Sciencemadness Discussion Board
Not logged in [Login ]
Go To Bottom

Printable Version  
Author: Subject: capacitor discharge
schatz
Harmless
*




Posts: 14
Registered: 15-12-2009
Member Is Offline

Mood: No Mood

[*] posted on 15-12-2009 at 06:46
capacitor discharge


Hello everyone,
I intend to build a capacitor discharge exploder for SFX squibs mainly, but might also do double duty in commecial quarrying.

I am going for 50 to 100 WS. The highvoltage generation circuitry is no problem since I have some experience in this area, however I have my doubts about safe triggering.

I was thinking that since switching 100 joules will require a very heavy duty mechanical switch/relay, I though of using a power thyristor.
My question really is that there must be some way to protect the thyristor because I am sure that sooner or later, somebody is going to dump 100 joules in a shorted wiring circuit; This might damage the thyristor.

I wanted to see how its done in commercial exploders and surfed around a bit for schematics, but found none!
I wonder if anybody here can give me some usefull hint.
View user's profile View All Posts By User
12AX7
Post Harlot
*****




Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline

Mood: informative

[*] posted on 15-12-2009 at 07:43


Spark discharge (ignitron, krytron, etc.). They don't make thyristors that handle the voltage anyway (>4kV).

What type of capacitors are you using?

Tim




Seven Transistor Labs LLC http://seventransistorlabs.com/
Electronic Design, from Concept to Layout.
Need engineering assistance? Drop me a message!
View user's profile Visit user's homepage View All Posts By User This user has MSN Messenger
schatz
Harmless
*




Posts: 14
Registered: 15-12-2009
Member Is Offline

Mood: No Mood

[*] posted on 15-12-2009 at 08:09
capacitor discharge


I have'nt built anything yet. But I have a fair idea from SS laser work what will be suitable for storage capacitor. I have some low ESR , 1 kv, energy caps in my junkbox I can probably try out first.

I have several 1500volt 1500 amp thyristors. I am not sure if they will survive the big jolt.
In that case just what is actually used in commercial exploder units ?

In the past while playing with laser power units parts , I fooled with EBW experiments and just enjoyed the supersonic plasma crack; But I always used a big mechanical contactor which pretty much had short contact life !

So whats the switching element in those compact 50WS exploders?
Thyratron?
View user's profile View All Posts By User
hinz
Hazard to Others
***




Posts: 200
Registered: 29-10-2004
Member Is Offline

Mood: No Mood

[*] posted on 15-12-2009 at 08:31


A triggered spark gap (Trigatron) is probably the simplest solution, as everyone can manufacture it on a lathe. Probably the raise time is not as short as with a Thyratron, but you can make it yourself and you can hardly destroy it.
It would actually be interesting if someone would compare the raise-time between Trigatron and Thyratron.

http://en.wikipedia.org/wiki/Trigatron
http://www.rossoverstreet.org/tesla/projects/Pulse/pulse.htm...



[Edited on 15-12-2009 by hinz]
View user's profile View All Posts By User
aonomus
National Hazard
****




Posts: 361
Registered: 18-10-2009
Location: Toronto, Canada
Member Is Offline

Mood: Refluxing

[*] posted on 15-12-2009 at 09:08


I've built a capacitor bank for fun, but its huge compared to the purpose for this. Short of going to complete overkill and using a pair of hockey-puck sized semiconductors, you can either use semiconductor, mechanical, or triggered sparkgap.

Voltage and current are all proportional if you want 100J, E = (1/2)(C)(V^2) (note, C is in farads, you must convert units properly). At lower voltages, the resulting voltage drop across your resistive load (wire to be exploded) is low, and the current rise (dI/dt) is low. If your wire is too beefy, or your capacitors too small, a slow current rise means more current is wasted just heating it up without getting to the stage where its red hot and starts to melt/evaporate.

For a mechanical switch, the key is to simply close it as fast as possible. Use spring loaded large copper contacts, or use a pair of copper wedges spaced apart, and use a pneumatic piston or other system to quickly push another copper wedge into the 'V' space between them.

For a switched sparkgap, its only practical on higher voltages in the kV range, and you can either have 2 stationary electrodes with a third one moving from the side to reduce the distance between both, or 1 stationary, one moving electrode.

