Black Hole
May contain traces of nut
What does the team think?:
I had 3 coil packs* (out of 4) fail within about 30 miles**, had all four replaced and my car has been fine since then (the fourth was replaced as a precaution, it could have failed any time).
[ * Actually coil-on-plug, but everyone seems to call them coil packs regardless. ** Firstly one failed, and I had that replaced individually and thought everything was fine, but then I had to abandon a journey the next day and limp back to my repair shop with two more cylinders misfiring!
]
To me, this seems like an unlikely coincidence. A coil pack is just a transformer (and perhaps a diode), taking a low voltage pulse in one end from the engine management computer (ECU), and outputting a HT pulse direct to the spark plug, one per cylinder. Saves all that messing around with distributors and HT leads.
The only point of commonality is the ECU, the coil packs seem to be independent of each other, so why should they fail as a batch? My mechanic says that's typical, and a friend who's worked in car sales agrees, but why???
I put the question on a car forum and the yobbos piled in like it was a stupid question – of course they're all the same age so they're all going to fail around the same time, like light bulbs. My arse. That just served to remind me why I don't spend time on car forums any more. There has to be a physical explanation. Idiots.
Bulbs have tungsten filaments, which gradually evaporate as a direct function of duty hours. Like black holes, once it has evaporated enough the filament goes into exponential run-away and the bulb blows. The service life is predictable. So where's the failure mechanism dictating the service life of a coil pack?
Even if there is a failure mechanism, the batch-failure implies a manufacturing tolerance of 0.01% (failing within 30 miles after 350,000 miles service). Really??
The alternative is some kind of feedback which causes a cascade failure, but I can't see where that is coming from either.
I had 3 coil packs* (out of 4) fail within about 30 miles**, had all four replaced and my car has been fine since then (the fourth was replaced as a precaution, it could have failed any time).
[ * Actually coil-on-plug, but everyone seems to call them coil packs regardless. ** Firstly one failed, and I had that replaced individually and thought everything was fine, but then I had to abandon a journey the next day and limp back to my repair shop with two more cylinders misfiring!
]To me, this seems like an unlikely coincidence. A coil pack is just a transformer (and perhaps a diode), taking a low voltage pulse in one end from the engine management computer (ECU), and outputting a HT pulse direct to the spark plug, one per cylinder. Saves all that messing around with distributors and HT leads.
The only point of commonality is the ECU, the coil packs seem to be independent of each other, so why should they fail as a batch? My mechanic says that's typical, and a friend who's worked in car sales agrees, but why???
I put the question on a car forum and the yobbos piled in like it was a stupid question – of course they're all the same age so they're all going to fail around the same time, like light bulbs. My arse. That just served to remind me why I don't spend time on car forums any more. There has to be a physical explanation. Idiots.
Bulbs have tungsten filaments, which gradually evaporate as a direct function of duty hours. Like black holes, once it has evaporated enough the filament goes into exponential run-away and the bulb blows. The service life is predictable. So where's the failure mechanism dictating the service life of a coil pack?
Even if there is a failure mechanism, the batch-failure implies a manufacturing tolerance of 0.01% (failing within 30 miles after 350,000 miles service). Really??
The alternative is some kind of feedback which causes a cascade failure, but I can't see where that is coming from either.
Last edited:
