Eating Ignition Coils?

[Randy]

I still love my A4, but I went through 2 coils on a 2 week cruise and we sailed more than ever!

The engine was running fine and then just stopped. This happened year before last and I replaced the coil and condenser and that fixed it. I had a couple of spares of each for this cruise so I got lucky.

The engine stopped (the second time we had only been going about half an hour) and wouldn't restart even hours later. Replaced the coil and, oddly, both times it didn't start first time touching the button but the second try it started right up like normal and ran fine.

Another problem is that the depth and speed instruments (2 and 3 years old) are starting to whack out. I suspect the two bugs are related.

I'm guessing there is resistance in plug wires, coil wires or somwhere along there but I'm just guessing.

Any ideas will be tried.

[rgoff]

Just a thought, but maybe you have a wire somewhere with an intermittent connection (at a terminal or inside the insulation). Replacing the coil might move the wire to a position where it works again. See if old coil was really bad.

Good luck.

[Randy]

My only problem is that I don't know how to check to see if the coil is good other than putting it back on. Any ideas?

[rgoff]

Using an ohmmeter you can measure for the proper resistance. There was a recent post on this. Search for "ohms".

Also, if you live in an urban area, your local auto parts store might check it for free (Kragen, etc.).

[Don Moyer]

Randy,

Your question is an excellent one, and I hope it will draw some interesting and valuable suggestions from our group. As you're suggesting, there are relatively straightforward ways to "functionally" test a coil while it's installed on the engine, but you're clearly asking for a check that you can make while holding your coil in your hand.

The trouble with functional checks of course is that they only tell us that a coil is working at the moment it's being tested, not whether the coil is going to work after it gets hot, vibrated for several hours, (or whether it's likely to fail during a two-week cruise). So, whether we're testing a coil on the engine or holding it in our hands, to be really valuable a test of the coil ought to also be qualitative so that it could tell us something about its "condition", and hence its reliability.

If I could ramble just a bit, I recall that during one of my last day-time jobs, I frequently used a meg-ohm meter to check the condition of the field windings in large submersible water well pumps. As their name implies, meg-ohm meters are able to measure resistance in millions of ohms, which is necessary to detect tiny leaks to ground through insulation.

The motors in our submersible pumps aged in much the same way that our coils do, which is to say that the insulation around the field windings would deteriorate over time until a short circuit would develop between the field windings and the outer case of the pump. Once reaching the outer case of the motor, the electrical energy that was normally contained within the field windings of the motor to do productive work, would take the easier path to ground provided by the water in the well.

In the case of the windings in our coils, the normal short circuit to ground is through the coil bracket to the block, which of course by-passes the route that we want it to take, which is across the electrodes in our spark plugs.

By being able to measure resistance in millions of ohms, a meg-ohm meter can detect even "normal" imperfections in the insulation. Each manufacturer of submersible pump motors publishes a specification for their normal meg-ohm readings, and those that represented an unreliable motor. Therefore, by simply checking the resistance of the fields on a routine basis, we were able to track the aging process of motors and remove them from service before they actually failed.

There are only two problems with using meg-meters in our application: They are frightfully expensive (approximately $600 even in the 1970s), and I don't know of any specifications published for ignition coils in terms of what would be an allowable leakage to ground through their internal insulation.

Don

[Jim Booth]

Don,

Here's a little math (or exercise in futility) that might yield a guesstimate for allowable leakage resistance between the HV winding and coil case.

The energy that gets stored in the coil depends on the peak current that the primary winding reaches during the dwell time of the points. E=(L*I^2)/2. The peak current I is calculated from VL = L di/dt where VL is the voltage driving the coil (12V), L is the inductance of the primary winding, and di/dt is the change in current divided by the dwell time. To calculate the stored energy, we want to calculate di to substitute for I in the first equation.

E=(L*(VL*dt/L)^2)/2 = (VL*dt)^2/2L ; VL=12V, dt=point dwell time, L is the inductance of the primary.

