Stan944

This is an update on my troubleshooting of stalling or not starting engine problem:

I installed a new ignition coil, and for about 10 days the car didn't stall even once. But eventually it did. Using ignition-wire tester I figured that the voltage from the coil wasn't consistently making through the distributor/rotor. The spark test showed poor spark (gap of less than about 6mm required for a regular spark). In fact the spark from the new coil was considerably weaker than with the coil I replaced earlier!

So I installed another ignition rotor (not a new one, but from previous maintainance job), and the spark got better (about 12 mm gap was still OK).

I think through the 10 days of driving with the defective ignition rotor (I think it was shorting to the grounded cam shaft) I fried the coil. I think so because shorting to ground through the rotor doubles the coil current (and for times more heat generated on the coil). This is because the coil resistance is 6 kOhm, wire_from_coil_to_distributor = 0 Ohm, wire_from_ditributor_to_spark =6 KOhm, spark_plug = 100 Ohm.

What do you guys think?

P.s. I'm not sure about my conclusions, because for these 10 days the temperatures here were about -25C, and very cold air might have hidden my problem.

Peckster

How old are your wires and cap?

Stan944

The cap is about 1 year old (so was the rotor), but the rotor might have partially broken when the bolt that holds the cam sprocket got loose. But it was several months earlier...

The wires are not too good: I replaced the ones from the distributor to the spark plugs with a generic fit about 1 1/2 year ago. But the wire between the coil and the ditributor is very old, more than 3 years, i.e. before I bought it. I should replace them too, although I don't see a spark show in the dark.

Stan944

Does anybody have experience with installing a coil from a different car? Not necessarily Porsche-related experience, but in general, e.g. Volkswagen and Mercedes? I imagine I would have to match the coil primary resistance (~0.5 Ohm), and inductance (don't have an easy way to measure it). Are coils similar from electric point of view? I could use the spark wires from that car model too, so that the plugs are matched.
I wonder which car could be a close fit? Volkswagen, Audi?

njwilser

Replace all the wires. When I put on a set of new magnecores I first did the four leads and not the coil wire (had trouble w/ the cover on the coil). Car ran better but a couple weeks later I replaced the coil wire and the car ran better still. Decent quality wires are essential.
MW

Stan944

I might have to replace the wires, but I can't buy Porsche stuff now. I'm broke now, and that's why I'm asking for experience with swapping ignition parts between different brands...

Stiil, does anybody want to comment on possibility of me having destroyed the new coil?

Stan944

bump

Zero10

I bought a coil at NAPA, which is virtually identical to the 944 coil. It was $12CDN. However I believe I lost the box a long time ago. Like the porsche coil, it required resistors on the plugs (or in the wires), and was very similar in physical dimensions. As a temporary fix, or for troubleshooting purposes, it would suffice.

The only real diference you might see, is it may build capacitance at higher frequencies (higher revs), and may have less windings on the secondary side, i.e. lower voltage output, but I'd bet it's rather close.

If it wasn't part of my spud gun, I would send it to you.

Jfrahm

You should replace the coil wire. It should be possible to put your beru ends on generic ignition wire and have a like new part. As for the coil, many people use the MSD Blaster 2 which is $45 or so USD. I would suppose a coil off an old Golf would work as well. Auto parts stores might also have a generic coil too. But didn't you say with the new coil and a good rotor it's working?

-Joel.

Stan944

Thanks Zero10,
I can check NAPA on Saturday. I sure could use it for testing for $12CDN, even if it's not completely good. What kind of plug would I need to connect to the distributor; same?

I don't understand this; is it related to impedance Z=R+iwC+i/Lw ?, and what are the consequences? voltage drop on the coil at higher rpm due to increased impedance?

Are there huge differences between coils from various cars? I believe most cars have resistive wires to suppress noise pickup by the radio, but I don't know that for sure. I would think most cars from the eighties (i.e. before invention of distributorless ignition) would have similar voltage on the output.
From physics course:
V = -L dI/dt
so you could have different combinations of inductance L and time derivative of current dI/dt, but maybe there is a standard?

Stan944

I guess I didn't make it clear. with the new coil it worked for 10 days, but then the problem reappeared. After putting a different rotor, the spark test showed improved spark, and the car drive more smoothly(!), but it kept stalling.

I'll check Golf.

Zero10

Perhaps I used the wrong term. Some of the cheaper coils act like a small capacitor, so when the revs get high, you no longer get nice 40kV sparks, since the low tension side may remain close to +12V since the pulses are so quick, and you will get virtually nothing on the output terminal. This will happen to every single coil out there at high enough revs, but on cheaper coils, sometimes it will happen in the high end of the RPM band.

I run my cheap-o coil at around 300Hz, which equates to roughly 2400RPM on a 4-banger (4800 pulses per minute, 2 pulses per revolution). I tried it up at 3000Hz, (well over 20,000RPM) and it did mess up, that was using a solid-state relay as an oscillator, I reverted to a mechanical relay to drop the frequency.

I would bet it's fine well past 10,000RPM, so I wouldn't worry about it.

You will sometimes find in older coils, that they have started to act more and more like a capacitor, due to the core in the transformer coming apart, allows larger eddy currents to circulate, which creates erratic output on the transformer.

That is somewhat sidetracked. Long story short, the cheap-o napa coil should be fine for testing. It's a little smaller than ours, but I bet it would mount in alright, and it would be easy to swap in/out to verify a bad coil, maybe even on other cars?...

Most coils have a very similar output voltage, which a lot of vehicle manufacturers will not admit to

Most are in the range of 40kV. I had my particular coil measured at 43.56kV

Stan944

Thanks for the explanation Zero10.

I sure hope they have this cheapo coil in NAPA here, cause swapping coils with any German car I checked makes no financial sense: most are $199 in Canadian Tire. So it would be cheaper to get another exact fit from US. The cheapest domestic was $70, which is not cheap either.
Wires: I haven't found cheaper wires than $130 CAD, and the connector that goes to the coil is rather unique, and I won't be able to fit a generic one. Unfortunately, the old wire connectors are in poor shape, so I wouldn't be able to reuse them. And buying an unknown quality used set is not very advisable, since I have enough troubles now.

I think that is my case too, as every now and than I see a brighter blink on the neon spark wire tester.

I just noticed, sorry to hear it.

More comments are more than welcome...

d_roedel

i had a very similar problem on a friends car, no spark and would run for a little while, turns out it was speed sensor on the flywheel, but also check the other sensor and the connections of all the sensors near the back of the cam housing, should be a couple of plugs help up by a bracket,

hope this helps you

Stan944

I think it's not my case, because:
1) the tachometer works while coasting in gear after the engine stalls,
2) I have LEDs mounted to monitor the injector pulses and pulses at the primary circuit of the coil: they both blink when I crank the engine, even though it doesn't pick up.

I can't rule out occasional missing pulses (Danno's suggestion), but I find it less likely here. Could be bad synchronization between the injector and the coil pulses, but that's hard to check without a scope.

In a moment of frustration, I thought about taking the car for profesional diagnosis, but the problem is intermittent, lasts from several minutes to 1/2 hour typically, and it would be hard to have the problem occur when I take it for the diagnosis. And then they would probably replace everything that even remotely could be suspecious. Guess what, the value of this repair would probably approach the value of the car... So I'm stuck with trying to figure it out, natuarally with rennlist help, which I appreciate a lot.

One thing I know for sure, the ignition rotor was defective (shorting to ground I believe), but I don't know if this was the problem, or a problem.