flash bulb firing current characterization?

Thanks for your 90 mA data point and for your suggestion about GE/Sylvania. Unfortunately, but not surprisingly, neither of their websites (nor that of Westinghouse) had anything.

Just to clarify, do you mean the total amount of energy that has passed through the bulb?

And also, how exactly are they constructed internally - is there a bridge wire or do the connections go straight to the metal that reacts inside the bulb?

Thanks,

Hi Niall, Yes and no. The amount of electrical energy to set off the Mg + O2 reaction is what is required. Flashbulbs are basically a small oxidation reaction in a bulb. The by-product, light, and in rocketry's case, heat, are just the useful entities that are functionally used from this small chemical reaction. What actually starts the reaction is a small spark inside the bulb. So, it's not the total amount of energy that has passed through the bulb, just the energy that starts the reaction. While the reaction is going, the ions inside the bulb will conduct even more electricity, but this is not needed to actually fire the bulb; it's basically wasted energy. What is interesting, is you can get a bunch of flashbulbs and bundle them tightly together and then only fire one of them with electricity. What will happen, is the energy given off from the one firing flashbulb will fire the rest of the flashbulbs in a chain reaction.

Internally, there are two electrodes (the two wires that come out of the flashbulb) that make contact with the thin magnesium wire inside the flashbulb. The "atmosphere" inside the flashbulb is pure oxygen with absolutely zero humidity or other contaminates. When the flashbulb is energized, the magnesium burns in the oxygen rich "atmosphere".

Hope this helps... Daniel

inside the

Hi Niall, Yes and no. The amount of electrical energy to set off the Mg + O2 reaction is what is required. Flashbulbs are basically a small oxidation reaction in a bulb. The by-product, light, and in rocketry's case, heat, are just the useful entities that are functionally used from this small chemical reaction. What actually starts the reaction is a small spark inside the bulb. So, it's not the total amount of energy that has passed through the bulb, just the energy that starts the reaction. While the reaction is going, the ions inside the bulb will conduct even more electricity, but this is not needed to actually fire the bulb; it's basically wasted energy. What is interesting, is you can get a bunch of flashbulbs and bundle them tightly together and then only fire one of them with electricity. What will happen, is the energy given off from the one firing flashbulb will fire the rest of the flashbulbs in a chain reaction.

Internally, there are two electrodes (the two wires that come out of the flashbulb) that make contact with the thin magnesium wire inside the flashbulb. The "atmosphere" inside the flashbulb is pure oxygen with absolutely zero humidity or other contaminates. When the flashbulb is energized, the magnesium burns in the oxygen rich "atmosphere".

Hope this helps... Daniel

inside the

A spark? Would that not tend to imply a high voltage (or a very small gap) in order for the current to spark across a gap - I thought flashbulbs required a fairly low voltage and current to set them off. Do you mean that the current flows through the magnesium wire, causing it to heat up, thus starting the combustion(oxidation) reaction?

What form would this energy be transferred in? I'm guessing that they put out a lot of radiation in the IR band as well as visible light when fired, so this could be how.

Just curious really, never something I've come across before.

Also wondering, is the magnesium inhibited or coated in any way, since Mg ribbon stored in air develops a layer of oxide fairly quickly, so very fine Mg in pure oxygen would be even worse, I would think.

One more Q, I'm also interested in how you would use a flashbulb to ignite motors in a cluster - surely with composites you'd want the igniter to be at the head-end of the motor, which I wouldn't have though possible with flashbulbs unless you're clustering *big* motors! (maybe everyone meant BP clusters)

Thanks again,

The traditional method is to use flashbulbs in conjunction with Thermalite fuse. The flashbulb lights the fuse(s) and the fuse lights the motor.

The old FSI black powder F100's had a large nozzle and core. In some cases it was possible to insert a flashbulb all the way into the motor and light it directly.

I thought about this and no, I'm positive that it's NOT the magnesium wire heating up that causes the bulb to fire. This would require too much current. It's a TINY spark. High voltage is not needed to cause a spark as long as the elements are super close together.