For a semiconductor switch, you can use a SCR (thyratron), but they are only good to about 1kV. These are expensive, tricky to use, but would give the best result (silent operation, relatively durable).

I would personally recommend using a SCR with a reasonable voltage of maybe 600-800VDC and using a little more than only 100J. You should be able to find them cheap on eBay, because alot of old industrial motor drives and phase-angle controllers are being swapped out and replaced with IGBT based modules.

Oh, and one other thing, design a backup system to discharge the capacitors. I recommend having a high value resistor directly bolted across the capacitors for slow discharge (capacitors can rebuild their charge over time), and a second smaller SCR used with a resistive load (ie: lightbulb) that can be triggered to discharge the capacitors in the event that the main switch or load circuit fails.
View user's profile Visit user's homepage View All Posts By User
schatz
Harmless
*




Posts: 14
Registered: 15-12-2009
Member Is Offline

Mood: No Mood

[*] posted on 15-12-2009 at 09:09
capacitor discharge


I understand your reasoning hinz, but surely, the smaller (50WS) exploders dont use these kind of devices do they?

Tried google for schematics but no real diagrams turned up.
View user's profile View All Posts By User
dann2
International Hazard
*****




Posts: 1523
Registered: 31-1-2007
Member Is Offline

Mood: No Mood

[*] posted on 15-12-2009 at 19:21


Hello,

A book by Cook!
See links in this thread, post of 12 Aug 2009
http://www.sciencemadness.org/talk/viewthread.php?tid=12414#...
Page 369 of book has lots of info. on exploding bridge wire detonators. Thread may be useful too.
You need a VERY LOW INDUCTANCE (low esr too) capacitor for 'real' exploding wire detonators, the ones with (no primary explosives) only PETN or RDX in them at the bridge wire. A photoflash capacitor will not do the job.

Direct link to book here:
http://www.megaupload.com/?d=0DIJE7NC

Regards,
Mr. Madinajad (spelling? :D)


[Edited on 16-12-2009 by dann2]
View user's profile View All Posts By User
schatz
Harmless
*




Posts: 14
Registered: 15-12-2009
Member Is Offline

Mood: No Mood

[*] posted on 16-12-2009 at 04:41
capacitor discharge


Hey guys,
my need is NOT for EBW or similar.
I am looking for schematics for blasting exploder machines for ordinary hot wire e-matches and commercial dets.
Power , 50 to 100 WS.

I have not yet figured out how to dump 100 or so joules without causing damage to the exploder itself.
Perhaps, thyristors can be used, but will probably need protection in case of shorted firing circuits.

This is why I was looking for diagrams or schematics of the smaller commercial exploders.
Thanks !
View user's profile View All Posts By User
JohnWW
International Hazard
*****




Posts: 2849
Registered: 27-7-2004
Location: New Zealand
Member Is Offline

Mood: No Mood

[*] posted on 16-12-2009 at 06:20


See this recent thread, which I started, on extra large capacitors for power storage: http://www.sciencemadness.org/talk/viewthread.php?tid=12817
View user's profile View All Posts By User
densest
National Hazard
****




Posts: 359
Registered: 1-10-2005
Location: in the lehr
Member Is Offline

Mood: slowly warming to strain point

[*] posted on 16-12-2009 at 14:04


Commercial inexpensive CD units use 10-25A or so SCRs chosen for their pulse handling rating. A multi-hundred-amp SCR will probably survive a short into most wires. They're very rugged. The big question is whether they turn on fast enough (many microseconds to fractions of a millisecond) for your purposes. It will be plenty fast enough for igniting something. It will probably not be fast enough for EBW. An IGBT would be (50-100 nanoseconds), but they're less robust. Assuming a few feet of wire, there's enough inductance and resistance for many large units to survive. Data sheets for most surplus units can be found online - check for the one-time peak current rating (time duration and current value).

