube. At one of the narrow ends of the cardboard ( the one the foil was folded over ) some of the foil was cut away so that the remaining foil formed a 1 mm wide bridge right over the fold.
As mentioned above, this assembly was inserted into the loosely packed MHN, a wad of wax was inserted on each side of the cardboard, and finally the tube was sealed with epoxy.
When connected to a cable using a clothes pin ( to pin the stripped leads to either side of the cardboard ) and fired using the flash cirquit, a partial ( or perhaps low order ) detonation was obtained.
The tube was unharmed, but the clothes pin was damaged by the explosion.

Following this, another test was made. This time the tube was turned from an Al rod. It had the same dimensions, but one end was closed. 0.072 g MHN was packed loosely into the cap, and a bridge-foil assembly of the same type as before was inserted. As before the open end was sealed with epoxy over wadding. When this was fired, a full detonation was obtained.
It should be noted that the cable was very short ( 1.5 m ) to eliminate attenuation of the electrical pulse.

[lucifer]

For a EBW system I use a very low inductance 1500V 10 uF capacitor and charge it up to 1500V DC.
This capacitor is connected with a very big relays to the bridge wire.
This relays can handle up to several hundreds of Amperes.

For the bridge wire I use 0.1 mm copper wire, wich is also used for making transformers.


Closing the Relays will fire the EBW and will give a load explosion.
I Over-voltage the Relays so that it will close very fast and it will make a fast good contact.

I tested it with 0.2gr MHN and it worked, The MHN must have been detonated because the housing of EBW was completely destroyed. (I tested it too without MHN inside and this time the housing stayed intact.)

I have one concern about this setup and this is how reliable will the contact of the relays be?
Is it possible the contact will bounce and that it will make contact for only lets say only for 250 ns, enough for opening the wire but maybe not enough for detonating the HE inside?

Or is it more likely that the current will keep flowing even if the contact of the relay will bounce, because the air in between will be ionized?

I think the second, but maybe some one can tell me if it is likely that I will get problems with this system?

[Tuatara]

There should be a healthy arc during your contact bounce, maintaining the current. In fact the arc probably strikes moments before the contacts actually touch! Don't expect a huge lifetime out of those contacts though

[Boomer]

As I promised to Rhadon (?) in another threat, here comes the circuit of my EBW B-Box. I did not include the internal parts of the CCFL inverter, if someone is interested I can post them later. It should be one made for a longer tube (>8 inches) to give enough voltage.

The resistors parallel to the diodes are necessary to equally divide the voltage. Those in series protect diodes + inverter in case the caps discharge not equally. The 4 diodes to the right and the LED are not absolutely needed, but useful to show when the caps are full. They also prevent overcharging.

I am open for any questions.

[nbk2000]

Here's a question.

Where is the schematic? :rolleyes:

[Boomer]

Rhadon deleted it.
He must have misunderstood me in the "personal info" threat. I scanned the raw text again and did not find ANYTHING that could lead to me, so here it is again, in case he does not have it anymore to re-attach it.

Rhadon, thanks anyway!

[Rhadon]

I thought what you meant was that the file you uploaded contains personal info again, so I deleted it. But I approved your attachment now, so everything's fine.

[Child-of-Bodom]

Hi,

15µF MKP at 1kV gave reliable dets
This is a total energy of 7,5 J (W=1/2 . C . V^2 Where W is energy in J. C capacatie of cap in F, and V volts)
In your B-box however, you have 14 275V 3,3 uF cap's parrallel. This means in total 275V and 46.2uF. Total energy would be only 1,75J
I guess you used MKP cap's - I can only find them in AC :confused:
they are 16-50uF 450 V AC.
I am interested in you battery drived CCFL...
Sorry for all these quetions - but I don't want to wast money on wrong components...

COB

[Child-of-Bodom]

Ok, I already find out something...

I now have 3 MKP cap, 220V AC, But they can be charged up to 320V DC (with the flash circuit of a trow-away camera - just cut off the cap and replace it with yours) They have a capacity of 16, 66 and 99 uF.

16 didn't work, the cap is broken I think, when I release it from the charge circuit it runs down in voltage...
66 gave a loud bang when discharged over a copper wire with r = 7,5mm.

As switch I use a 24V driven relais from a car, I can get many of them, so maybe I will make some snap-in montage.

greets,

COB

[Child-of-Bodom]

I got it working!!

I used an empty .22 shell, filled with 3mm MHN. As bridge I used r=7,5mm copperwire, and as cable 1,5mm². I make the 2 cables together, cut them off, and on top I solder the wire, this is a little difficult, but it works. (I can't find TR5 fuses in a store)
The end of the shell is closed by pressing.

