58 replies to this topic

[Pretty green flames]

It depends on the size of your ball milling jar.
Basicly you would want it to be half full of milling media.

As for how much green meal.....well i can't say since don't know the size of you ball milling jar.

Take care

PS: i think this may be the wrong topic for a ball mill.

Edited by Pretty green flames, 11 December 2004 - 04:57 AM.

[alany]

Today I pieced together this capacitive discharge firing box:



It charges a 130 uF capacitor to 340 V, storing 7.5 J.

A significant fraction of the 7 odd Joules is wasted in the flash lamp used as a thryatron, but most of it makes it to the e-match. A short stainless steel turning from a pot scourer explodes into a shower of sparks at the end of a metre of cat-5 cable.

The continuity test is implemented with a Neon bulb, which also acts as the charge indicator. The test loop current is 150 uA. The only disadvantage of this is that several hundred kR is sufficient to light it, even your fingers across the output will work nicely. In practice, even with a hundred metres of shooting wire the insulation leakage is far below lighting the neon. One issue with the way I have wired it is that there must be something across the output to light the Neon, open it looks just like it is discharged, so you must put a shorting link across it when it is not wired up to the shooting circuit - that's a bit of a safety issue, but there just wasn't any more room inside the box for a more complex circuit.

With the unit I can easily fire this composition through several metres of wire;

12 Potassium Nitrate
2 Sulfur
1 Lampblack (conductive)
1 Charcoal (airfloat)
+5% Dextrin

It results in an e-match that measures 45 Ohms or more and will basically not fire with a conventional system. (Don't use this composition it was only a test to see how insensitive the e-matches could be and still relibably ignite. I passed an Amp or two DC through it and managed to roast the sulfur out of it as a yellow vapour, but not ignite it. The CD unit exploded it with a sharp crack. It does make a good pyrogen for bridged matches though). My other bridgeless e-match compostion works just fine, as do bridged matches. With a bridge wire of 34# or finer you can use over 100 metres of wire (the longest I've tried - probably *much* longer).

[adamw]

Nice work. I would consider adding a 2-button safety feature whereby you must press 2 buttons at the same time to make a circuit to help prevent accidental firing.

[alany]

Nice work. I would consider adding a 2-button safety feature whereby you must press 2 buttons at the same time to make a circuit to help prevent accidental firing.

I'd love to, but I physically can't get another button inside. I was actually going to have normally closed button that shorts the output unless pressed, but murphy had other ideas. It still has to be charged before it can fire, so it is basically no worse than my battery based system once armed.

[evilgecko]

A significant fraction of the 7 odd Joules is wasted in the flash lamp used as a thryatron,

Did you consider using a high current SCR instead? It would mean most of the power would go into the e-match

[paul]

Nice work, alany! I work with a very similar ignition system. I built that handy little thing for a friend a few months ago:



Its uses a 320V rated photoflash-cap..... Yes, i know there is no second security switch, but in my next one I?ll integrate one The cap is loaded with a single-use camera electronic and two 1,5V micro batteries. With 3V the circuit lioads the cap way faster.

EDIT: And your conductive powder for ignitors sounds good. I?ll try that stuff

Edited by paul, 17 January 2005 - 01:05 PM.

[Andrew]

All sounds good, they look very neat. I've found that after a while the capacitors get knackered, when they are used for quick discharge systems. When I was at college, I used a photo flash module, in a remote fireing system for one of my A-level projects. In the end I had to add a way of changing the capacitor because they only lasted around ten to twenty fires. It would be interesting to hear how long your ones last.

[alany]

Did you consider using a high current SCR instead? It would mean most of the power would go into the e-match

Yes, I did the math, but nothing I had in the junk box had the surge capability to survive a dead short with the ESR of the capacitor I was using. I couldn't physically fit a large wirewound resistor inside the box to make what I had available safe. Dick Smith carries theTYN816 for about $4 which has an 160A+ surge capability in a TO220 package, with a relatively small value resistor in the loop that should be easy to design around. Maybe the next system I build, with multiple channels...

