Intermittent Loss Of Spark. Possible Pcm Failure?

Scott_96GT

New Member
May 3, 2017
10
2
1
*First post on Stangnet*
Greetings! I come in peace from the bow-tie side of the aisle. Recently went looking for a quick and easy project that would go faster than my 30 year old Suburban, with at least slightly better gas mileage, and a craigslist ad for a cheap 1996 Mustang GT with standard transmission and a few unknown problems caused me to make the decision to add a blue oval to my GM family. After all, I'm no brand snob, and what kind of car guy can hate anything that's American, reasonably reliable, rear wheel drive, V8 powered, fast, and cool looking. (even if mine is apparently the "most hated" Mustang body style.)

But enough about me.

After buying the cheapest GT on craigslist, I proceeded to waste all the money I saved buying a cheap car on the basic stuff it needs to have a hope of running. She has new plugs, wires, intake manifold, fuel filter, fuel pump, battery, and little misc things. While all that stuff was bad, the heart of the problem, the reason the car won't run, still eludes me.

The car tries to start when I hit the key, sputters and dies almost immediately. Every once in a while it will run if I rev it up as soon as it catches. When it decides to die however, no pumping of the gas can stop it. ( thus not IAC valve) It seems to be a loss of spark, as a spark tester made from an old plug and wire only had visible spark during cranking while the car was sputtering, and nothing during the fruitless 75% of the time the engine was cranking. TPS shows consistent ohms across its rotation, crankshaft sensor was good, replaced it anyway because desperation. cam sensor is tested and good, alternator was probably good, but also replaced because apparently the diodes can fail and screw up the cam sensor signal, and for some reason she seems to run better with no belt (Although when it decides to run even briefly seems pretty random) Haven't formally tested the MAF, but that shouldn't affect spark much. Once it randomly decided to work, and when I got out to observe the engine running the worn out door fell shut, and the smarter-than-you anti theft whatevers decided to automatically lock the doors, and I had a hard time killing the engine. pulling the MAF plug didn't do a thing to slow her, and it only died after I pulled both coil's plugs out. I guess I at least have compression.

From everything I've read I can't seem to find anything that directly messes with spark besides crank sensor, coils, wires, and plugs, other than the PCM, or maybe the wiring. I've done all the googling I can do, so I figured it was time I went and found some Mustang guys. What do y'all think?
 
  • Sponsors (?)


Welcome.

If looking for recommendations about how to trouble shoot and repair your car, resist the urge to drop into "parts changing" mode as the primary trouble shooting method. Slow down. Test more. It's generally cheaper to test a part than to replace it.

We all have to make educated guess on what's wrong and change parts from time to time. Best to keep to a minimum.

Let's see if we can help. Regarding your question about what affects spark during cranking. The list was missing PATS and power.

I also don't see any reference to what the fuel pressure actually is.

1996+ Crank with no start check list
http://www.allfordmustangs.com/forums/4-6l-tech/336452-1997-mustang-wont-ignite.html#2984838

Ok onto more detailed questions. Hopefully these will point you in the right direction. Or at least rule some more things out.

Absolute first thing. Double check all of the grounds around the radiator core support. Are they clean and tight? Check to see that the motor grounding strap between the left hand motor mount and frame rail is in place, tight, and clean. Double check the battery terminals. Are they clean and tight? Pay special attention to the battery negative as it's bad about splitting.

Bottom line. Today's cars simply will not run right without a strong battery and charging system.

Put an accurate Volt-Ohm meter (VOM) on the battery and monitor the voltage while testing. What we are looking to see is IF the voltage drops or varies and exactly what is going on at the time.

Also a good idea to monitor power at the coil pack R/LG wire (or the radio interference capacitors) with a VOM to see if the power drops out when this problem occurs.

Consider investing/renting a set of "noid" style test lights. These can be used to monitor the primary side of the coil pack signal return line. The noid testers can help confirm a strong grounding signal from the PCM and if the signal drops out during the start sequence.

Are you positive the spark plug wires are routed correctly and connected to the correct coil pack terminals?

Does the tach work? What is the tach doing just before/during/after this problem occurs?

Does the check engine light go out during cranking? The PCM will turn off the check engine light during cranking when a good CKP signal is received.

When the motor does run, how does it run?

Some consideration should be given to performing a compression test to confirm the base motor's health.

In general PCM replacement should be considered only after all other possible causes have been ruled out. Replacing a PCM as a trouble shooting method often creates as many problems as it solves. But it would be a good idea to inspect the PCM. Looking for evidence of water or rodent activity.

A set of Ford service manuals and wiring diagrams could be a big help restoring this project. If interested in getting a set for yourself I maybe able to help. PM if interested.
 