Once the energy is calculated, you could claim that the plug should still fire even if, as a guess, half the energy went to the wrong place. What does that mean for leakage resistance Rleak? Well, you would now have to know what voltage the high voltage side gets to, and for how long. Let's call it Vpeak, for some time = Tspark. Energy = Watts*seconds. Watts = V^2/R ; Vpeak^2/Rleak. E=Vpeak^2*Tspark/Rleak

Rleak = Vpeak^2*Tspark/E ;use E from the equation above.

Of course there are plenty of simplifying assumptions in all this, like perfect energy transfer in the coil, no primary resistance etc.

So now "all" you have to do is measure the primary inductance of the coil, and knowing what you do about point dwell, spark duration, and coil high voltage in the A4, you can calculate the leakage resistance.

Or you could keep a spare coil on hand! :)

Have a great weekend. You probably notice it's Friday and almost time to go home...

Jim

[Don Moyer]

For those of us with only modest computer skills, would you please provide a few more details on where to search for the ohms check procedure for coils?


Don

[Jim Booth]

Hi again Don,

I don't know where to find an ohms check procedure for an ignition coil. My note was just a "Rube Goldberg-ian" way an electrical engineer or physicist (like I studied to be in the distant past) might try to answer the question in the last sentence of your post ["I don't know of any specifications published for ignition coils in terms of what would be an allowable leakage to ground through their internal insulation."]

It was a way to address the problem of the $600 dollar meter, but it forced a much more expensive inductance meter into the mix, just to produce an estimate for Rleak that one would have to eventually measure with the ohm meter anyway. :rolleyes:

I thought most coils failed by shorts within the winding rather than to the case. If that's true a regular cheap DVM would probably work, if you could get the coil specs. Maybe measure a new one and use those numbers for a reference. Actually the cheap DVM's these days sometimes have a pretty high impedance resistance scale. Maybe they would work for the coil to case leakage resistance number too.

Jim

[Don Moyer]

Jim,

It's probably true that coils can fail winding to winding, as well as winding to case.

The reason I suspect secondary winding to case as the biggest culprit is that I intuitively believe controlling 24,000 volts to be more demanding than controlling the 12 volts in the primary windings. We also have had quite a bit of feedback from the fleet to the extent that if a coil is removed from the block and suspended on a non-conductive string, the coil has worked long enough to get back to home port.

Don

[Don Moyer]

Jim,

Thanks! I think we may have at least stumbled onto the reason that we continue to be so confounded over the condition of coils. Your astute explanation also reminds me of the many exciting years I spent investigating airplane accidents in our country's fine Air Force.

If, in the course of an investigation, we were finding a lot of evidence that pointed to an ignition failure in a reciprocating engine and we came upon the coil relatively unscathed, we'd rush it off to a testing laboratory and wait impatiently for someone like you to get back to us relative to the likelihood of the coil failing prior to impact. The local maintenance shop (like ours in MMI) wouldn't likely have had the instruments to make necessary measurements, and the accident board (like Ken and I) wouldn't have had the brains to do the math even if they had the measurements.

In the case coils in Atomic 4's, if we were only talking about saving the relatively low cost of replacing a coil, I'd say let's replace every coil in the fleet, and get back to our real work. However, the real cost of our not fully understating coils is the time and expense generated in delaying our finding the real "primary cause" when an engine fails intermittently. The cost is unusually high in the case of intermittent shutdowns, because we're likely to pay whether the coil is good or bad.

The scenario is quite simple. An engine shuts down unexpectedly. The coil is believed to be the cause of the shutdown, and (as luck would have it) there is a spare coil on board so the presumed "bad" coil gets replaced. After replacing the coil, the engine starts and runs fine, only to shut down again in a few days or weeks. At that point, most folks will quickly assume (since replacing the coil worked once), that they must have gotten a bad coil, so they replace the coil again. The engine starts right up, and the cycle repeats over and over, because in these cases, the real reason the engine is shutting down is due to something other than the coil.

If you sense an element of humor in this situation, it's only because when an Atomic 4 shuts down it doesn't usually result in a crash - like airplanes do. It's equally true however that when an Atomic 4 shuts down, it can get quite expensive since we have to add the cost of towing to the replacement of wrong things, trashed vacation plans, and most importantly, even in our "pleasure boating" experiences, an engine failure at the wrong time can turn into a life-threatening situation.