I was considering a system where the slats have an individual capacitor on each cue, charged from a HV AC source sent up to the slat from the firing control as part of the "arm" protocol. A simple test/arm/fire/safe controller for each channel could be designed, probably with a microprocessor in each slat also powered off the AC. You could implement the entire system over a single pair, with the control signalling being modulated onto the power carrier.

Andrew: Capacitor life is one thing that worries me in the long term. My current unit uses a photoflash rated capacitor, designed for pulse service into fairly low impedance loads, so I expect reasonable life from it. I've fired a lot more than 20 shots so far and notice no signs of deterioation yet.

Conventional HV electros from TV supplies aren't design for that kind of punishment, and even a single shot can ruin them from internal melting or fractures. By putting the capacitors in the slats (and perhaps running higher voltages) you wouldn't need such large devices and could probably get away with smaller values that could be implemented with cheap and robust capacitors.

[evilgecko]

On this mini protocol one I made...I used an SCR pulled from a TV. Its rated for only 10A but I fired it at least 50 times and its still going strong. I think because the pulse is so short they are able to handle it. Damn I checked Jaycar and the highest they go is 500V 8A.

[alany]

On this mini protocol one I made...I used an SCR pulled from a TV. Its rated for only 10A but I fired it at least 50 times and its still going strong. I think because the pulse is so short they are able to handle it. Damn I checked Jaycar and the highest they go is 500V 8A.

A modern SCR typically has a surge capability of about 10 times its rated current for 20 ms or so. It isn't wise to exceed the value on the datasheet for reliability, but they are usually conservative. You can do some di/dt analysis and with more detailed study of the SCR data sheet you may be able to safely get a bit lower source resistance, but that is pushing your luck IMO. Firing systems should be robust, they get abused in practice.

Yes, Jaycar only has a limited selection of SCRs at the moment. Dick Smith is carrying a bit better device for a little extra. You can always troll eBay for larger devices, SCRs with continous current ratings of over 4500 A are available, but they aint cheap or small. The C122E Jaycar carries has a surge capability of 82 A for a 50Hz half-cycle, or 90 A for a 60 Hz half-cycle (10 and 8.3 ms respectively). That could be made to work with a large enough protection resistor at the expense of some peak current - if all you want is high voltage to drive long/thin lines to conventional e-matches it would be perfectly suitable.

Assume you are firing the circuit into a short, which is the worst possible case. Work out the peak current that would flow, based on the ESR of the storage capacitor, ignore the SCR resistance. If that is larger than the surge capability of the SCR you need to add a few ohms of resistance to reduce the surge current to something the SCR is rated for. The resistor you use has to be mechanically robust or the current will open it almost instantly, a 5W wire-wound device is good, fortunately the 5 Ohms or so you'll need can be achieved easily in a wire-wound device with thick wire.

[evilgecko]

Are wire-wounds the big resistors which create resistance by making a magnetic field?

[alany]

No, wire-wound resistors are essentially ohmic, any inductance is parasitic, a side-effect of their construction. They are made from a length of high resistance wire, like nichrome and usually incased in a ceramic. The wire is coiled up to make it fit in a smaller area, that adds some inductance. Some wire wound resistors are wound specially to minimise the inductance, but they are expensive and relatively rare (some film resistors are also made the same way to reduce their inductance too).

The current handling capability of the relatively thick resistance wire in the device is what I want, not the inductance. The inductance will help reduce the surge current, but it will also increase the back-swing voltage which tends to be quite stressful on the dielectric of the capacitor, especially electrolytics. How significant that effect is will depend on the rise time of the disharge pulse, I don't think it will be too much of an issue in practice.

[evilgecko]

Oh so they are just big so they can take alot of current without overheating. Now I understand.

To stop back EMF I put a large diode in reverse polarity across the capacitor. I think you could even put it in reverse across the ignitor.

[alany]

To stop back EMF I put a large diode in reverse polarity across the capacitor. I think you could even put it in reverse across the ignitor.

Yep, but it needs to be quite fast, with a good recovery time and surge capability to be effective. I'd put it straight across the capacitor and SCR, that's what you are protecting. I don't think a more complex snubber network would be required, the load isn't all that inductive compared to say an electric motor.