Misfire Mystery

Have a code misfire cyl #3 (rear cylinder, driver side). Checked the plug it was cracked so replaced it with appropriate part. Old plug was Motorcraft Platinum and looked really ancient, suspect it was the original. New plug still not sparking, tested the wire got 6,400 ohms which seems normal, the wire to #3 is the shortest of the bunch at about 12 inches long so resistance probe is pretty straightforward.

Did the spark-in-the-dark test by unplugging each wire boot one at a time from the coil pack with engine running, much easier to do it there than pulling boots from those nasty rear plugs. Got the sparks (and really rough and hot idling as it fell down to 4 cylinders each time I pulled a boot) on all cylinders but #3. Pulling the boot on #3 has no effect on idling and there is no high-voltage stream of sparks between the #3 coil outlet and the cable connector, engine keeps idling relatively smoothly considering it has only 5 cylinders firing.

Logically at this point I should conclude that coil pack is bad and replace it. However in my experience a coil failure should cause misfire in two opposing cylinders, not just on one side. Is it possible that the coil is faulty even if only one cylinder is affected? Is there a way to do do a basic resistance test on a coil pack?

So I would appreciate your opinion(s) before I throw $140 on a new pack. Thanks in advance :smile:

Ok found the info on how to run basic resistance tests on the coil pack and both primary and secondary coils are within range for resistance and have continuity. However this only tells me that the internal wiring of the pack is intact, what happens under high-v load remains to be detected. But taking into account that sparks check good, wires check good, both primary and secondary coils have continuity, imho that only leaves one possibility. And it's not the PCM otherwise it would affect a whole bank (in this case Cyl #3 and 4) not just one.

Following the circuit path eliminating one possibility after another leaves me with a single suspect, something has to be amiss with the transformer in the 3-4 bank. Faulty isolator? Anyway I am now resigned to purchasing a new coil pack, I'm reasonably confident it will solve the problem.

.... However in my experience a coil failure should cause misfire in two opposing cylinders, not just on one side..... Is there a way to do do a basic resistance test on a coil pack? .......

My experience is different: A coil pack failure usually affects one cyl, but not its "sibling".

The coil pack has three coils in it, each coil supplying a cyl at each coil-end. #3 and #4 share the same coil. Swap the coil wires between these two cyls (usually easiest to do at the coil pack itself) and see it the misfire moves to cyl #4.

Good luck.

BTW If one checks continuity to ground at each cable terminal, he will usually find the coil in question has failed to ground ... usually at one end of the coil.

My experience is different: A coil pack failure usually affects one cyl, but not its "sibling"

Sorry my bad, I meant to say an external coil pack failure. That is to say a no-signal fault on one of the primaries will affect both "bankmate" cylinders.

The coil pack has three coils in it, each coil supplying a cyl at each coil-end. #3 and #4 share the same coil.

[anal mode] You mean there are 6 coils inside the pack, 3 primaries and 3 secondaries wound inside the primaries (in the magnetic field) but not in physical contact with them, in fact they must be isolated from each other. Coupling between primary and secondary coils happens through electromagnetic induction. Each end of the secondary coils feeds a single tower (part of a bank of two towers) in opposing polarity. [/anal mode]

This is very similar to most power transformers used in electronic appliances, but configured in reverse to convert a low-voltage/high-amperage current (roughly 12v/20A) into a high-voltage/low-amperage current. In such a configuration an isolator fault would leak amperage to the secondary coil. If I recall my high school physics of decades ago this could cause depolarization (is that a word? lol) slowly killing one of the impulses as the isolator deteriorates further. This would happen over time, starting with random misfires, progressively rough idling, jerky handling, etc, until finally a permanent no-spark condition occurs in one of the cylinders. Those are my symptoms.

Logically one should be able to determine if this is their problem by doing a resistance check between the primary and secondary coils of the affected bank. With no power being supplied this should be an open circuit (0 ohms), any resistance reading means there's an isolator failure. We can't replace transformers encased in the pack so I guess the only solution is a new pack. But at least we know why we need it.


