Testing coil packs

[mcandmar]

Has anybody devised a method of testing coilpacks? I borrowed a set which confirmed mine have a problem, but i cant figure out which one is at fault. Problem is they work fine for daily driving and only seem to start playing up under heavy load or 1bar boost. Even then there not bad enough to missfire, they just feel fluffy..

I'm trying to work out how to build a test rig using one of the spark testers below. I'm working on the basis you can keep increasing the gap until they miss fire and then compare the results of all five and hopefully highlight a weak one of the bunch. Anybody BTDT?



Cheers,

Mark.

[UrS4boy]

You might want to read this procedure that I wrote up a few years ago for the C4 UrS AAN crowd on AudiWorld.

http://forums.audiworld.com/showthread.php?p=18838360

Basically, you are going to remove the coil pack from the engine, install old plugs in the other 4 boot connectors, ground them to the cam cover and use the spark gap tester that you show to see which coils are the weakest (can only jump the spark over a small gap). Make sure that you pull the fuel pump fuse before you get somebody to try to start the engine (it won't).

Replace the weakest coil(s). Or just switch to another coil set up, e.g. 1.8t coils from Apikol or LS2 coils from EFI express.

[S2central.net]

This Q has come up before - IIRC an expensive conductance meter (reciprocal of resistance) is required to to do this accurately - but these aint cheap and are therefore hard to justify by any one person.

I do like the idea of a test rig to spot potential weak coils by the method of increasing gap. Could be lashed up with a simple timer circuit, set for correct pulse shape, pumped thru a couple of POS units and onto a jig to earth with coils and plugs.... Mmmm might be a fun project if I had time.

Only small issue might be that all coils could run perfectly at ambient heat and pressure - may be almost impossible to visually spot differences on the bench - in saying that it would be worthwhile to try out the idea - as its the only way to know if 'bad coils' can present themselves under such a test.

Care is needed on the circuitry so as not to overheat the coils - would be worthwhile building the driver circuit from astable (timer circuit) that can be programmed to fit the basic shape of pulses coming from the ECU.

I'd be game to certainly help with this if required.

[mcandmar]

Perfect, so the theory does work in practice. Can you remember off hand the range of gaps sizes you were seeing? Would be useful to know as a baseline.

Good point on the ambient heat as they do deteriorate as they heat up. Guess i'll try it with a fully hot engine and a pair of gloves. I had originally thought of bench testing them but i have no idea how to simulate the signal from ECU to POS unit. I only sort of follow you on the pulse shape stuff...thats all over my head i'm afraid.

Cheers,

Mark.

[S2central.net]

Good link from Dave F there - have a vague memory of that... should be useful way to spot the weakest spark by grounding on the engine as stated. If that works well in ambient heat, light and pressure then it would make the notion of building a bench based tester more of an academic exercise.