Mondial 3.2 ignition troubleshooting

Things I learned today:

• My exhaust temp warning thermocouple system doesn't work.
• My tool roll seemingly includes two different styles of lug nut wrench but no spark plug wrench.
• The tow loop in my tool roll is of a totally different size than the threaded boss on my car.

...which is to say, that I seemingly lost ignition on one bank of cylinders, as it was intermittently down on power, and when we stopped, the muffler was glowing red hot. (But it fired right up again to drive onto the flatbed after we discovered issue #3 above.)

So I'm posting here to see if there are any failure points known to be more common than others, where I should start troubleshooting, before I tear into everything (at this point I don't even know which bank was affected).

One [now-] suspicious clue is that when I bought the car last year the box of parts in the trunk included what look like a pair of new-in-box Hella 5 DA 006 623-43 ignition modules, suggesting that maybe I'm not the first owner to encounter an intermittent ignition problem! But I'm a little confused because it looked to me (both under the hood and on the wiring diagram) that the ignition module in this system is built into the coil assembly—can I remove that portion and bolt on these spares? (If so, can anyone confirm that the aforementioned Hella part [or H-5DA006623-431, as it appears on the outside of the box] is a legit cross-reference?)

Thanks in advance,
Joe

 

Thanks—that's certainly easier to reconcile with what I was seeing on the wiring diagram!

Thing I learned today: I don't own an 18 mm (11/16") spark plug socket. (I was hoping maybe the condition of the spark plug electrodes would give me a clue as to which bank wasn't running, although it seemed to be running totally fine when I drove it on/off the flatbed, so that probably destroyed any such evidence.)

Is there a resistance spec for the coils that will tell me if one of them might be the culprit? Is there anything to be learned via the diagnostic connector shown on the wiring diagram? (Why do I feel like I'm going to end up driving endlessly in circles around the neighborhood trying in vain to get the failure to reoccur?)

 

I gather that the plug wires are secured to the distributor caps mechanically (I think I'm seeing little screw heads in there), but what about the coil wire? Is that one supposed to just pull out, or am I missing something? And how springy is the carbon button in this distributor cap supposed to be? Mine is springy, but barely—only a couple millimeters of travel.

Thanks again!

 

Well my button (I've only pulled one distributor cap thus far) only protrude about a millimeter—so that's definitely getting replaced—but I'm frankly not sure I can even manage to grip its edges with to pull it out!

I found a statement on an older thread saying "On the microplex system one coil sends a signal to the fuel pump circut to run the pump. If you install the bad coil on that bank the motor will not run." Is this true of my car? Should the fact that it was still running on the other 4 cylinders tell us which bank wasn't firing?

 

Well, I'm seeing essentially infinite resistance (except briefly, if I wiggle it just right) from the carbon button in the aft distributor cap to the coil end of the coil wire, so it looks like I found my problem! But that little carbon button is definitely too worn down to grab ahold of, and I haven't had any success trying to pry it out with a pin, either. So I guess I know how I'll be spending this weekend... maybe clamp it up in the drill press and try drilling into the button until it shatters?

 

A toothpick didn't have nearly enough surface area to pull the thing out via gluing, but a 2.5 mm diameter dowel did! Thanks for the suggestion!

Also at your suggestion, I checked the resistance on the coil wire after separating it from the distributor cap, and it's measuring >800 Ω. But so is the other one, that checked out when I measured from the carbon button—is this by design? (Is this why the actual conductor is so much tinier at the distributor end than at the coil end?) If so, do I need to know anything before I cut off the chewed up end (it looks like it wasn't pushed into the distributor cap all the way before someone tightened the little screw into it) and pierce it anew? (Or—now that I realize these are available separately from the complete wire set—does it make more sense to just replace the pair of coil wires?)

Thanks again!

