Ignition coil resistance

[Diver300]

This is on a Ford Zetec engine. There's a coil pack, with four HT leads and three LT connections. The coil pack is two double-ended coils as can be seen from the construction.

It's much like this:- https://www.ebay.co.uk/itm/316432985538 but it's made by NGK

The three LT connections are two primary coils of about 0.5 Ohms each and a common connection.

The HT connections for cylinders 1 and 4 have a resistance of about 8300 Ohms between them. The resistance between the HT connections for 2 and 3 is about the same.

That all seems fine. However, between the two HT coils there is a resistance of about 2.3 MOhm, which I'm fairly sure is wrong as it should be open circuit and indicates that there is a significant fault. There is no connection between the primaries and the secondaries.

Am I correct that there should be no connection at all between the two secondaries?

 

[MaxHeadRoom78]

The common arrangement now is the waste-spark method of ignition, where each coil feeds two cylinders with the primaries being isolated from the secondary now.
The may be a form of this , but a pair in one unit.?

 

[augustinetez]

Sounds like internal leakage in the coil pack = faulty - I'm assuming your getting misfire or low power problems.

 

[Diver300]

The common arrangement now is the waste-spark method of ignition, where each coil feeds two cylinders with the primaries being isolated from the secondary now.
The may be a form of this , but a pair in one unit.?

I've seen cars where there are two separate double-ended coils, but this car has two coils in one unit, like this:-
https://www.rsshop.com.au/product-page/zetec-coil-pack-ze001

I also found three coils in one for V6 engines.

 

[Diver300]

Sounds like internal leakage in the coil pack = faulty - I'm assuming your getting misfire or low power problems.

I am getting a misfire. It's intermittent. I just wanted confirmation that any conduction between the two secondaries is a fault.

On another car, there were two separate coil assemblies, so no possibility of any conduction between the two secondaries and they were in different assemblies. When one of those failed, there was conduction between the secondary and the primary. That was easy to confirm as I wasn't getting any spark on one cylinder, and I could compare the two assemblies.

I looked at several online videos about testing coil packs, and I didn't find any that checked leakage between the primary and the secondary, or between secondaries. They only checked the resistance of the windings. In this one, which I think has failed, and the previous on which I know had failed, all the winding resistances are correct and the failure is the insulation between the windings that should be separate. Checking the resistance of the windings would not show the faults that I have experienced.

 

[rjenkinsgb]

It's a common problem with those, I've had to replace one on a previous Mondeo.

The underside of the old one on mine had visible cracks in the encapsulating resin. It was causing an intermittent misfire, especially during acceleration.

Those encapsulated block types can sometimes also be damaged by someone doing a misfire test by pulling a plug lead - the open circuit HT voltage can be high enough to cause the insulation to break down.

 

[ahsrabrifat]

This is on a Ford Zetec engine. There's a coil pack, with four HT leads and three LT connections. The coil pack is two double-ended coils as can be seen from the construction.

It's much like this:- https://www.ebay.co.uk/itm/316432985538 but it's made by NGK

The three LT connections are two primary coils of about 0.5 Ohms each and a common connection.

The HT connections for cylinders 1 and 4 have a resistance of about 8300 Ohms between them. The resistance between the HT connections for 2 and 3 is about the same.

That all seems fine. However, between the two HT coils there is a resistance of about 2.3 MOhm, which I'm fairly sure is wrong as it should be open circuit and indicates that there is a significant fault. There is no connection between the primaries and the secondaries.

Am I correct that there should be no connection at all between the two secondaries?

The 2.3 MΩ resistance is not normal, and the coil pack is likely faulty. This leakage could cause weak sparks, misfires, or ignition instability under load. Replacing the coil pack.

 

[Tony Stewart]

I recall my '63 MGB stalled in a blizzard in Winnipeg with 1ft drifting snow -20'C windchill -35 on a lonely road late coming home from Uni. I opened the hood and tried the manual solenoid button under the hood and after a while it started. This repeated every few minutes until I noticed blue arcs over carbon trees in the plastic formed after a minute on the Distributor Cap. The spark occurrence was like night and day with intermittent for less than a minute. Then I realized the warm engine with an open hood and strong wind quickly froze the invisible moisture and elevated the breakdown voltage of carbon surface creepage. AHA, luckily I had some WD40 so I sprayed the distributor cap and wiped a bit to remove the invisible moisture and raise the withstanding voltage. Luckily I made it home to them confirm my failure in the plastic-burnt carbon-trails of the distributor cap to replace it ASAP.

