Ignition Troubles
 

I love you guys.

My car's having some starting trouble. When I lift the sirflos sensor plate while the key is on, I hear the injectors spray, so I'm pretty sure it's getting gas. It doesn't want to start at all, so I think I'm getting no spark. Tomorrow I'll have an assistant, but in the meantime I noticed that I do not have voltage at the non-ground post on the coil. Looking at the schematic, I notice again that this ignition system is different and maybe that post is not supposed to have voltage just with the key on.

It looks like Terminal 15 of the CDI is supposed to have voltage, and I'll check that tomorrow. My green wire looks fine. If I get a pulsing millivolt signal at Terminals 7 and 31d at the CDI box, and if it's getting voltage at Terminal 15, and a good ground, then it should work. Right?

I guess I'm looking for a CDI quick-check method. I hope you're out there, Warren.

Jim

Jim,

Yes, if you have low AC Voltage at terminals #7 & #31d, as well as +12 Volts at #15 of the CDI unit ... you should get a pulsing 350 - 460 Volt signal at the 'A' terminal of the CDI.

You can try looking for the HV pulses at the coil with one of those neon electrician's test lamps ... ground one side and connect the other to the 'A' terminal output. It will be a dim signal, and only on one electrode of the neon bulb [DC illuminates one side, AC -- both electrodes] ... should be visible in a dark garage.

http://forums.pelicanparts.com/uploa...1106788931.jpg

Thank you, Warren. As a followup question, does anyone know what the resistance values should be between the various terminals of the coil?

Warren, when I had Hall Sender problems before, it seemed to me I could see some faint voltage on the green wire when that unit is working. Just with my cheap multimeter. If the coil tests properly, then not getting a spark means it's not getting that 350 - 460v pulse from the CDI box. Or maybe my cheap meter can measure that voltage. Probably.

Bad news. Car started perfectly. Twice yesterday it failed to start.

Jim,

The Bosch 0.221.121.001 and 0.221.121.006 CDI 'coils' should measure 0.4 - 0.6 Ohms on the primary, and 650 - 790 Ohms on the secondary.

Hard to say for sure whether you meter will measure the trigger signal from the magnetic pickup coil correctly or not ... 'good' signal readings could be anywhere from 0.6 - 3.0 Volts AC. Since it isn't a sinewave ... meters display different values from what it really is. If the meter gives you any indication at all during cranking ... the signal is probably getting through the coax just fine!

Try the neon circuit tester on the coil 'hot' lead ... while the car is running, just to see whether it is a usable testing technique!!! It was the only thing I could think of early this morning, when trying to figure out a simple, cheap way to check the 'A' signal from the CDI unit ...

Thanks, Warren. I'm posting here mostly so that other folks can benefit from this data point. The car starts and runs probably better than it has in a while. Smoother. My suspicion is that the connection, at the CDI box, for the two conductors in the green wire, was just not a good enough connection. That signal is very very wimpy, so any resistance in that connection can have a profound impact on performance. My car is probably more fuel efficient now too, and with gas prices where they are...... All from unpluging and replugging a connection. What I really should do is Warren's continual suggestion to us, which involves q-tips, vinegar, alcohol and silicon grease.

Thanks for the data, Warren. I should jot those values down in my manual, which stays in the trunk. I'm not sure what you mean by the neon tester, but I think it might just be one of those cheap things that look like an ice pick with a tail and a handle that lights up. Yes, I was able to at least "see," though I could not measure, the green wire signal, when I was having Hall Sender problems a couple of years ago. I'm a little surprized to hear that the Sender could possibly muster 3 volts. I think it's just a little electric generator where six signals per revolution (of the distributor shaft) have to be generated by a little star wheel. That's got to be a pretty wimpy signal.

At any rate, thanks again Warren. Thank you for keeping our cars running these last five or six years. For those of you who don't know, if Warren had a dollar for each problem he has solved here over the years, he'd be retired.

Jim,

Thanks for the compliments!

The following link is what I had in mind for a neon circuit tester ... usually under $5:

http://www.awsperry.com/cgi-bin/db_search2.cgi?name=ET-201A&sort_by=8&submit_search=More&session_key=&exa ct_match=on&item=et-201a

I probably have 6 - 8 of the same type tester around the house, and in several tool boxes ... first piece of electrical test equipment I got from my dad in 1958!

You're certainly welcome Warren, and deserving. I'll see if I can locate one of those testers. My testers look like an ice pick, and I think they have a regular filament in them.

My ignition troubles are not over, and it's looking like it might actually be the CD box. I'm posting here to ask whether CD box problems are ever intermittent. That's my problem, intermittent. Runs fine, then occasionally dies. Tach falls like a stone and the engine is off. Fiddle with stuff....taking perhaps ten minutes or five, then she starts and runs fine. I thought it was the green wire so I improved those connections, but then it happened again. When it did, I put my (DC) multimeter on it, and got 0.02 volts across the two green-wire conductors during starter operation. Of course, it's not a DC signal, but I think the presence of any signal means the Hall Sender is sending. I'm beginning to think my CD box (A Permatune) is just failing intermittently. I need to travel to my storage locker and fish around for the working Bosch unit I have.