Actually, flashbulbs release A LOT of UV radiation. UV is much more energetic than IR. IR is slow also. IR may be what ignites the fuse wick that ultimately ignites the motor, but it's the UV radiation that will start off the multiple flashbulb chain reaction. I just tried it to make sure I know what I'm talking about! The flashbulb that didn't receive any electricity fired at the same instant that the electrically fired flashbulb went off.

Pure magnesium; not inhibited. Magnesium oxidizes with oxygen plus humidity. Humidity is required for oxidation to occur. The humitity creates an acidic (or maybe it's basic) environment which enables the oxidation to occur. The elements need to be ionized to some extent for oxidation to occur.

The flashbulb ignites a piece of Thermalite (fuse) that then burns up into the composite motor. The thermalite is sheathed (sp?) with heatshrink tubing except for the ends. This prevents the motor from lighting anywhere but the top of the grain. One exposed end is taped to the flashbulb and the other exposed end is inserted all the way up into the motor. Tie all the thermalites together (one for each motor in the cluster) and tape them to the flashbulb for cluster ignition. Hope this answers your question. Daniel

How fine is the magnesium wire? Does it touch the electrodes? How close are the electrodes?

If you can pass 90mA through the bulb without firing it, what is the current flowing through if not the magnesium wire?

Sorry to bombard you with questions, but since you have flashbulbs and I don't..... :-)

Ok, well I'll trust you on that one...haven't studied chemistry since/any further than age 16, and I'm studying electronics now. (Perhaps hence the interest).

That sounds just like what I do with quickmatch for BP motors. Does the thermalite burn quite fast when sheathed? It sounds quite a lot like PIC (Plastic Igniter Cord, IIRC), which is available here in fast and slow varieties and burns hot enough to ignite composite motors.

Thanks again for your info,

I just Googled flashbulbs and found the following at

"What makes a flashbulb flash? Well, flashbulbs have, placed between their terminals, a piece of tungsten or zirconium filament. This wire is covered with an explosive primer paste. When current is applied, the wire heats up, igniting the paste, which then ignites the tin, aluminum or (in later years) aluminum wire (or wool). An oxygen atmosphere would increase the brilliance of the flash."

Also

seem to be a manufacturer/distributor of new flashbulbs.

Maybe new, modern flashbulbs, but not the old rocketry standard AG-1 flashbulb. Only magnesium, oxygen and two electrodes in those; no heating element, no explosive primer.

I have played with flashbulbs (AG-1) for over 20 years now. I know under what conditions where they will work and where they will not. If fact, I'm alive today because while experimenting with them as igniters when I was a teenager, I used to crush them thinking that they would transfer more heat to the propellant. Thank God that this ruined the flashbulb so it wouldn't work. Otherwise I wouldn't be writing this right now!

What I'd suggest is refresh yourself with chemistry and read about Activation Energy of a Reaction. The energy can be electricity, light, or any other type of high energy radiation. An AG-1 flashbulb firing is a great example of this.

You should also look into purchasing some AG-1 flashbulbs through Ebay. I have seen many English sellers offer them for reasonable prices. The best way for you to learn is to experiment with them yourself. Good luck, Daniel

Hey Niall, how to send you a PM? I tried your return address - it bounced.

Hi Daniel,

Perhaps you have AG1s from a different manufacturer.

I just crushed a GE AG1b (date code 8/1972) and carefully removed the electrode portion. And while it certainly was pushed up against the internal filaments, it also had an extremely fine bridge wire permanently bonded across the electrodes. With all of the mesh removed, I retested the resistance, which remained at the 1.0 ohms I had measured prior to disassembly. Then I connected 5 volts to it, and there was a very tiny flash as the bridge wire flamed up. The two now-separated portions of the bridge remained attached to the electrodes, still embedded in the bonding material. The bonding material did not burn.

permanently

retested the

That's very interesting. I have never seen that. Of course I'm kind-of a brute when it comes to taking things apart. I guess I learned something new today; thanks. If that's the case then ohm's law should be in effect for the firing of the flashbulb. I think what happens is that at 12 volts, your flashbulb will pass up to 12 amps; but at let's say 100ma, the bridge wire gets hot enough to start the magnesium - oxygen reaction.

I believe that's correct and appears substantiated by that "igniter characterization" study you mentioned earlier:

The current peaks near 10 amps for a very short period (about 20 uSec) until the bridge fire breaks, at which point the current drops to a level supported only by the burning gasses inside the bulb.

