Parallel is ALWAYS the way to go, other than with some e-matches, and here's why:

First, in series, you are dropping the voltage that each igniter sees. If you have a 12V launch system, with 3 igniters, each only sees 4V. Then, since with this low voltage, they may take some time to light, after the first lights, the other two loose continuity, and only one lights. E-matches are different for 2 reasons:
1) they require very little current
2) they light INSTANTLY

However, with regular igniters, nothing but parallel. Wired in parallel, each will see the full voltage availible, and behave as if it was hooked up alone. Even if one goes first, the others are still recieving power, and will continue to go.

Use parallel: it works.

Chris,

I am familiar with the arguments, and yours may be precisely why some with tons more experience than either of us recommend series wiring-- somewhat slower but more uniform heating of the pyrogen?

If I'm not mistaken Dave Bachelder is a series advocate, ask him his thoughts next time you talk, and get back to us. Like you, before I would have bet the house on parallel, but am now uncertain. Both work. Series may actually work a bit more reliably. ❓
JS

As an electrical engineer, I will strongly dispute the use of series igniters for the reasons that Chris gives. Unless there is a great deal more uniformity in the manufacture of igniters and ematches, the predictability of when an igniter pops is just not there. I've had igniters draw a pretty wide range of current when they finally pop and to make series work right, you'd need a helluva lot better predictability and repeatability from a product that is inherently designed and made to sloppily to provide that. E-matches may be a bit better, but... I sure wouldn't trust them.

Warren

I too had heard the auguments for using series e-matches and in fact I have tried doing it. For those of you who will be around Friday evening I have some close ups of motor ignition with the high speed cameras using series ignition. I used four 38 mm motors in the first stage and I was lighting them off of 30 V. Although they all lit, it was not simultanous. Now, I was not filming the ignitors but rather the motors coming up to pressure and it s clearly that some lagged others...

Doug

Doug,

Sounds interesting as in real data vs opinion based on theory. Though the fact that some came up to pressure later than others really doesn't tell us much given the variability in motor ignition to begin with. How many drag races have been lost bigtime even when the motors, ignitors, and power source are the same?

Warren,
not sure i understand why that variability in the igniters would lead to the conclusion that parallel is best. One could flip that argument around and make a case that in the case of a car battery, we are dealing with a current source, and that if one of three sinks has half the resistance, it will draw twice the current and heat much, much faster,(given that heat is proportional to the square of I, and will have an I^2R product twice as great) leaving the other two in the dust.....

I've never tried series myself, but thought it would be an interesting discussion as there are staunch advocates for both camps. Bob Krech gave a balanced view on a Texas HPR thread recently. My personal view is that simultaneous ignition probably depends more on other factors than series v parallel such as the propellant type, igniter pyrogen type, placement, etc assuming one has a decent set up and batteries. There have been notorious even catostrophic late starts using both. And there may be situations with multiple clusters that it may even make sense to do combinations of series/parallel. This would most likely arise in an airstart of multiple motors with a finite amt of amperage.

But a good discussion,
JS

The problem is that when they are not all identical in series, they will not all heat evenly (the one with the most resistance will heat first). Then, when it goes, the circuit is open and none of the others will go. However, if they are all in parallel, even if one goes first, the others still have current to them.

simultaneous ignition probably depends more on other factors than series v parallel such as the propellant type, igniter pyrogen type, placement, etc assuming one has a decent set up and batteries

I agree. One side aspect I wanted to do with the high speed cameras is some ground testing with ignition systems. Besides multiple ignition methods, just shooting the match some distance away from pyrogen, pyrodex, or even just the propellent as you described would be useful. Of course it would not be enclosed so it wouldn't be fully accurate since it would not be coming up to pressure but it still may give an insight to which method is best.

Doug

The problem is that when they are not all identical in series, they will not all heat evenly (the one with the most resistance will heat first). Then, when it goes, the circuit is open and none of the others will go. However, if they are all in parallel, even if one goes first, the others still have current to them.

But I think Chris you are still missing the point so i edited my post--lets assume a supply of 12v with very low int resistance like a car battery.
We have 3 igniters, two with two ohms and a third with one.

Parallel: The two ohm igniters each draw 6 amps, heating is 36*2 =72 J.
The third draws 12 amps, = 144J.

Series: All igniters draw 12/5=2.4A. The two 2 Ohmers then heat at 5.76*2=11.52J while the third heats at 5.76. The proportionality is maintained but all heat slowly igniting the pyrogen, there is no break in the circuit as is often feared, until the pyro is started.

Just playing the devil's advocate, as common sense isn't 😆
J

Yes, and therin lies the problem. As you pointed out, the one will heat 1/2 as fast as the other two. The other two will light first, and break the circuit, and the one will never light. Also, if the voltage is not high enough (3 identical igniters in series will only see 4v each), then none will light, or possibly one. This is about the worst case scenario in a large cluster - if only one lights.

My gut feeling would be to go parallel. Although I have no emprical evidence to support this. Estes ignitors which use nichrome wire for example burn completely through thus breaking the circuit. Don't know if this is a result of the motor ignition, the pyrogen material on the ignitor or the wire itself. It the first fire ignitors (or others) for example sever the wire in the same way, then it follows that a parallel configuration is best. Ignitors in parallel will draw more current at the same voltage than ignitors in series and would thus ignite faster causing quicker ignition of all motors.

kp

Assuming I do try and cert with H motors, other than the casings and reloads what else do I need?

Chris,
But how many series clusters do you personally know of where the circuit was broken? None I bet. How many anecdotal cases. maybe one? The problem is much more complicated than an ee 101 class. You have the rise time of the current to consider, the capacitances of the circuit drop in series, and triple in parallel so a factor of 9:1, the actual ignition point of the pyrogen, the heat capacity of the pyrogen, in other words its delta T/dT of the bridge, and the dependency of the pyrogen to spark at whatever temperature. Life just aint simple as a physics textbook, at least as currently written 😀

I'd say go parallel, as well, just cuz I don't know enough.
JS

I have never seen one in a rocket that failed this way. I tried 3 estes igniters in series, however, and 2 lit, one did not. I then tested the one to make sure it was good - it worked instantly. Never tried it with HPR igniters, but it should work the same way as the estes.

Never tried it with HPR igniters, but it should work the same way as the estes.

I would never make that assumption and I rarely use the word never. If there is one thing I've learned it is that what works for model rocketry rarely works for HPR.

Doug

All I am assuming is that their resistance characteristics are similar. I know that they break when fired(I have tested that), and I know the resistance is generally significantly lower, but if anything that would seem to make them less reliable than estes when in series(they need more current to fire).I am not assuming that they are the same in all ways, just that the resistance and temperature characteristics can be scaled (A reasonable assumption if the HPR igniters use nichrome - not sure about conductive pyrogen).