I don’t think the problem is electrical. If you measure the voltage going to the battery, I bet it will be 13.4 volts. The bad thing about do-it-yourself and parts store code pullers is that it is either only as accurate as the person operating it or only as good as the detector itself. Who knows how many times that code puller has been dropped or bumped around. Spend a little money and do two things: One, spend some money on a compression tester. If you can change the lugs, you can check the compression of the cylinder that is giving the code. This is a good thing to do because you don’t want a hungry mechanic telling you that you have poor compression when you probably have marginal compression - the kind of compression that a 100K engine would have. Then if you don’t have poor compression, take the car to a shop and have them pull the codes professionaly. An honest shop won’t charge you for a full hour of labor because it doesn’t take more then 10 minutes to pull codes. Be sure to tell them that that is all you want them to do because you sound like you are intelligent enough to replace your own bad sensor/component.
My Father (Now 83) owns a Chrysler van conversion that has a similar problem but I know the engine is a bit worn. I have been replaceing the passenger side, rear spark plug every other month for the past five years. There is a lot of mileage on this van because he lives outside of LA and has Grandkids as far away as Missouri. So, if it happens that there is poor compression in that specific cylinder and your G-Dad is like my Father and wont spend the money or allow anyone else to spend money to fix it - there are alternatives. Seems like the bigger a pain in the ass you are to your parents the more vengeful they become in their later years.
I have scanned through this post and have not seen anywhere that the coil for #4 was pulled, a spare spark plug actually put in the plug socket, the base grounded, and the engine run to see it there actually is no spark being produced at #4. If there is indeed no spark, your tests so far point at the coil driver module (PCM if they are in the unit). If you are getting a spark, change the spark plug, check the compression on #4, and/or look at a fuel injector for #4.
I apologize if I have missed coverage the above trouble shooting.
I just posted the same thing along with a bunch of other information and it’s gone. This BBS sucks! I’ve had it with lost and missing posts. And this wasn’t the familiar double post to the same thread bug.
You’re welcome for the help and glad you learned some things on this.
There is nothing else that you need to ckeck on this. The PCM is bad and the one from Ebay sounds reasonable. You may be able find one at a salvage yard that is less expensive but it may require some reprogramming and that will cost some extra money.
To Researcher and Twin Turbo:
The OP mentions that he did try swapping coils and installed new plugs and neither produced a change to the problem. Since he did see a small difference in voltage readings between a known good PCM coil contact and the bad one I think he has pinned down the trouble.
You want to backprobe electrical connectors to see what “live” voltages are doing in the circuit. To use a backprobe pin, screw it onto your digital multimeter test lead, slide the pin along the wire into the connector until it contacts the metal terminal. This way, you don’t have to disconnect the electrical plug. You could file/sand down, to 0.040 inch, your test meter probes, or get them from Fluke, etc. Here they are: http://cgi.ebay.com/Fluke-TP88-Ridgid-BackProbe-Pin-Set-w%2F-Tip-Protectors--_W0QQitemZ320258100627QQcmdZViewItem?IMSfp=TL0805311478r13842#ebayphotohosting You can also use simple straight pins.
Thanks everyone for your input, this has been extremely helpful and my grandfather thanks you too.
I think I’ve done pretty much all the testing that makes sense for this problem. I’d consider compression on that cylinder, Norm, if I wasn’t getting a different (bad) reading at the connector, before it even reaches the cylinder.
I also hooked up a test light ground wire to the positive terminal of the battery and the main contact of the test light to the coil driver wire of the coil connector for cylinder 4 and started the car…no flashing light. Did the same with cylinder #2 (a properly firing cylinder) and got a proper flashing light. This was a Haynes manual test and the book states that if you don’t get the flashing light, the problem lies either with the crankshaft position sensor, camshaft position sensor, or the PCM.
I’m guessing that if either of those two sensors were bad, I’d either not be able to start the car, or it would be running much worse…and giving mulitple, random misfires…not just cylinder 4 all the time. Is that about right?
Norm, I’ve got a decent code puller that has never been mishandled and I’ve also got a setup that allows me to hook up my laptop to the car and run different software to monitor the ODBII system…so I think I’m ok there. Thanks for the suggestions though.
Researcher, I think the above test I did is about as good as the spark plug to ground test…but I see what you were getting at.
Well, unless anyone has any differing opinions on the camshaft/crankshaft sensors above, I think I’ll be going down the new-PCM lane…thanks again for everyone’s help, especially Cougar and hellokit!
I can’t believe that Haynes suggested testing an electronic circuit with a test light! This is dangerous to do to electronic circuits. The use of test lights, and cheap ohmmeters, can allow too much CURRENT (called: “loading”) of such circuits. The results can be burned out circuits, and burned out electronic components. This is why I instructed you to to use backprobes and a quality digital multimeter.
I agree that it was highly probable but myself, I like to be absolutely positive before I replace something like that. Glad to hear it has been proven to be the problem but prior to the noid test, I still thought it could remotely be something else like lack of fuel, dead cylinder etc. Hooking up the spark plug externally and seeing the lack of a spark would have been definitive (in the absence of test eq).