As charger I use the circuit of the trow-away camera, I am busy with a CCFL driver circuit.

The caps were charged to 300V, but the strange thing was that after the det 220V was still over the caps. (when I am testing 'dry' the caps always completely discharge) Could this mean that schockwave blow out the arc? If this is treu, I can use less power to detonate. (=less space in my d. box)
As swich I use a relay which I can open with a screwdriver when it is metled together :rolleyes:
Thanx for all the info, Boomer, Microtek and others!!

[Marvin]

I have the feeling the MHN test is too easy for this application. I was under the impression it was a primary explosive, Ive just checked Davis, and he considers it 'almost, but not quite a primary explosive'. He also says that it will detonate from the flame of a match that causes local overheating. So Id question if its really the shockwave from the bridge thats causing the detonation, or just local overheating in many of these tests.

You might be forgiven for wondering if this matters, but given the substance is stated to be about as shock and friction sensitive as liquid nitroglycerin, and yet rather less stable chemically, do we really want to be making detonators out of it? ISnt doing away with unstable primaries the whole point?

I have the feeling electrolytic caps are slightly too slow for this application and my experience of camera flash circuits is that they dont tolerate abuse well, particulally the disposable ones.

[Microtek]

It is true that MHN is rather sensitive and not particularly stable chemically, but as with so many things this is a matter of compromise:
A sensitive base charge such as MHN will be more reliable when it becomes time to initiate it, and a less powerful blasting box can be used. It will also be somewhat more liable to explode accidentally ( though still less than a cap with primary explosive ).
A less sensitive base charge such as RDX will definately not explode unintentionally from any realistic accident, but it will require a state-of-the-art blasting box and even then, misfires will be rather likely.

In my opinion, well-neutralized MHN is not overly unsafe but things like PETN would obviously be preferable from a chemical stability point of view.

[Child-of-Bodom]

Well, the big (in my point of view) advantage of MHn is that you can store t for a rather long time, if you dissolve it in hot EtOH, it will be very hard to detonate.
To let the EtOH dry to get good MHN doesn't take too long, within a day it must be ok.

The charging circuit I am using allows itself formidable for abuse, it is doing the same job as usual, anly a bit longer. However, for 10 $ you can buy a CCFL inverter which works on 12V and gives an output of 600V AC. - two 2 KV diodes will do to rectify.

Maybe this is a good option if this circuit is broken, but so far this works nice.

[Boomer]

I never had it DETONATE (i.e. high order) from heat. I experimented a lot with caps containing only MHN, plus a power resistor to overheat it. They SOMETIMES worked, but it war a low order detonation / deflagration. The cap metal was only split, not pulverised like when I started adding some 25mg of SF.

COB: You might have had a low order det using electrolytic caps. You can easily test this by putting the cap on a test plate, and compare the hole / dent with that made by a cap with an additional 0.2g AP (or 0.025g SF etc).

I agree that PETN or at least ETN would be better. I guess that is when you will need a circuit like mine. Instead of giving you 150A within 5µs, it gives 1500A within 0.5µs. :D

BTW you can double the efficiency of the cap if you press ¾ of the base charge REALLY hard in a vice (20000 PSI = density 1.65 = 8km/s), then hand-press the last ¼ and add the bridge wire! As soon as I get the digicam back, I will post pictures of the difference it makes. :)

[Child-of-Bodom]

I use MKP caps, electrolytes didn't work out...
I use 2 220V AC caps in a parallel situation. They are 66 and 99 uF. charged to 300V as explained before.
Switching is done by a 24V driver car relay, it can handle up to 50A. Sometimes it is stuck, but with a screwdriver I can open the contacts again, and so it work already quite some time...

Boomer, where did you 'find' the tr5 fuses? I am playing with wires r=7,5mm, but that is hard to standardise, I mean, it is hard to make everytime exact the same cap.
I cannot find them at Conrad or in any electronic store...
I can order from germany (I'm 'Hollandisch'), that's no problem.
thanks!!

[ProdigyChild]

I've read in one of the links above, that commercial bridge wires are extremely thin - at most 0.0025 " IIRC , that's about 60um for metric peoply - and made out of gold. That's less than the thinnest copper (0.1mm) wire I can buy.

I wonder if carbon fibres could be used, too? Carbon fibres filaments are about 8um in diameter. I know about the positive temperature coefficent of carbon (graphite), i.e. resistance drops as it gets hot. This seems to be the major drawback of this material, besides it's relative high resistance. But is is really a problem WHEN the greatest power dissipation is achieved? Why use thousands of amps to heat a metal wire if you can use thousands of volts to heat a carbon one?