So, I don't think we ought to let the subject drop just yet. Jim, would you have time to stay in the lead in sorting out the feasibility of "condition testing" our coils a little while longer? I'm getting the idea that except for checking the condition of the insulation around the windings (maybe only of the secondary windings ?), the other checks would be very straight forward.

Is it wishful thinking that if we could narrow the scope of our insulation checking requirements, we could come up with a very limited (affordable) "Insulation Checker" that would work only on our genre of coil? Even as I'm writing this, I'm thinking that I recall seeing such an "insulation checking meter", but I have no recollection of what scale it was designed to check. Maybe our astute web-master could do a search for limited use affordable insulation checking equipment in cyberspace.

Meanwhile, the rest of us laymen could be checking in with our local auto parts stores and service stations to see what kind of coil checking service they offer. If any of you have a coil that you recently replaced in the aftermath of intermittent shutdowns, it would be a good idea to have that particular coil checked to determine whether or not replacing it really solved your problem. If the coil checks good, the real problem causing the shutdowns is likely still ahead of you - sorry.

Don

[Jim Booth]

Don,
Sure, it would be interesting to make some measurements to see what I can find out. I work at one of those big electronic companies so I have lots of equipment to use. My boat comes out of the water in a few weeks so I'll bring the coil home and make some tests then. It would be nice to get my hands on one that was diagnosed as "bad" also, especially one with a good primary winding. I agree the HV side of the coil should be the most fragile, and it makes sense to isolate it from ground as a first step if you suspect a coil failure.

Getting back to Randy's initial problem I'm reminded of the "fool me once shame on you, fool me twice, shame on me" adage.

Last year I thought I had a bad solenoid because if I tapped on it, the starter worked. I replaced the solenoid. That turned out not to be the trouble so now I have a spare solenoid and a new starter.

Randy reported electrical issues with depth and speed instruments as well so I would suspect low voltage wiring problems somewhere else, if not directly in the primary circuit. I think he's correct suspecting the two are related. Intermittent problems can be very difficult to track down. Since he replaced the coil and things worked again, I would first look very carefully for problems with any wire that could have been bumped during the repair, then at wires related to the gages. I'm guessing the bumped wire is the real problem.
Jim

[Don Moyer]

Jim,

Thanks for assistance in general and especially for your willingness to remain on our "coil project" a bit longer. I'm sure that there has to be at least one or two bad coils somewhere among our group that could be sent to you for analysis.

Don

[Marty Levenson]

Dear Don + Jim,

I'd be happy to send my failed coil if you can't find one closer...I'm in Vancouver, Canada.

-Marty

[Jim Booth]

I'm about an hour from Chicago. I would think there are enough A4's around Chicago for someone to have a bad coil to donate. My boat is in Waukegan Harbor. Anybody else?

Jim

[Don Moyer]

Marty,

Thanks for the offer. Jim will probably check in with an address for you to send the coil.

Don

[Randy]

if not the cause it might another symptom. I had a battery that wouldn't hold a charge. It was half-way through it's 5th season so maybe it was due.

I still think I'll go with electronic ignition and a new wiring harness. I just have a feeling that it's in that neighborhood. This boat has been hard on electronics since before I bought it. When I bought it the LORAN (yes Loran) wouldn't work while the engine was running and the depth sounder was finicky. (this is a Pearson 30) I replaced the fuse panel and all the wire from the new breaker panel to the busbar. I put in a ground bus and brought everything to that and traced down and diagramed all the major wiring aroung the engine. New depth sounder and knot meter. After that everything seemed fine for about 4 years. Year before last the engine stopped and I replaced the coil and condenser and went on my way, last year the depth sounder started getting flakey. This year the engine stopped twice and each time I replaced a coil and continued on. Don't we all love intermittent bugs? :confused:

[okred]

I have an A4 that was using up coils, they were getting extremely hot. This was completely resolved after a set of new ignition wires. I was running electronic ignition also.

[Don Moyer]

Okred,

This is very exciting! Could you please tell us which ignition wires you replaced? What type of wires were they, and what did you replace them with?

Don