Reliable, reproducable triggering either requires a fairly stiff (1-10A depending on what you're triggering) fast rise time (under 1 microsecond for the SCR, under 100 nanoseconds for the IGBT) driver. There are commercial ICs which will do the job for under $1 available from the usual surplus sources, under $2 new. Common part numbers start with "TSC42" though there are many many other types. Use short (under 6 inches) twisted pair wires from the trigger IC to the power device. Use a schmitt trigger (74C14, HC, LS, etc,) to drive the gate driver from your switch. An old MC144490P contact bounce eliminator ($3, goldmine-elec.com) is a good thing to use. Put a disable switch shorting the gate to ground for safety.

I'd -strongly- advise against using a lash-up like this in a quarry. If a $100 pre-built fireworks grade CD unit is out of your price range, you can't afford any hospital bills.
View user's profile View All Posts By User
quicksilver
International Hazard
*****




Posts: 1820
Registered: 7-9-2005
Location: Inches from the keyboard....
Member Is Offline

Mood: ~-=SWINGS=-~

[*] posted on 17-12-2009 at 13:41


Does anyone have schematics with defined components of either the commercial item or a pipe-dream proposal? I MAY have some that are much better than patents - from a company called Ideal products.
View user's profile View All Posts By User
woelen
Super Administrator
*********




Posts: 7764
Registered: 20-8-2005
Location: Netherlands
Member Is Offline

Mood: interested

[*] posted on 18-12-2009 at 01:10


Quote: Originally posted by aonomus  
I've built a capacitor bank for fun, but its huge compared to the purpose for this. Short of going to complete overkill and using a pair of hockey-puck sized semiconductors, you can either use semiconductor, mechanical, or triggered sparkgap.
This capacitor bank looks very nice, but it is fairly low-voltage and this can be handled with more or less normal components. I myself have three pulse discharge capacitors of huge size (5 kg each) having a capacity of 40 uF at a voltage of 5000 volts (max. voltage is 7500 volts, but recommended is not to exceed 6000 volts, I stick to 5000 volts). Discharging such a beast is not easy at all without causing big damage. I once tried using a smal bottle with a rubber plug, having two wires in it and making a vacuum such that a spark can be made more easily. I only tried this at 1000 volts and the effect was rather scary (very loud noise from bottle, I fear explosion of this at higher voltages). A lot of energy is released in the switching device itself and not all of the energy goes to the load. I still did not find a suitable and safe method of discharging the big capacitors in a non-destructive way.



The art of wondering makes life worth living...
Want to wonder? Look at https://woelen.homescience.net
View user's profile Visit user's homepage View All Posts By User
WaveFront
Harmless
*




Posts: 31
Registered: 1-1-2005
Location: Southern Europe
Member Is Offline

Mood: curious

[*] posted on 2-1-2010 at 09:36


Quote: Originally posted by woelen  
A lot of energy is released in the switching device itself and not all of the energy goes to the load.


...unless the switching device is the load itself. It would be nice to see a common 1000 V diode exploding when the rupture voltage is achieved.

With a few thousand volts the explosion could reach 1 "microton", the energy of one gram of TNT. Wonderful capacitors.

[Edited on 2-1-2010 by WaveFront]

[Edited on 2-1-2010 by WaveFront]
View user's profile View All Posts By User
quicksilver
International Hazard
*****




Posts: 1820
Registered: 7-9-2005
Location: Inches from the keyboard....
Member Is Offline

Mood: ~-=SWINGS=-~

[*] posted on 2-1-2010 at 18:05


I have seen VERY expensive (EBW) units. The devices was developed from an Inverter (& yes it could run from a truck battery). It had a "reset" that was (ThBoMK) some type of a circuit breaker. It has "ARM" & "FIRE" (strongly shielded) buttons. and had several very serious "WARNING" stickers all over them and were in the two thousand dollar range.
They would work off of 110, 240 (AC) & 12V(DC) automotive. One warning stated it delivered lethal amperage/voltage, etc & to make all personnel aware of that issue to the wire-lead out roll.
When casually examining one I am fairly sure that someone with a bit of electronic engineering background could make one fairly quickly but it needs a quality inverter and some design room (box space) for extra items and hoods for the buttons & heat sinks.

I was asking questions about these things for awhile. They CAN be made by the home workshop hobbyist but the trick is to recognize that to really blow the bridge wire you need some complexities if the thing is to be safe & be usable more than a few times.