Another thing I'm wondering about is: can't I use a normal diode as power switch?
A one way power switch that uses avalanche breakdown of the diode - kind of a solid state sparc gap ;)
Use one capacitor charged up to below the diodes reverse voltage. Connect the capacitor to the bridge wire via the diode in wrong polarity. No current flow. Now apply some overvoltage to the diode and it breaks down and reverse current increases dramatically initially, destroys the junction and you have bridged the diode (forever).
Application of overvoltage can easily applied by another diode in series with opposite polarity and applying voltage to this diode (reverse). I should have the possibility (post count) to post a diagram soon.

SCRs (thyristors) capable of switching high currents are SO UNBELIEVABLE expensive :eek:

[ProdigyChild]

Finished my first EBW circuit today and blew up a few mg of MHN with it :)

Nothing much improviced but a electronic circuit build from scratch with a handfull of standard devices, around 30$ in total.
It charges two 10uF/500V capacitors used for AC motors.
The result is seen by the bridgewire as a 5uF cap charged to 1000V. This was shorted over a 5mm short carbon fibre (7um). Bazzzzz!!
The little bit of MHN sprayed the tape I put on top of it into my face, that's we we ALL use wear goggles in our dreams, don't we?

Carbon wires (fibres) explode very well, too (not only PTC metals). At least as far as I can say right now.

I someone is interested in a schematic, I can draw and post it.
The ciircuit is as follows:
2 transistors (BD135) work as an astable, timing determined by two 10uF caps and base resistors of about 8k+protections resistors of a few hundred ohms. Collectors wired directly to a cast 1.6VA transformator (2*4.5V, 1x220V). Secondary voltage is doubled and rectified with two 1N4007 and the two beforementioned (non-electrolytic) capacitors. That's all! Quite resistant to abuse (shorting), low power consumption.

However, I should have chosen a 2*3V trafo, because I need 15V= input for 1000V= output right now. I wanted it to be operated by a single 9V battery.

Oh yeah, if someone wonders: over-driving the 3V coil of the trafo with 15V is OK - I use a higher switching frequency than 50/60Hz. However I count on its insulation to stand the 500 Voltsc :D

[ProdigyChild]

Thanks to my intense research recently I've got news for you and can answer some questions I posed myself.

1) Carbon fibre wires suck.
Not, that they don't explode. But the need far more voltage to perform well (compared to AL foil, for example). I succeeded to detonate a 10mg 'chargelet' of MHN once, but failed to do 2 times.
Detonating PETN failed, too.

2) Aluminium foil is wonderful!
Cut a stripe of 0.5 to 1mm width, 2cm length off a household aluminium foil. Tape it between copper foil contacts of 3mm to 5mm distance. It explodes violently with my EBW-box described above (5uF/1000V/2.5J, non-electrolytic). Even PETN in smallest amounts is set off reliably :)

3) I've understood that EBW are an excellent way of setting off tiny amounts of HE !
I've used PETN/MHN charges of 3mm x 2mm x 1mm in volume, 'confined' by a tape on top, stuck on top. The HE well blows away the copper foil contacts. This never happenes, when exploding the Al foil wire alone.

4) My idea of using a diode an applying overvoltage to breakdown is shit. It does not break down resistance fast enough and EXPLODES. Thereby of course consuming most energy itself :mad:

@Boomer: you've convinced me at last.....

[ProdigyChild]

Well I got my 160A (4000A surge)/1200V SCR (thyristor) today. I could not resist trying to blow up a aluminium foil with it.
I engaged mit HV charging device, checked the tension with my volt meter - BAZZZ!!?????
OK, I exploded the series resistor inside it. Must not happen, but after opening I saw it was f***ing Chinese work inside and two resistors nearly touched each other and the high tension just ionized the air in between. So my volt meter was gone :mad:

I concluded, my HV charging device definitely worked and charged to about 1kV. Next I build a neat bridge wire from aluminium foil, taped it on a plastic plate, attached the 1kV capacitor to it with the SCR in series. But what was that sparkling when closing the circuit ?? Obviously, there was enough current trickling through the SCR to show a visible tiny arc, when closing the circuit. While my brain was trying to find an explanation, I was repeating this one more time when suddenly ** BAZZZZZ ** the SCR decided to switch without gate trigger and was destroyed at once (shorted). :eek:
What the f***, I thought, already gone after a few seconds?
Yes. Really. Boomer is so right, when he says SCRs won't do the job.

There are few more reasons agains SCRs in EBW-triggering:
How the hell to avoid a too rapid rise of the voltage at the SCR (dU/dt). You can't simply plug your detonator to your blasting box and then trigger the SCR. The sudden voltage rise will trigger immediately by exceeding critical dU/dt. So always:
1. plug
2. charge
3. trigger.

Moreover, you have a too high trickling current to use a low power charging unit, like I do. Voltage breaks down to a few 100V and that will definitely not give a detonating wire. That's a major problem with slim battery powered systems.

So do what? I HATE mechanical triggering, really.
I spontaneously decided to build a triggered spark gap from nothing more than a bit wire and tape. See the attached image below.
It consists of a 50um PE foil. The first foil I could get hold of in my room. On the upper side I taped 2 copper foils (rather thick - 0.2mm at least) and left a gap of 1mm. On the lower side I taped another copper foil with 1 sharp edge pointing into the middle to the gap. Know, the 50um foil separates this 'gate' from the other two electrodes. On the left electrode I taped a aluminium foil stripe of 1mm witdth and 7mm length in series. All that connected to the HV-capacitor charged up. Then I wired a car ignition coil output to the 'gate' and ground to the right electrode. Then stoke the ignition coil with a 30V power supply ** BAZZZZZ ***. Aluminium bridge wire vapourized :D
I don't claim this to be a practical solution, but it shows at least how easily you can build a (unlimited) power switch rather than spending you money on expensive silicon, that dies at the very first attempt to do something unuseful.
The spark gap shown only works once. I have to clean it and re-tape the electrodes before the next use, cos they're bent upwards and oxidized a bit. A bit more solid construction along with a better foil (higher dielectric constant, perhaps polystyrene) and especially more massive electrodes would make it an attractive alternative to mechanical switches!
This was my first triggered spark gap ever! I've tried years before but always failed.
I wonder if I should patent it as a EBW film thyristor ;)

[Child-of-Bodom]

Why do you hate mechanical triggering? I use an over voltaged car relay, which can handle 50A. Although you put 1KA thru it, the only thing that happens is that the relay gets welded together. I’ve made a small hole in the plastic cover of relays so I can re-open it later with a screwdriver. This relay can stand at least 600V, maybe more, and you can simply adjust the distance between the switching contacts, they are made of copper.

Your solution is very nice and is working, but it also has to be operative in absence of light, in the middle of the night, with cold and not easy to manoeuvrable fingers, in a wet gras land. (For me at least)
That’s why I like to use a relays, just use a long wire and a simple charge circuit to charge a 40V cap, and press the button.

No offence to your idea, I have some troubles with wires, so I'm really going to test it, but then with a piece of alu foil between 2 copper plates, which are soldered. It's a pity that alufoil is inimpossible to solder, and copper foil of the same thickness is hard to find.

[ProdigyChild]

COB, definitely try alu foil wires. I have found nothing that works performs like that. Although the low melting point of Al should be a drawback?
You don't need soldering. At all. Instead, you cut a long thin stripe off the Al foil with a sharp blade and a ruler. As thin as you can.
Stick this stripe/wire on your favourite tape. Here you have your EBW tape! Tape it between two contacts - done.

There's no real reason, why I don't like mechanical switching. It's the more practical solution, you're right.


I've bought a high voltage capacitor on ebay. 2100V AC (= nearly 3kV DC), 1 uF. Sounds good but is sh1t. This kind of capacitor is usually found in microwave ovens. Unfortunately, they have a parallel resistor inside of about 10M. Bummer!
Not much you say? Oh yes, at 3kV this is U*U/R = 0.9W that leak away. Too much for my up to 4kV charging unit. It reaches 1kV only with the 10M load.

I build a capacitor myself today. household alu foil + 80um PE foil. A bit clumsy, but I managed to detonate some PETN with it at last. I reached 0.25uF capacity by stacking 10 layers of foils (10xAl foil 0.45m x 0.3m). In the end I got an internal short that destroyed it. Pity!
Where do I get a leakage free capacitor of a few kV and from 0.5uF upwards? What machines contain such?

[Child-of-Bodom]

It's a pity that since the iDefence incident a lot of info from this thread was gone, but I had some of the theads offline, since I don't have a continious internet connection.

This is part of the thread, where in the last reply Boomer gives a solution to your problem, ProdigyChild...

Boomer
As promised, here is a picture of a 1 inch water pipe I used for EBW cap testing (indoors, in a sound-proof blast box!). You can see 5 very small dents (circular), marked with lines to reuse the pipe. They were caused by 0.4g MHN each, hand pressed into small metal tubes. They had a TR5 fuse bridge wire, B-box had 1500A with 0.5µs rise time (see my other posts).