The bottom line is that ignition coil is just a high voltage transformer that depends with magnet wire over cold-rolled grain-oriented steel (CRGOS) laminates with silicate coatings. The rating of each layer increases cost for thinner layers and also reduces the eddy losses in W/g of material and if the edges of silicate are broken, they only shunt the series capacitance of each layer but may increase eddy current losses. The ignition wire is high resistance carbon wire and I forget if the ratings are 10kohm/foot or whatever to reduce ignition current to extend plug life, EMI noise for AM radios as once ignition occurs , the arc risetime being negative resistance depends only on the speed of light and it becomes self-energized to burn for a longer slower duration proportional to the cost of the fuel which also reduces the impulse losses and slightly raises power transfer per stroke.

The conclusion to this anecdote is that contamination is the cause for leakage resistance in isolated coils which can significantly reduce the breakdown voltage rating of a transformer but is not an accurate measure of the cause of failure. If you had an HVDC generator and you could slowly ramp up the voltage until you heard a crackle on your AM radio (which when current limited might be silent) this is called a Partial Discharge(PD) in some part of the surface creepage in the insulation. and not a "full discharge". (When external and repeated every AC cycle, it is called Corona) This threshold is called the PDIV or PD inception voltage. Like any Relaxation Oscillator with capacitance to ground the feedback resistance when reduced slowy causes a inverse reduction in the discharge interval time and rapidly increases frequency. But this is a quasi-random frequency. The PD itself is like a self-healing wound igniting the weakest shortest path with the lowest breakdown voltage so the next arc interval is somewhat stoichastic or random but in a homogeneous mixture of surface contaminants limited in range. The resistance is not the cause a high voltage breakdown but rather a clue that the withstanding voltage of any insulation cannot handle the E-field in kV/mm. So if you are lucky the contaminant is external and can be easily cleaned and rejuvenated before the plastic carbon tree get burnt into it. Even Silicone RTV or spray can extend the life but access to the part is now labor intensive. They used to have spark plug testers in garages to see the arc shape, which requires a much higher voltage under 9:1 compression and transformer Meggers do not be high enough in voltage to test it. I have tested over 100 kV DC and it raised the hair on the back of my neck from more than 10 ft away.

 

[Diver300]

I've fitted a new coil and the car works fine. There is no connectivity at all between the two HT windings on the new coil.

The resistance between the windings on the faulty coil varies a lot. It was a lot higher after a day inside a warm house, which explains why the fault was intermittent.

 

[Tony Stewart]

Yes likely controlled by rel. humidity. But we’ll never know for sure where the fault lies.

When it comes to insulation due to the breakdown time called ionization which can be a microsecond or more. Insulation may withstand 10x higher impulse voltage than DC or line frequency before arcing occurs. In grid components this is called the Basic insulation Level or BIL. Thus a ceramic bushing over 300 mm long rated as BIL200 can withstand 200kV induced lightning voltage yet fail from surface contamination with 25kV DC.

This analogy also applies to automotive flyback transition times and spark plug ignition times.

Experienced auto technicians can diagnose faulty plugs or coils from the shape of the transient voltage.

 

[Tony Stewart]

The normal ignition voltage should rise with revving up an engine as the pressure buildup raises the breakdown voltage maybe from 20kV to 30kV for some engines. If there is an external insulation breakdown failure below this peak then intermittent ignition of fuel will result which then lowers the internal threshold.

 

[Diver300]

I cut open the old coil, and I found that there was nothing other than a couple of millimeters of potting compound between the two HT windings. The HT windings are separated into 8 sections which I assume makes it better at resisting the 30 kV or so.

Also the core down the middle of the windings is a magnet. I guess that is so that it takes more current to saturate the core in the direction that is to be used.