Swap the coil wires between these two cyls (usually easiest to do at the coil pack itself) and see it the misfire moves to cyl #4.^

That was my first impulse but to my amazement the result was both cylinders went dead. WTF? This being a Ford it then dawned upon me that the procedure was too way too straightforward to please Ford engineers. So googled it for hours until at last I found the explanation. Turns out the original factory-installed plugs are polarized. 3 fire only under negative charge and the others positive charge, that's why they have non matching Motorcraft part numbers. Swapping the wires caused plugs to receive wrong polarity charge, and both wouldn't fire. Ford does this to save on platinum coating, however the replacement Motorcraft plugs have yet a different part number and work on both sides. Using an NGK replacement on #4 and swapping cables again the problem did move to #4.

Oviously I need a new pack, and I ordered one online. But being "proby" by nature I stopped by the local LKQ salvage yard and picked up two used coils for $20, actually they gave me the one from the 2000 and only charged me for the one I got from an '02 which looked brand new (the pack not the van) I had noticed the new cables and suspected a recent replacement at the other end so dove into it. The '02 has a different pack, the same as the Mustang. Power connector is different but nothin' a little splicin' can't fix so I got it as a backup. Also picked up the cables. Just Belden Premiun, but they are new.

The plugs should have still fired with the wires swapped, the platimum is only there to prevent erosion of the electrodes. Without platimum 1 polarity (I forget which) will erode faster, they do sell non platimum plugs for the windstar that work fine but do not last long.

PS: Never pull a coil tower or plug wire to check misfire, doing so can cause the spark to jump internaly in the coil and may damage the insulation in the coil. Always ground the plug you do not want to fire.

Yes, by "coils", I meant coils in the traditional auto jargon. What I should have written was "three transformers" ... but who is that correct?

I am also surprised that the original plugs did not fire when reversed. I agree with Kevin's comments on this. But I have never done this experiment myself, having changed out the plugs, early on, to "double platinum".

The plugs should have still fired with the wires swapped, the platimum is only there to prevent erosion of the electrodes. Without platimum 1 polarity (I forget which) will erode faster, they do sell non platimum plugs for the windstar that work fine but do not last long.

PS: Never pull a coil tower or plug wire to check misfire, doing so can cause the spark to jump internaly in the coil and may damage the insulation in the coil. Always ground the plug you do not want to fire.

If that is the case that they should fire anyway then my (admittedly tentative) theory is that the isolator fault is beginning to have an effect on both ends of the secondary coil in that bank. I know how it feels to drive on 5 cylinders, it is not that bad. Test-driving the van around the block today it felt like it was falling on 4 at times and that is just godawful, felt like some sort of massage bot. Also some fuel-smelling vapor was coming out the exhaust, which did not happen on 5 cyls because I had pulled power from injector #3 but not its twin.

The original platinum plugs on mine lasted very long, like 13 years now. Easy to tell they are the originals, the part numbers are reserved for factory use only. Official replacements are "EE" instead of "E" and "EG". I am replacing them along with coil and cables because finding one cracked means the others are probably worn as well. I am using NGK double-platinum because the Motorcraft plugs are too expensive considering the age of the vehicle.

I suspected that pulling the wires to check for spark was likely not the safest approach with regards to possible damage to the coil isolator but in this specific case I knew the isolator was already damaged and I had already decided to purchase a new coil... and I had misplaced my spark tester :redface:

Ok so I was too impatient to wait for the new coil pack to arrive tomorrow and the weather is nice so I installed one of the junkyard coil packs I got yesterday, the older EDIS 6 style one from the '00 that I got for free. Had to spend a few tedious hours cleaning it as it was laying in the dirt next to the van in the junkyard. Anyway I hooked it up and voila, everything back to normal, engine purring like a contended cat.

Contrary to some people I don't agree that the EDIS style coil pack used prior to '01 is more prone to failure than average. Physically it is one of the sturdiest packs I have ever seen, the tower connectors have an unusually secure design, though not the most practical kind. I noticed it is also sought after as an upgrade in many non-Ford cars and trucks, and for later Ford models with the plastic pack introduced in '01