You also see two big dents, where the steel is partly torn, plus the torn right end of the pipe. These were also 0.4d caps, but 3/4 of the base charge was compressed to density 1.65 - 1.7 in a vice (at 20000 - 30000 PSI).

The difference you have from a properly pressed base charge is amazing!

(Sorry for the picture size, it was 16k at 540x300, and cutting it to 496x300 to allow upload INCREASED it to 36K. How that? I must get photoshop pro)

Edit: photo completely on the bottom!!

COB
Nice pic!!

Boomer, I have some questions about the scope, how to set it in the right 'mode' for measuring very fast currents. Maybe this is too specific for here, would it be possible for you to e-mail me? It's a new model Fluke scope.
Thanks! (I tried before, but is was difficult to get them appropriate on screen...)

Marvin
Just a suggestion and assuming its a digital scope, but I would try using 50 ohm cable, terminated (50 ohm resistor) at the explosive end next to the fine wire, battery over it, and set the scope input to ECL mode. That would alow you to inexpensivly keep the scope a good distance from the charge.

Some experimentation to set the trigger level, try seeing how frequently you get random triggers for some very low settings....
Set it so you see whats going on before it triggers (negative time delay).

COB
Hmm, I test at a company...I won't go in there with explosives...

I'll try to trigger with a trigger wire over the bridge wire, I hope the plasma destroyes the trigger wire. The relais takes tens of milliseconds to close, so that doesn't work for triggering.

Thanks anyway.

Btw, I just found out that I made a small mistype in a post before, I use wires r=0,075 mm, not 7,5 mm

Marvin
Whoops my fault, I read the post and assumed you were trying to do a VOD measurement. Ignore everything I said about trigger wires. By fast currents, you mean the fast/high current of the bridge det circuit.

You need the charger circuit isolated from the mains (say by running from a battery). Put a very small resistance in series with the bridge, it needs to be much smaller than the bridge itself. 1mohm would be good for 1000A using a scale of 1V for example. Take the input over this resistor, preferably with a 10:1 probe which vastly reduces the chance you will do perminant damage to the scope.

For measuring voltage again use a 10:1 input, but you will need an aditional voltage dividor. 10:1's will normally take 300v or so but its best to reduce this with an external dividor, if the charger is a 3kv job this is clealy essential. You need a dividor not as high as the probe impedence (usually 1M or 10M) but not so low that you draw noticable power from the circuit. The dividor will go over the bridge.

If measuring voltage and current at the same time be sure that both grounds are at the same point. This should be the only point your circuit is grounded given its running from a battery.

Ignore the trigger input alltogether and use the current input, it should be reliable. Again use negative triggering to see the whole pulse. For the first attempt go down to the noise level and back off until you get no random triggering.

Use TTL mode for these measurements, not ECL. Very very important.

Boomer
OK Marvin, I was wondering what you meant. I measure current with a 1 mOhm resistor towards ground, and voltage with a 100:1 (2.5KV) probe towards the same ground. But most Fluke scopes (DSOs) are insulated, and my B-box runs from 2 x 9V blocks and a CCFL driver anyway. It charges to 500V (standard cap) in 3 seconds, and to 1.5KV in 13 seconds!

COB, put the caps in series if possible, even if they are not exactly the same value (or add 33µ to the smaller)! The energy is the same, and the higher the voltage, the faster the current rise. And parasitic impedance (wires etc) have less effect.

If they are X1 or X2 types (noise suppressors for mains voltage), they have a test voltage of 1KV. You can charge them to 500V each (1000V total) without problems. My MKPs are 275 VAC/X2, I charge to 750 VDC (1500V total). You can use a simple voltage doubler.

PS: Conrad part numbers are 536768 / ...84 / ...92 for the 1A / 1.6A / 2A TR5 fuses (47 cent each at 10 pcs + VAT). Maybe they are only in the business catalogue, not in the one for private users. But they should deliver to you without question if you are registered there.

[Anthony]

Child-of-Bodom, what is the isolation level between the switching parts, and solenoid parts of an automotive relay?

You've got potentially (some member's blasting boxes) several thousand volts on the switching side, just waiting for an earth, and a few millimeters away are the wires which lead to your hand with the firing switch...

[Boomer]

At COB:
Thanks for saving those posts! I was wondering where that photo had gone, was just about to re-post it, as I referred to it in another thread (about the power increase of highly compressed base charges). But I didn't have the texts saved on HDD.

At Anthony:
Unfortunately some car relays have lower insulation than those for mains voltage, but some have the same insulation, creapage and clearance distance, because they use the same plastic body. There are types with guaranteed 4KV insulation, and 10mm between coil and contact pins. These should be safe to use, I tested one for breakdown and it took 17kV!