Another one I saw was gas motor driven and was actually portable (85+ lbs) but was also $3000+. I believe the circuit was an almost typical inverter with VERY heavy duty SCR, circuit breaker design (looked very custom but could be available or substituted).
This was not a transformer-type but designed like the inverter in a modern Microwave Oven as the heat sinks for the switching transistors were HUGE. Both units were NOT small, hand-held cap-discharge, easily portable, light units: they were beefy.

There was certain demands regarding the shooting wire (thicker than 20). There was unquestionably certain shorting resistors for the capacitors, which (I was told) had multiple protections for the parts. Everything about them was top grade and not cheap. They had long guarantees & were made (one anyway) in Norway. They took special care to protect the poles, buttons, etc.

I was told that a Microwave Oven inverter would be a good starting point for parts. IF you used an old fashioned transformer from a Microwave Oven and an inverter with short-circuit protections; that may come close. Using two of the BIG 50/60Hz caps for serious heat / spark, one of the biggest issues would be the safety issue. The rectifiers in Microwaves would handle things just fine. The current from those are .25 - .5 amps. But the commercial one will go hotter. All of them would be deadly however.

I've seen quite a few Jacobs ladders and Tesla coils from Microwave parts and the output would pop most any wire - way thicker than most anyone would even consider for a EBW.

I have thought of making one, but without the use of seriously expensive stuff. Especially since the guts of the modern Microwave have much of what you need. The way I would imagine it, -it needs to withstand a short: NOT for the bridge wire but if the long leads were to become abraded & short.

I was told that with EBW caps, the formula for required current was unique to the cap manufacturer & generally it was a individual setup of industrial blasting as a totally different thing from the series/parallel formulas common cap-discharge boxes.
Generally, single units were backed up with a standard cap backed up to it for surety and Nonel or detcord was used extensively for multiple or timed shots. The EBW issue was used for very safety-oriented, highly controlled agendas.... So a few amps could do it.






[Edited on 4-1-2010 by quicksilver]
View user's profile View All Posts By User
Contrabasso
National Hazard
****




Posts: 277
Registered: 2-4-2008
Member Is Offline

Mood: No Mood

[*] posted on 3-1-2010 at 10:08


The NATO shrike Mk V exploder uses 319 - 400v and therefore 6.8 - 12 joules

The Mk IV is 719 - 800v 35 - 48 Joules.


The clever thing about the Shrike is that it checks tht the firing line resistance is below 400 ohms ( MKV) or 350ohms (Mk IV) before priming. This means that you cannot fire it into an operator person. This sort of discharge is well into cardiac defib range.

Even for SFX or blasting 600v is plenty Many times 50v is plenty.
View user's profile View All Posts By User
densest
National Hazard
****




Posts: 359
Registered: 1-10-2005
Location: in the lehr
Member Is Offline

Mood: slowly warming to strain point

[*] posted on 5-1-2010 at 21:18


EBW requires a fast rise time. Below 1 microsecond is a figure I've heard; there have been threads on that topic here. Hockey puck sized semiconductors are remarkably inexpensive. An IGBT rated at 1200V and 400A is usually in the $10-30 range on EBay or from a surplus dealer. Any IGBT from the last 10 years or more will turn on in less that 300 ns if driven from a fast rise time source capable of supplying 5A or so.

There are electronic methods for sharpening pulses using things like saturable reactors and other magnetic or capacitor/magnetic units. Saturable reactor cores are in many current computer PSUs. I will admit not knowing how the distributed magnetic ones work - perhaps Mr. 12AX7 might point out a reference?

Getting high voltage is easy if you aren't in a hurry. A cold cathode fluorescent light driver (about $5 from a surplus dealer) and a fast recovery diode rated over 1000V will get you 500V or so from 12VDC at lowish currents. The output is in the microamp-milliamp range so be prepared to wait for a minute or three to charge 50 uF. It's inherently current limited making it fractionally safer than a MWO derived unit which can deliver 1KW or so at high voltage. A fast diode is imperative - a standard 1N4007 does NOT work.

A Cockroft-Walton multiplier would probably get you another KV or so at some cost of complexity (multiple diodes and some high voltage ceramic capacitors).

