Ignition coil arcing from tower to ground post

My '83 400i has an ignition coil problem and I'm wondering if it's normal and/or if it's repairable. The previous owner replaced the Marelli spark box with an MSD 6AL and also swapped in an MSD Blaster II coil. (See attached photo.) The engine stopped running last Wednesday and I discovered the problem was that the coil had shorted. The short occurred because a persistent arcing between the coil's tower and its ground post had burned a hole through the epoxy tower, such that the arcing was shorting the secondary discharge entirely. The no-run condition occurred immediately after I had run the engine up to 6500 in first. I speculate that the arcing was so intense at that rpm that it burned through what must already have been a weak spot in the coil tower. There is a visible hole in the tower now. I will add a photo later.

The solution to the no-run condition last Wednesday was to swap in another canister style coil that I had in the garage -- a Pertronix brand "Ignitor" -- and the car runs as good as ever. However, when looking at the coil after dark with the engine running, I can see a constant faint arcing happening with the new coil as well. So this Ignitor coil's days also are numbered.

So now it's question time. Is this a common condition? Or is there something unusual about my car? I have a fix in mind: to swap in an Accel Super Coil. (See attached photo.) The Super Coil has a large distance between the tower and the ground post, and the cap molding even has a wall feature to prevent arcing. The price is $69 at Summit Racing, plus handling ($8.50?). So it's not a lot of money. But if there is a simpler fix, I would prefer to try it.
Image Unavailable, Please Login
Image Unavailable, Please Login

 

@ jm3: Questions arise: Are you saying a Chinese copy of the Bosch is acceptable? What is the carbon button? The spring? What is the "phasing" of the rotor? If my radio exhibits no interference problems, does this tell us anything about the resistor? Thanks in advance.

Edit: searching about a bit I am advised that the Bosch red coil is available for between $45 and $50, that there is no difference from one pn to another, and that one make/model to use when sourcing a Bosch red coil is 1974 BMW 2002 tii.

 

Here are some photos of my defunct MSD Blaster 2 coil. The second one shows the hole in the tower that I believe got burned through on the unit's last Blast to 6500 rpm.

This evening, in my effort to cure the on-going arcing between the coil tower and the coil's negative post, I measured the impedance of the "carbon button" in the center of my distributor cap and I got 10.3 ohms in the 200-ohm range. I checked a second time with another multimeter, a pocket-size one that has no range selector, and I got a wandering value of 12.5 to 13.9 ohms.

Does anyone know the correct value?

I also checked the Pertronix Flamethrower coil that's now in the car. For the primary I got 1.5 ohms on the 200-ohm range, and on the secondary I got 10.65 on the 20K range, i.e., 10,650 ohms. A web source -- http://www.aa1car.com/library/ignition_coils.htm -- says 0.4 to 2.0 ohms is the acceptable range for the primary circuit and 6,000 to 15,000 is the range for the secondary. So for whatever it's worth my Flamethrower checks good, at least until it burns a hole through the tower.

I wanted to check the voltage to the coil but I don't know how to do that with an MSD 6AL ignition. If anyone knows I would appreciate the advice.

Also, how do I check the other things that jm3 mentioned: the "radio suppression" condenser and the phasing of the rotor? I watched the MSD video and I don't see the difference between rotor phasing and base timing. In the case of base timing, the rotor stays stationary and you move the mounting plate on which either a breaker point set or a magnetic eluctor sensor is mounted. In the case of rotor phasing, using an MSD-specific rotor, the mounting plate stays stationary and you rotate the rotor to change its position relative to one of the plug wire contacts in the distributor cap. It looks to me as if both methods do the same thing: to change the relationship between a plug firing and the position of the piston.

Anyway, I appreciate all the help and I welcome any additional comments.
Image Unavailable, Please Login
Image Unavailable, Please Login

 

Searching the web for a spec for the correct resistance of our distributor caps' spring-carbon button assembly, I came across this page regarding the Jaguar V-12: http://www.jag-lovers.org/books/xj-s/05-Marelli.html . Evidently the Jaguars can suffer some catastrophic problems that arise when their Marelli distributor rotors fail and shut down one cylinder bank while the other continues to run.

The lengthy text on the linked page includes the work-around quoted below. It consists of pulling out the carbon button and spring and stretching the spring and then reassembling. The author also suggests turning the spring upside down. I am reluctant to make that kind of modification; I tend to think that engineers generally design how things should go together for a reason.

I note that the author claims that the resistance of the spring-button assembly should be "well under 1 ohm." If that is correct then the 10 to 14 ohm values I got from my cap are way off.

So anyway here is the workaround text. Please share your comments.

AVOIDING THE PROBLEM -- STRETCHING THE SPRING: I can see no inherent problems with the aftermarket cap; the electrodes are aluminum, but I've had no problems with aluminum electrodes in caps for other cars. Offhand, I would suggest that the aftermarket caps would be OK to use with one proviso: prior to installation, the carbon brush and spring should be pulled out of the cap, separated, and the spring stretched. You will find that twisting the spring counterclockwise makes it easier to remove and reinstall; a counterclockwise twist pulls the coils a little smaller.