Moreover, the voltage “looking for an earth” will only hurt the operator IF he is earthed, and rubber boots with 10+ mm soles should protect somewhat. Plus, the low terminal of the charger circuit would have to be earthed too. If the output is without any earth connection, it would take a triple breakdown to do harm: From the contacts to the coil/user, through his boots to earth, and from earth to the low terminal of the circuit. The latter is the biggest problem, as most people will be using speaker wire (e.g. 2 x 6mm, 1/3 the resistance of 1.5mm mains cable) which is NOT rated for HV (500V max test voltage, not even allowed for 120 VAC).

[ProdigyChild]

I happened to look at a broken computer motherboard and noticed, the Cu lines on it are about 0.1-0.2mm in widht. Given the thickness of 37um of the copper layer, this is a suitable wire similar to a 60um wire - perfect choice for a bridgewire, isn't it? You can cut if and peel off the mainboard after heating above a candle easily.

They perform as good as the aluminium wires I normally use. Possibly you can cut out a suitable section of a PCB and already have a sturdy carrier for the BW :)

[nbk2000]

Or use a PCB etching kit to make your own bridgewires and solder your EBW box leads to it.

[Boomer]

I tried that but without success. But that was before I upgraded from electrolytic to foil caps with 10 time faster current rise. Will try again later.

BTW here is the pic of my 1500V/1500A/12s-charge-time EBW box which I promised:

[Centimeter]

I recently purchased 12 2kVDC/2uF pulse type mica-paper polymer capacitors. On the information page for the capacitors, it specificly states that they are often used for millitary EFI devices. Before I discovered the existence of this thread, I decided to make two sub-capacitor banks in which each has 6 capacitors hooked up in series. I would then attach the two capacitor banks together in parallel so as to create an overall value of 12kv/2uF. I also purchased a HV transformer that steps up 150V to 9KV DC which is just right for my capcitor bank. I then purchased an adgustable tungsten electrode spark gap. I also purchased some HV diodes so that I don't have to worry about damaging my expensive transformer. Overall, this cost me about 100USD. The stuff should arive some time this week. Is this the optimal capacitor bank that can be created from the 12 capacitors for EBW aplications? Should the charging circuit be in the form of an RC charging circuit? If so, what kind of resistor would I use for such HV aplications. I was planning on using a conductivity meter to indicate when the capacitors are fully charged. How do I go about setting this up so that it doesn't blow up? I don't plan on using the blasting box for explosive aplications right away. Rather I intend to use it in an art piece representing anger. Thus I would love to have the whole thing powered by a hand crank generator. I know that I can make one that easily puts out 4.5V which can then be stepped up to 150V no problem. Is this feasible? Also, in your "professional" opinion, is this going to blow the fucking shit out of a one inch long hair thin strand of copper wire? I want it to be a fantasticly loud and bright explosion, hence my attempt at making the capacitor bank have a farely high voltage. Any and all help is most apreciated. :D

ps- The capacitors cost about 8USD a piece and if anybody wants I can give you the contact informaion of the guy that is selling them.

[ProdigyChild]

Nice toy! How much current does it draw while charging and how much in trickle charge mode?
You used 3 varistors to stabilize the voltage at 1500V instead of Z-diodes?

I've built a low power 2kV flyback converter using mainly a CMOS 4093 and a high voltage MOSFET (IRF710 happend to come into my hands). I used 10 (!) 200V Z-diodes for generating an feedback signal indication target voltage reached. Feedback current is only 6uA. I haven't expected Z-Diodes to have so low trickle current below Zener-Voltage!
Is the same true for varistors??

Despite of the energy saving design, my circuit still consumes around 30mA @9V=270mW) in trickle charge mode (only the 6uA trickling! = 12mW). I believe that's because it's so inefficient at high charging voltages.
The circuit avoids high diode reverse voltages and especially slow voltage reversion. That's why I don't (believe I) need the series resistors when cascading diodes.

At the moment, I'm rather disappointed about my EBW experiments. Since I use a cable (5m only!), things have become MUCH more difficult and I'm misfiring more often than not :mad:
Since total resistance is less than 0.1Ohms I blame it on inductivity of the cable.

I use open detonators, i.e. confinement is done by plasticine only. I have either detonation that ruptures the plasticine ball (2-3cm) into a hundred pieces or I have a misfire. Low order dets are avoided by this method. And I can do everything indoors with only a few mg of HE.
I found, that the type of wire isn't really important. Aluminium foil (household foil > chocolate wrapper foil > capacitor foil) is my favourite. Electroformed copper foil can also be made easily (estimated 10um) but isn't worth the effort in my opinion.
Super fine copper wire - thinner than 0.1mm is available from tiny electronic coils. I broke a flash light ignition coil (15mm high, 15mm wide) while experimenting with spark gaps, disassembled it and found that it contains the ideal bride wire :D
It's copper wire hence solderable after burning and cleaning.