400-450V high surge rating capacitors are pretty easy to come by since they are used in a lot of mains-driven electronics in industry. 2 400UF at 400V gives you about 32 J. Snubber capacitors used in inverters need to handle high current pulses with fast rise time. They are usually in the 1-2 uF range so you'd need to stack 10-100 of them using low impedance (wide flat ribbon pairs) conductors.

View user's profile View All Posts By User
12AX7
Post Harlot
*****




Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline

Mood: informative

[*] posted on 6-1-2010 at 10:40


Nonlinear energy storage, like ferromagnetics and ferroelectrics, are interesting. Ferromagnetics increase the inductance of a transmission line except under large currents (where they saturate), so the average impedance is higher, which isn't too handy for this subject. Ferroelectrics increase the capacitance of a transmission line except under large voltages (where they saturate), so the average impedance is lower, which is more useful for this subject.

I don't have a clue what it would take to make a loaded transmission line blow a bridgewire. It might be interesting to play with. It would take a lot of playing to build something practical, though.

For this purpose, it's easier to take a primary energy storage medium (i.e., capacitor) and dump it into the load. Note you need this anyway, plus the transmission line has to store its energy for sharpening, so it's not really useful unless you absolutely need picosecond pulses.

As for switching, SCRs are kind of slow, on the order of 10us risetime. For quarter shrinkers, they're just fast enough, not quite performing as well as a trigatron. IGBTs aren't as robust and tend to explode at the slightest avalanche (they don't tolerate excessive forward or reverse voltages). With SCR type latchup all but eliminated from modern devices, you won't get quite as much conduction: even at 20V gate, you may get into current limiting and end up dumping your capacitor into the IGBT instead of the load.

Keep in mind it's very easy to get kiloamp discharges from a rather pedestrian capacitor -- it's NOT easy to conduct kiloamps through pedestrian semiconductors!

Tim




Seven Transistor Labs LLC http://seventransistorlabs.com/
Electronic Design, from Concept to Layout.
Need engineering assistance? Drop me a message!
View user's profile Visit user's homepage View All Posts By User This user has MSN Messenger
quicksilver
International Hazard
*****




Posts: 1820
Registered: 7-9-2005
Location: Inches from the keyboard....
Member Is Offline

Mood: ~-=SWINGS=-~

[*] posted on 9-1-2010 at 08:09


Tim:

You're right on the money! I asked a guy who is really into Tesla coils and he said the same thing. He said the best be is to find junk yards with old arc welders for many parts. My thinking was a cap bank from microwaves, keep the thing at 50/60Hz. & see if the switching could take it. I was told that's most likely too dangerous.....
But I was told there were better ideas. He gave me this, which I am passing on. I think something like this could be done, especially with a good parts selection. The layout seems workable.
Opinions?

Attachment: inverter_transformer_design_and_core_selection.pdf (1.1MB)
This file has been downloaded 591 times

[Edited on 9-1-2010 by quicksilver]
View user's profile View All Posts By User
Contrabasso
National Hazard
****




Posts: 277
Registered: 2-4-2008
Member Is Offline

Mood: No Mood

[*] posted on 11-1-2010 at 12:13


Do you have a specific reason for looking to fire EBW? There are almost NO civil uses for them and the military uses come too close to the UN Non proliferation treaty for comfort (unless a room with bars and a hard bed is your style).

A standard exploder for SFX or quarrying HE use will only put out about 20Joules maximum at around 250 - 600vDC there will be an electrolytic cap in there and a small DC -DC converter. Allow 2 seconds to charge each time.

The basic circuit is cheap and easy to build, Making it work reliably in all situations and keeping it dry is much harder an adds a lot to the build costs.
View user's profile View All Posts By User
quicksilver
International Hazard
*****




Posts: 1820
Registered: 7-9-2005
Location: Inches from the keyboard....
Member Is Offline

Mood: ~-=SWINGS=-~

[*] posted on 12-1-2010 at 13:58


Quote: Originally posted by densest  

A Cockroft-Walton multiplier would probably get you another KV or so at some cost of complexity (multiple diodes and some high voltage ceramic capacitors).