After some fiddling with the one I had to play with, I suggest the following procedure: Once the spring is out, you will note that it has a large diameter coil in the middle (to hold the spring in the cap), widely-spaced coils tapering gradually smaller in diameter toward the carbon brush, and a smaller constant-diameter set of coils stacked on top of each other (no compressibility at all) toward the cap electrode. I suggest turning this spring upside down and using the constant-diameter coils to push the carbon brush. It is a simple matter to stretch the coils out so they provide some springage. It is a little trickier but still not too difficult to compress the diameter of a couple coils at the end so they will grip down on the carbon brush. Gripping the carbon brush is only necessary to get the cap on without the brush falling down into the works.

Once you have that spring stretched out to a total length of about 1-1/2", install it tapered-coil-end first into the cap, making sure the upper end of the spring is firmly positioned against the electrode at the bottom of the hole. Again, twisting counterclockwise will help, possibly holding the end of the spring in a pair of hemostats or small needlenose pliers while twisting and pushing. Then the carbon brush should be pushed back into the center of the spring. To get the carbon brush to snap in, it helps to turn it clockwise; a clockwise twist makes the spring coils larger. When done, the carbon should protrude at least 1/4" from the plastic each time it is pushed back and released. If you have a VOM, check the ohmage between the carbon itself (without compressing it back into the cap) and the electrode up on top. It should be well under 1 ohm, but it won't be zero because carbon isn't really that good a conductor. If it's an open circuit (infinite ohms), you didn't get that spring seated against the electrode; pull it apart and try again.

 

Hi Steve. Thank you for the hands-on info on resistance of the carbon button. I think I will cross that off the list.

I don't know if that is a crack in the tower of my defunct Blaster. It went out with the rest of the trash on Wednesday so we'll never know.

The thing is, I continue to have a constant arcing between the tower and the ground post of the Pertronix Flamethrower that is now in place in my car. The chances that two coils from two makers would have a crack or other defect to cause this are low. The arcing seems not to be present until everything is very well warmed up and I have ticked off a couple bursts to 6500-7000 rpm. (I don't know how high the number actually is because the tach needle continues swinging rightward even after the revs have audibly decreased with the shift from first to second.) But once it sets in it persists. There is also a vague feeling of surging at steady speed, as if I was driving on undulating pavement, but I'm not.

I have another work-around idea that I got from the Jaguar page, to squirt a blob of RTV silicone on the coil's negative post. After that I will try closing the gaps on the spark plugs, on the theory that this should lower the secondary voltage at the coil.

When you think about it, these single coils are asked to do a lot with a 7000 rpm V12. 3,500 sparks per minute: my calculator says that's one cycle every 0.017 seconds. It's no wonder if they overheat. This job is a lot more demanding than the 5,000 rpm V8's that most of the Flamethrowers and Blasters have to deal with.

 

I am going to do that this weekend. I'll post the results.

 

Today I checked the wiring between the MSD 6AL and the coil and I found that the small black wire is connected to the negative post and the orange wire is connected to the positive post. So I am still looking for an explanation for the persistent arcing from the coil tower to the negative post.

What I think I know at this point is that I believe that there is no resistor between the MSD and the coil. I also believe we have eliminated the button and spring in the distributor cap. The consensus is that 10 to 15 ohms resistance in that location is negligible in the context of 20K - 40K volts running through it.

 

Update: on July 4 I replaced the Blaster 2 coil, that had the hole burned through the tower, with a Pertronix "Flamethrower." The Flamethrower also had arcing going on between the tower and the ground post, as I reported in the original post to this thread. I loosened the eyelet-style connector on the ground wire and held it in place while I tightened the nut, so as to maximize the distance between the eyelet and the coil tower. After this adjustment the driveability problems that I had experienced with the Blaster 2 did not return, nor could I see any arcing while observing the coil after dark with the engine running, as had been the case before this adjustment.

This repositioning of the eyelet may have cured the problem entirely, and the Pertronix may have spent many happy miles under the hood of my '83. But we will never know, because I have broken my usual rule against throwing parts at a problem and swapped in a Bosch "red" coil -- picture attached -- costing $53 delivered from Amazon.com. If I have any further problems I will add to this thread.

The Bosch "red" coil has no internal resistors. I measured its primary resistance and it's 1.5 ohms. I am running it connected to an MSD 6AL, with no ballast resistor.

BTW, if you search for this coil, "Bosch coil 00013" is a good query.
Image Unavailable, Please Login

 

Well, fez1238, since nobody who actually knows something about these cars has responded to your question, I'll give an answer, on the theory that, in these on-line boards, a wrong answer is more likely to get a response than is the original question.

My answer is that the only reason that the Bosch "red coil" would not be suitable for a Marelli spark box is if the Marelli system requires a resistor coil. It is my understanding that the red coil is not a resistor coil. However, I also understand that the Bosch red coil is very robust and that the unit itself will survive in any application.

So there you go: a completely off-the-cuff and possibly wrong answer. If it *is* wrong, the corrections should come flying in.