Also I found that putting a bit of paper beween explosive (PETN or MHN) is almost an insurance for misfiring.

I have to calculate or measure the inductance of the wire.

EDIT:
I've done this right now: @3.3Mhz the cable behaves like a 4uH inductance (Z=83Ohms). What a bloody shit! This limits dI/dt for a 2000V power supply to 2000V/4uH = 500A/us. Not really what I would like to have :mad:

This explains a lot and generally rules out speaker cables. The DC resitance is neglectable compared to its AC impedance! Unless I've made a big mistake in calculation...
I'll compare to a coax cable soon.

[ProdigyChild]

@Centimeter: You should put resistors in series, if you diodes are exposed to high current peaks while connection your capacitors or while discharging. Also expect, that the cable + caps form an oscillator thus reversing the voltage of the caps at radio frequency. I've seen up to five sine waves on the oscilloscope (when the bridgewire did not explode) in at a few MHz. This voltage reversal can easily kill recitifer diodes. I suggest using a few kOhm, limiting the current peak to 1A.

One problem with checking the charge is power consumption. I use 10 10MOhm resistors and a 50uA analog meter to measure voltage up to 5kV. However it consumes 50uA at 5kV!
Do not try to measure 1kV with you multimeter. Most of these toys only stand it for a few seconds despite a 1kV range :mad:
In the data sheet you will find the actual restrictions (duration!)
I've killed 2 of them, damn! There are no varistors inside - too expensive????

blowing 1inch requires a bit of energy. 1uF won't be enough, 10uF are (at 2kV). You're in the right order of magnitude of capacitance, if you charge up to full voltage.

[Child-of-Bodom]

I did some testing, and have solved my problems partly. I use new caps now, 11 1KV caps with a capacity of 2.2uF, and 9 caps of 250V/4.7uF. I put the 11 caps parallel, The 250V caps are set in series by 3, and 3 of these blocks are in series with the 11 1KV caps. (I hope I was clear…).
This is all charged to 1.3KV, my disposable-photo-camera-print couldn’t get more, it started fuzzing over the 1.3KV. I blew my CCFL inverter to the electronic heaven when I tried to charge my caps up to 1.2KV with it. So I use it at 1.3KV…
This delivers 30J of energy, what should be way enough. It actually is, but not with some cable length, with 50cm of cable it worked very well, but when I tested it with 4m cable (separated 1.5mm² cable), I only had one full det, but it wasn’t reproducible.

So the length of the cable is essential, as Prodigychild has already explained me by email (thanks therefore!!), so my question to the ones who actually have their box working (Boomer?): What cable do you use? Just ordinary speaker wire, or coax wire? The latter should give a better performance due to the earth layer around it. I can get very thick coax cable, (1.5mm² at least in the centre) as I cannot imagine that a thin coax cable can beat an ordinary 1.5mm² cable.

BTW: For the ones that would like to try 12um thick copper foil, drop me a line in the mail… I stepped over to wires, since they perform better in my lab, for one or another reason.

[ProdigyChild]

Today I shot a few photos of my EBW equipment and experiments.
The printed circuit boards are in fact 'hand drawn circuit boards' made from scrap yard copper coated boards.

Image #1 shows on the left from top to bottom:
- 8V power supply
- 9V to 2000kV (stabilized) flyback converter
- 5kV voltage meter
In the middle you can see 3 0.33uF/2kV caps paralleled, below these a spark gap triggering at 1.9-2kV
On the right side, there's a epoxy board with copper tape connections and a 60um bridge wire soldered between
All experiments were made with this setup (enery dose).

Image #2 shows EBW for fun: A 30mm Al foil stripe (15um x 200um) is blown in a way, that it does NOT go plasma but only fractured in a hundred pieces. Only on th bottom a bit plasma formed. Although looking spectacular, this is not suitable for setting off HEs. Also, percussion output isn't much.

Image #3 shows EBW for blowing HEs. Al foil is only 5mm long and no shrapnel can escape vapourization. It forms a typical bright flash ball. Too bright for my camera and the number of gray filters and sunglasses I used (1 each) :cool:

Image #4 shows shows the 'plasticine block test'. You can get an idea of energy output of your circuit, much better than with your ears. However, I CAN'T decide if a circuit is good enough from this test only. Only if a circuit does not make dents like that it has NO chance of detonating HEs, really. Not much louder than eating pop corn.