Anybody with any experience with intermittently-failing Permatunes?

Okay, I did a little researching of past threads, and Permatunes do seem to have intermittent issues. More information: When this happens, the engine is always warmed up, but the CD unit is barely warm to the touch, at all. Not at all what I would call "hot." No more than 150 degrees, to the touch. Probably much closer to 100 degrees. Farenheit. Not hot at all. Still, I'm becoming very suspicious of this Permatune unit.

http://forums.pelicanparts.com/porsche-911-technical-forum/215424-my-car-stalled-twice-today-whats-going.html

supes,
check out the above thread that art started..it sounds quite similar, reference: permatune, etc.
ryan

Yea but, but but.....it's not hot. If it's a temperature-related thing, then maybe it cuts out when it gets to room temperature? Doesn't make sense. Ah well, I just need to find my Bosch. Thanks, guys.

I did the non-sense act w/Permatune starting in Area 51. I didn't bother finding out if coil or box. Both replaced.

This is extremely frustrating. Drove the 1.5 hrs each way to my storage locker. Didn't find my spare Bosch unit. I need an entire day and a helper for that. I'm using some fairly colorful language today.

Car died on the way back. The Permatune CD box was barely, if at all, warm to the touch. It was nearly cold.

Very very frutrating.

Jim,

Could the original Bosch CDI and coil be stashed away in the smuggler's box as spares?

Is the Permatune out of warranty, now? If so, I suggest popping the top cover to see if it is potted in epoxy???

Intermittent problems without a clear pattern are the worst possible scenario! Solder joints going 'flakey' are often temperature sensitive, but not always ... sometimes they are vibration sensitive, and sometimes even position sensitive!

I need to perform these tests also due to the following problems:

http://forums.pelicanparts.com/porsche-911-technical-forum/216716-did-i-kill-my-plug-wires.html#post1867991

If I put my Fluke portable scope on, will I detect the pulses? And, should this be sinusoidal, or more of a square wave? I'm seriously considering the MSD unit.
Pat

Pat,

The following document has the waveform from a magnetic pickup coil [as between CDI terminals #7 & #31d] illustrated ... kind of like an ocean wave breaking:

http://www.st.com/stonline/books/pdf/docs/4347.pdf

If your scope probes are rated for 500 Volt inputs ... the output signal at terminal #A is essentially a square pulse of 350 Volt - 460 Volt amplitude.

Warren:
Thanks! I sincerely appreciate this. Should I find a non-repeating (differing max amplitude) waveform from terminal A, could I thus conclude the pickup in the distributor is not functioning correctly. Maybe that's a reach. The reason I ask this is that my plugs were all a bit carbon fouled, with the exception of #1 and 4, which looked like they hardly fired. If 1 and 4 aren't firing correctly, this would have to show up as an abberation in the waveform. The #1 and 4 thing have me a bit baffled for an explanation. My Fluke is rated 600VAC.
Do the magnetic pickups ever wear out?
Thanks again!
Pat

I had the worst scenario in time, miles, heat, ice cold. It was a wait of 10-30min till it would start.

Pat,

The green coax cable from the distributor can cause misses or intermittent no-start conditions. Often they will be found to be discolored brown and with cracking insulation near the distributor. The replacement part [about 9" long] is around $30, and an easy job to replace. Some cars don't have the connector in the run between distributor and CDI connector, but that is a common VW part that can be installed to mate with the replacement 'pigtail.' A search on 'green coax' will turn up many threads!

I doubt that you will see much variation in the output pulses to the coil. Once the input circuit Schmitt Trigger circuit switches, the SCR will be triggered and that sequence of discharging the main 1.4 mFd cap is pretty much identical every time. I documented the 6-pin CDI circuit in the following thread:

http://forums.pelicanparts.com/porsche-911-technical-forum/203100-6-pin-sc-turbo-cdi-unit-repair-documentation.html?highlight=6pin

One problem that can occur is what Bosch calls 'flashover' ... internal arcing inside the coil that happens when the coil secondary doesn't have a proper spark gap load, and the 35 - 50 kV spark output jumps to the case or back to the primary and back to ground! Eventually, the oil inside the coil is overheated and can leak out causing a noticeable goo! Hence, the suggested tests for 0.4 - 0.6 Ohm primary, and 650 - 790 Ohm secondary test limits.

I suspect the fouled plugs are due to spark plug wire problems ... either cracks and pinholes in the Neoprene/Hypalon insulation, or defective resistors at the ends. The ends of the OEM braid-shielded cables unscrew, and the wires can be replaced with bulk ignition wire from a auto or motorcyle supply store. Only branded cable such as Accell, Packard, Belden, Magnecor, etc. should be used. I prefer a silicone-insulated cable, but the Black OEM style works just fine, too. New crimp-on ends will have to be installed. The parts are available at Porsche or Mercedes dealers, and most Mercedes dealers will crimp them for you, free, too. Numerous people have rebuilt their own old wires, and JWW does it, too ... with great results! Save the big buck$$$ for something important!

Thanks Warren!
I'm putting on my troubleshooting hat as I type.
I guess you've done this before, huh? The EE profs used to tell us at school, " you ME guys ought to pay attention...the world will be electronic someday"
Pat