My next concern would be why? Why did that driver fail? I would measure the primary coil resistance very carefully and compare it to the other coils along with inspecting the wiring and connections. Component failures do occur but I would want to reduce the chances that the replacement PCM could be damaged in the same manner over the long haul.
I also agree that it’s best not to excessively load down the PCM outputs if you don’t have to. But considering that the drivers are intended to service a pretty low impedance load already, a lamp isn’t going to hurt them even when paralleled with the coil. Those are designed pretty tough and can handle a lot of power. I read somewhere that Chrysler was using a capacitor to snub some of the transient noise from the coil switching. If there is one and it’s external to the PCM, you might want to verify that it is good. If it were open or bad in some way, then the driver might get punctured over time. A few random thoughts anyway…
You’re welcome for the help Jad2007. You did a good job on this. I think the new PCM will fix this problem up. The only ‘gotcha’ I can think that could happen is if there is something wrong on the crank sensor that tells the PCM to fire #4. I’m not sure how the sensor is designed.
The Haynes manual specified to use an LED test light, which I did…they did warn that a regular test light could damage the circuits.
I did also use the multimeter but was only getting the differing voltages on the coil driver wire (-0.06 for the “good” cylinders, +0.03 on the “bad” one).
TwinTurbo, I see your point, but the coil has already been replaced so if it was overloading the PCM, that’s no longer an issue. (Grandfather took it to a mechanic when this first started happening, and without testing much, the mechanic just replaced the “bad” coil and that fixed the problem for a while, if I heard the story from my grandpa correctly… it sounded like they had “fixed” the problem but weren’t sure how. Great.)
Also, the Haynes manual wiring diagrams do show an external capacitor, but I’m not sure where it is located…and there are two, each drives 3 cylinders, so I think there would be problems with more than just one cylinder if that was the case.
Just a bit apprehensive about those crankshaft/camshaft sensors before we do the PCM…
Thanks Cougar. The only thing I haven’t eliminated are those crankshaft/camshaft sensors…I really think there would be random misfires and a really bad running or non-running engine if either of those were bad…but if someone can confirm that, I’d appreciate it. I just hate to suggest a $200 part to my grandpa if that won’t fix the problem.
One thing you can do to see if something is different inside the PCM is to compare the resistance of the bad pin connection to a good one. I like to use the diode function of a DVM when doing this. Measure between the pin and a ground lead for the PCM. I suspect you will see a difference between the two readings. You can also reverse the probes to change the polarity and see what you get. If there is a difference then you know for sure the PCM is bad.
This sounds like a good test…just to make sure I’ve got it right: switch my multimeter to the Ohm’s setting that has a symbol like this " -->|- " (is that the symbol for diode’s?), attach the ground probe from the multimeter to a proper ground, and the positive lead to the different pins for the coil driver’s on the PCM connector (with the engine off, battery disconnected)? And then reverse the probes to see what happens, comparing all the different readings with both setups.
I’ll try that and post back!
and there are two, each drives 3 cylinders, so I think there would be problems with more than just one cylinder if that was the case.
OK, if that’s the case, I agree that it is unlikely.
Just a bit apprehensive about those crankshaft/camshaft sensors before we do the PCM…
From what I understand, they are hall-effect sensors and there is a corresponding plate with notches that encode the position. If it were marginal, I would expect the problem to move around between the coils. The potential for a fixed obstruction (ferrous material) staying put and consistently messing up one cylinder is remote IMHO. Hope I don’t have to eat those words
Thanks TwinTurbo! And you are correct, the Haynes manual describes the sensors as “Hall effect” sensors…I think your explanation makes sense and it’s what I had felt, too. PCM it is, after I do a couple more checks.
Haha…I hope I don’t have to eat those words either! 
Yes, you got it right though you don’t really need to disconnect the battery. Just make sure power is removed to the circuit. Before testing, removing the connector going to the coil may be a good idea also as it may change the reading you get. Since all the pins to the coils should be close to the same values, using either polarity you will be able to see how things measure up.
FINAL UPDATE
It’s fixed! I took the PCM out today and found a used one that matched the P/N at a junkyard nearby, bought it for $75 with a 4 month warranty. Put the “new” one in, hooked everything back up, started it up and it runs great now! No missing, misfiring, shaking, or other problems. The check engine light has stayed off, there are no codes, and a quick test drive went well in all gears and RPM ranges.
I’ve heard that a PCM going bad is very rare, so I’m curious how this one went bad, or more specifically, how just the one part of the ignition module that fires cylinder #4 went bad but everything else worked fine. Probably will never find out, but I’m just glad it’s all working now and without throwing a lot of parts at it first!
Thanks to everyone who provided some input, suggestions and help on this issue, especially Cougar, hellokit, and TwinTurbo. I appreciate the help and so does my grandpa! Felt nice to actually go through a proper testing/diagnosis procedure and to replace the right part the first time.
Till next time…
Jad
Praise the Lord for success!
“Why it failed?” comes under the heading of: “Stuff happens”. If something happened enough times, research would be done to find out why; but, for fewer occurrences, you’ll only get theories----it’s not worth the effort, and time, to find out.