I was thinking about that document too...

What must be remembered is that the current figures from that test do not necessarily represent what is *needed* to fire each igniter tested. The test used a high-current supply (lead-acid battery), which means that effectively supply current and duration were unlimited (even small lead-acid batteries will deliver 10s of amps continously). In effect its a measure of the resistance of the igniters during firing (which is useful to know, especially for onboard electronics where your power supply is likely to be limited).

If the resistance of the bridgewire in a flashbulb or e-match is ~1 ohm, then sure, you'll get 10+ amperes current (initially) when you connect it across a source capable of delivering that sort of current. However, you may only need to deliver 1 ampere (or less) for a few milliseconds to actually fire the device reliably.

This said, if your electronics presents a low-impedance path from battery to igniter when firing, you will draw large peak currents - with (cheaper) PP3 batteries the current capability is quite limited (compared to NiMH packs for example) so you will, as the paper shows, pull down the battery voltage significantly if you present a low-impedance path between the battery and igniter.

I think this is why some altimeters offer a low-current output option. Since you don't need 10's of amperes to fire e-matches or flashbulbs (while staging igniters may be a different story), you can get away with a single battery, but you need to make sure you don't deliver more current than is really neccessary.

I personally like the idea of firing ejection igniters with a big capacitor supplied by a current-limited supply (i.e. a resistor in series with the igniter). This way when you open the MOSFET switch you get a high peak current, which will hopefully fire the device. After that you then have current sufficient to fire the device but not such that it brings your battery voltage down too much. I'm pretty sure the RDAS lets you do this.

Thats my thoughts anyway, feel free to pick holes :-)

The G-Wiz document would be well-supplemented with a similar set of tests to determine typical levels of current required to reliably set off the device. I think in many cases this will be much lower, and the current will be required for less time, since, in theory, once the pyrogen starts burning the current shouldn't be required any further. Of course this would involve the destruction of many igniters, so would be an expensive process. I also think that it would be best to test each igniter at distinct and separate current levels - i.e. attempt to fire it at 300, 400, 500, 600 mA separately rather than 'ramping up' the current until it fires, which wouldn't accurately represent what happens.

Another idea I saw (David Schultz I think) was to use step-up circuit to charge a capacitor to a higher-than supply voltage, allowing a very large pulse of energy to be delivered even from a low-current source.

I'll stop here, these are just some thoughts I've had, if anyone else has any ideas....

I think the answer is alot easier than you expect. Simply make up a sticker and put it on the controller. "Not Flash Bulb safe"

For a flash bulb, it seems that cap could be quite small. A quick calculation (using 10A, 30 usec, and a voltage drop on the cap from 12V to

6V over the course of the firing) yields 50 uF. Of course, if one only needs 400 mA, that drops to 2 uF (assuming it would still trigger in 30 usec, which I doubt).

Its not clear that one needs to current-limit the flow into the igniter. Perhaps its better to current-limit the flow into the cap (from the battery). That way the full energy of the cap is quickly delivered to the igniter, while the battery voltage never droops regardless of the droop of the cap.

Gosh, Robert, what an epiphany! I would never have thought of that. ;^)

Yeah, that is what I meant in my original post, but it probably wasn't that clear. To quote myself :-) :

"I personally like the idea of firing ejection igniters with a big capacitor supplied by a current-limited supply (i.e. a resistor in series with the igniter). This way when you open the MOSFET switch you get a high peak current, which will hopefully fire the device. After that you then have current sufficient to fire the device but not such that it brings your battery voltage down too much. I'm pretty sure the RDAS lets you do this."

Except where I wrote 'in series with the igniter' I really meant in series with the igniter, but between the capacitor and power supply. Just have to watch out for the time constant for charging the capacitor up again, but I wouldn't think that would be too much of a problem for most rocketry applications.

I did wonder whether it would be an idea to simply wire a large capacitor across the power supply terminals of an existing altimeter. This way when the outputs fired, the battery would initially 'look' like a much lower impedance source than the battery alone. I don't know how much this would help, since to discharge the capacitor its voltage has to drop, but in terms of quick energy delivery it could augment a 9v PP3 or such.