We were thinking along the same lines but I did some looking and a Croft Walton may not be able to handle the current - unless money was no object. Parts are a real challenge; even on an intellectual level of "dream-up" a EBW machine.
I DO have a source for very serious resistors. All I can come up with for Capacitors would be Microwave-Oven (05/60Hz stuff that COULD handle close to an amp) and some Tesla coil caps that would not be able to handle the current.
I thought of a pedestrian idea. but sometimes simple is where you need to be (cost?).
Get a small gas driven motor, from a old chain saw, grass trimmer, etc - & put together an inverter for 12v to 110 (depending where you live; I'm in the USA) and simply use the Microwave transformer with two or three caps in parallel with a high quality circuit breaker. Box it up, make it safe and keep testing if it will stand up to multiple wire pops.
Make two separate units in small boxes. One would be the gas-motor and inverter; the other would be the electronics to maintain safe distance of the gasoline from ANY spark.
Awkward but it could work without a rechargeable battery pack or a vehicle.

[Edited on 12-1-2010 by quicksilver]
View user's profile View All Posts By User
densest
National Hazard
****




Posts: 359
Registered: 1-10-2005
Location: in the lehr
Member Is Offline

Mood: slowly warming to strain point

[*] posted on 12-1-2010 at 17:39


A couple of notes: Mitsubishi published a "how to use our IGBTs" document "powermos4_0.pdf" which has a number of useful instructions on how to make your huge IGBTs last. They specify peak current at 2X continuous current. So if you want 1000A for a millisecond or so, you need at least a 500A IGBT. They also specify drive circuitry, drive voltages, layout, and other crucial details.

I'm not sure why people are discussing high power high voltage supplies. Unless you are trying to fire multiple times in a few minutes, a low current supply will charge even a very large capacitor. Diodes, etc. for 5000V are available for a dollar or two, new from a reputable supply house. Tesla is not the way to go for DC because it is difficult to rectify the pulsed high frequency AC. A voltage multiplier for 60Hz is easy to make and not expensive if you are only expecting 50 mA out of it. One for 1000Hz is not too bad. Finding high voltage fast recovery diodes is difficult and expensive which makes high frequency inverter supplies problematic, so bigger magnetics and lower frequencies are necessary. A neon sign transformer, an oil burner igniter transformer, etc. would give more voltage than one would probably need (5KV? 10KV? that's pretty harsh stuff), and those work at mains frequencies so at most four relatively inexpensive diodes would be sufficient. I'd probably use a variac on the input to keep the voltage to "reasonable" levels.

It would be instructive to check the self resonant frequency of one's capacitors to determine a pretty hard limit to rise time, as well as the equivalent series resistance for peak current.... Both are usually far worse than one might like.

I'm curious about EBW because of the possibility of doing shockwave imprinting on hard metals, for instance, without having to make and detonate high explosives. Emphasis on -without- explosives, even though this is a chemistry forum.

View user's profile View All Posts By User
12AX7
Post Harlot
*****




Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline

Mood: informative

[*] posted on 12-1-2010 at 19:58


Quote: Originally posted by densest  
Tesla is not the way to go for DC because it is difficult to rectify the pulsed high frequency AC.


It's tempting to put a vacuum tube diode at the top, heater power from a couple turns at the top, big volts DC at the anode (or ground the anode and insulate the other end of the coil if you want positive output).

But that's no big deal, a flyback transformer running at 15-150kHz will generate 40kV at up to 2mA easily (not that it lasts long at that power level). You can do it resonant (ala TC) or quasiresonant if you like, but an ordinary flyback works at those voltages.

Quote:
It would be instructive to check the self resonant frequency of one's capacitors to determine a pretty hard limit to rise time, as well as the equivalent series resistance for peak current.... Both are usually far worse than one might like.


This is the main reason for high voltage: low capacity means high SRF, and a relatively high ratio of sqrt(L/C) means you can couple it to standard 50 ohm transmission line.

There are other techniques to make stupid fast, stupid tall pulses on transmission lines (e.g., diode step recovery), but the short pulse width means the energy is small.

Tim




Seven Transistor Labs LLC http://seventransistorlabs.com/
Electronic Design, from Concept to Layout.
Need engineering assistance? Drop me a message!
View user's profile Visit user's homepage View All Posts By User This user has MSN Messenger
bfesser
Resident Wikipedian
Thread Pruned
8-7-2013 at 17:57

  Go To Top