Image #5 shows 'plasticine block test' with a HE charge of 1mg (one milligram; approx 1 cubic mm) of MHN. It lifts the block and puts a round dent of about 12mm diameter in the plasticine. The bridgewire alone (+a 130um tape on top) causes a dent of 4mm diameter only. This test can be performed indoors with people living next to your door. Use a plasticine disc of 30mm diameter and 10mm height at least.
Using larger amounts of HE (about 100mg) results in fracturing the plasticine and blowing a hundred pieces all over your room :mad:

Also, I touched the the charged caps today with one hand (thank God) . I must say it's quite painful! I bridged on the wire's behalf with thumb and first finger and saw the spark gap firing (flashing) :eek:
In contrast to 'normal' electric shocks (limited current!) I felt a burnig pain inside the first finger for a few minutes along with lacking sensation at the tip. After about 5minutes the sensation recovered and the pain was gone.

[ProdigyChild]

Hi,

I read your post about capacitor arrangement and found a small error:

Putting 6 2uF/2kV caps in series yields a 2/6uF / 2*6kV cap.
Paralleling 2 of these is: 2/3uF / 2*6kV = 0.66uF/12kV !!

Huge amount of pulse energy though ;)

I'm interested in purchasing such a cap (or two) in order to compare performance to more easily available ones.

[Boomer]

At COB:
I had no misfires with a 2.5m and a 4m speaker cable till now. When you say “separated 1.5mm cable” do you mean you split the two conductors (for better insulation)? This might be a reason for misfires, as it greatly increases inductance by forming a loop.

At Prodigy:
The inductance you measured at 3.3MHz is probably a little high. Some HIFI gurus measured for cable resonance which is claimed to influence sound, and found 2.2 – 8.5 MHz depending on length. Shows that for HIFI resonance is a myth, but your measurement was probably effected.

I looked at several cable data sheets and they are between 0.4 - 1 µH/M for cross wired 4-conductor types and 1.2-2.4 µH/m for standard twin wires. This would give around 2µH for your cable.

BTW my 2.5m cable was standard 2x2.5mm2 twin cable, the 4m one was two 2x1.5mm twin cables in parallel (halves inductance). From this close, the shockwave of 50-90g Plastique is like a punch in the guts, and if your finger slips out your ear consider the ear drum gone. That is why I will buy a 100m ring for 14 Euro, and take 6 twin pairs of 15m length in parallel soon.

PS: The rise time I measured with the scope was THROUGH the 2.5m cable!

[mil&co]

I surfed on a Dutch/Belgian pyro site, (This one (http://www.chemixtry.tk/) ) and I came across the idea of putting an aluminium wire in an staw filled with destilled water. The reaction that might take place: The aluminium-oxide coating on the wire falls away because of the heat, disposing the aluminium plasma to react with the destilled water.
In fact, aluminium (whitout oxide coating) would react with water just like potassium or sodium would.

Would this be an idea to "boost" the effect of an ordinary EBW.

[Child-of-Bodom]

I wonder how you get that idea from chemixtry, as I was the one that started the EBW thread around there.
It seems bugger to me, as Al is not very reactive to water, if I toss 80um Al in water, almost nothing happens, and a piece of Al foil does not weigh too much (a piece of 0.1 * 1.5 * 0.012mm), so the amount of Al dispersed in the water won’t be so much. I cannot see the water act as booster, no way. Anyway, why use water??? Water is not really the material used in detonators.

Anyways, ontopic:
I had some struggle with wires ed, see post upwards, I read carefully through the whole RISI website, it has a shitload of info in the technical documents, also concerning the local laws, written in a funny way, you don’t expect from a company.

I solved my problem with wires by using COAX wires, and then not 2 separate, but using the mantle and the core. The best line would be constructed yourself, by taping a foil on tape, adding a thin-as possible insulator, and adding foil again, like a capacitor. This type of wire has a small amount of inductance. An optimum must be find in capacitance and resistance of this wire…

I only ‘suffer’ one thing, but I don’t know or this is normal.
If I use a plastic drinking straw casing for the MHN, in the casing just a few mm go DDT, and the rest gets launched. When I put my EBW in a 9mm shell, filled with 150mg MHN, and I crimp the 9mm shell, I have never duds, it always works.
I would like to set of the primary in a drinking straw since I live in a rather crowded area, and for setting of small amount of HE, less then 20gr. I am afraid that the charge gets flyed away by the detonator when I use MHN in a metal casing.
I don’t know why actually, does it make sense anyway? I’ve not so much experience in setting of explosives by a detonator…. mostly inside with small amounts.