 

I've only checked half of them so far; stopped when I found one with seemingly infinite resistance. Three others are 13 kΩ, 0.6 kΩ, and 37 kΩ, which is a bit more of a spread than I had anticipated? What's going on here? Is the outlier my only good one (based on Steve's 700 Ω/ft)? Of course I'm especially curious what to do with the infinite resistance one; surely there's just a marginal connection somewhere that the ignition voltage can overcome, because I'm pretty sure it was never running on just seven cylinders. How are the extenders attached? Do they just pull off, or is something threaded?

Thanks again!

 

Well, after a second one measured infinite resistance I went ahead and cut them back and reinstalled them to the distributor cap, and they're still measuring infinite. But I guess I really shouldn't be surprised; if ignition voltages can jump to ground from a deteriorated cable, of course they're going to be able to jump an internal break in the wire somewhat seamlessly.

So it's time for new plug wires, and Superformance appears to have the best price (at least among legit OEM-looking sets that I was willing to trust), so I went ahead and ordered everything I've been putting off buying from them (including new distributor rotors--downright affordable at their price--especially when it's expressed in UK Pounds) until I had a good reason to justify the shipping costs. So I guess I won't be driving the car this weekend, but I'll eventually have mostly-new ignition components (I already had the aforementioned transistor modules from the PO and a new set of plugs on the shelf) as well as fun stuff like Superformance's facsimile of the OEM 127589 tool roll!

But speaking of tools--can anyone recommend an 18 mm spark plug socket that actually works? The first one I bought last week (Amazon) had too loose a grommet, such that it was no better than a regular 18 mm deep socket, and the second one has such a tight grommet that it stays attached to the plug and pops out of the socket (even after I hot-glued it in place)! At the risk of sounding like Goldilocks, what's a guy gotta do to get an 18 mm spark plug socket that works at least as well as the hand-me-down 21 mm and 16 mm ones in my toolbox??

 

Wow; I can't say I've ever seen an internally-shorted distributor cap before. (Moisture infiltrating micro-cracks in the plastic would have to be the only plausible explanation, right?)

In any case, thanks for all the tips. After numerous distractions (most not Ferrari-related, but I did take the opportunity to install an aftermarket oil cooler fan where my stock one was missing, while I had the coil assemblies off the car) I finally put the car back together with the new ignition modules, plugs, wires, distributor rotors, and—crucially—carbon buttons in the distributor caps. And, thankfully, everything seems to run fine now.

I was too lazy to replace the oxygen sensor preemptively, but paranoid enough that I wanted to measure the frequency valve duty cycle after the car had warmed up, and to compare open-loop vs. closed-loop operation. This is the first K-Jetronic car that I've owned, and so my knowledge of this was purely theoretical until yesterday—is there a definitive explanation somewhere as to the best way to measure this? (And yes, I realize this probably warrants starting a new thread, but if there's an existing one that I just couldn't find, I figure you guys can just point me to that one.) I used a digital multimeter set to the 4-cyl dwell scale, and got a value of 62 (69%?), which seems reasonable enough, except that when I grounded the full-throttle switch to test open-loop operation, the value dropped, which I wasn't expecting, to 59 (66%), and the idle sounded a little smoother (and maybe a tick faster), which I really wouldn't have expected as a consequence of leaning the mixture. So it stands to reason that my oxygen sensor is, in fact, toasted, and the ECU is responding to a constant low-voltage sensor output by enrichening the mixture, but I am curious if my measurement is just totally flawed, for a couple of reasons: 1) my multimeter also has a duty cycle mode, which produced an unrealistically low value, 2) some internet sources (now I forget if it was something I found on here or elsewhere) say that you can't actually make this measurement on the frequency valve circuit itself, and need to go to pin 17 of the ECU for a clean square wave, and 3) one of my (third-party) Bosch books says to hook up the dwell meter between the valve and ground (which actually makes more sense, if you think about how a dwell meter is supposed to be used, suggesting that my measurement may have been backwards, except that an open-loop duty cycle of 35% makes no sense).

Thanks again!