Last edited:
The fuel pump was building up to about 40 psi, but the pressure fell sharply 5 psi the second it shut of, and slowly bled down to 0 over a minute or so. Felt like it was going too fast to just be injectors leaking, so I replaced pump and filter, and the fuel pressure is steady now. So at least that money went to good cause. The intake manifold had been replaced before, and was overtightened, causing it to crack and coolant to leak into the 5-8 spark plug wells, shorting the wires, the plugs were the crappy Bosch 4 prongs, so all that got replaced. alternator I got via a warranty swap, so I didn't feel too bad about the time wasted there, and the crank sensor was cheap, and has actually been the only part I wasted money on so far.

Tach works great. Seems to match engine rpm whether cranking or running.

CEL does go out during cranking. Annoyingly, it refuses to provide any helpful codes.

Battery terminals were pretty crappy. Replaced one of them. In all likelihood, if it isn't my PCM, my problem is going to be in the wiring nightmare around the battery. All kinds of absurd splices, a bunch of inline fuses, things being extended that didn't need to be, and the remnants of someones sloppy sound system install/removal. I would love some detailed diagrams and photos of what that's supposed to look like.

PCM doesn't look fried or rat eaten, but it is mounted just a little haphazardly.

Firing order was checked and rechecked.

No angry antitheft light, forgot to mention that in op, but it's pretty much ruled out. Car also sometimes will runs for several seconds before it shuts down. Not the one second that PATS gives you.

When it runs it seems to be okish, maybe an occasional small backfire, a little blue smoke on the rev, might just be the penetrating oil I sprayed on the spark plugs burning out of the exhaust. No knocking though.

As far as manuals go, I found a Haynes manual online with some basic wiring diagrams. I think today is going to see me poking at the PCM wires and checking grounds. Hardly even know where to start though.

Not my first rodeo in car repair, or even electronic fuel injection, but wiring diagnostics have always been a weak spot of mine. My last project was beat to crap like this too, but at least the wiring was all solid.
 
Major discovery. Found out that if I disconnect the driver side coil, the car will fire up every time, and run good considering it's down 4 cylinders. Was even able to put it in gear and park it a little more straight. Coil tested good, but I guess I've narrowed it down to the wiring on the driver side coil. (Or PCM signal)
 
  • Like
Reactions: 1 user
Noid light test shows the pink signal wire is short to ground with key on and ignition off.
PK/W is the driver coil "C" circuit 528. Sounds like you have a major break through. If this were my car the next test I would do is pull the PCM and see if the ground fault goes away.

OBTW, Ford PATS does not act the same as GM. When tripped Ford PATS does not allow the motor to start even for a second. Regardless if the theft light "proves out" at initial key on, the PATS is out of the picture.

Good call on the fuel pump check valve.
 
Last edited:
Well, I pulled the PCM, and the short went away. As I feared, my PCM must have an internal short. My local auto parts stores don't have my model PCM available within a week, (and those that can get it want around $200) and the local junkyard doesn't have any Mustang GTs period.

Has anyone every repaired their own PCM? Based on the fact that, other than one signal wire being shorted to ground, the PCM works fine, I'm hoping that a single voltage regulator or something may just be shorted. A few precursory Google searches turned up little useable info, although I did locate the faulty pin. (PIN 78) Pulled the unit apart and see no signs of water damage or burned components.
 
Try car-part.com which is a searchable list of salvage yards in your area. What you are looking for is one that will mail you a new unit. Be sure to have all the codes from the PCM and calibration codes.

For the 96-98 model year PATS can be reprogrammed by the end user. The procedure takes 45 minutes and the steps must be followed exactly. The procedure is documented in the Ford service manual.

An option is to use the extended license Forscan product, a compatible dongle, and a laptop. The Forscan product is well worth the effort to set up even with the free version. It's a real game changer.

Amazon product ASIN B01F0GVBWYView: https://www.amazon.com/OHP-ELMconfig-compatible-interface-diagnostics/dp/B01F0GVBWY/ref=pd_sbs_263_3?_encoding=UTF8&pd_rd_i=B01F0GVBWY&pd_rd_r=BB7HDHN8NHH411NPMJWN&pd_rd_w=Rl3DH&pd_rd_wg=PcWX7&psc=1&refRID=BB7HDHN8NHH411NPMJWN

Forscan: http://www.forscan.org/

Forscan extended license needed for PATS programming: http://forscan.org/forum/viewtopic.php?f=5&t=836

There is the board level repair. The driver components are in plain sight on the board. YMMV. I don't have any additional help beyond that.

Or try United Radio.
 
Last edited:
Try car-part.com which is a searchable list of salvage yards in your area. What you are looking for is one that will mail you a new unit. Be sure to have all the codes from the PCM and calibration codes.

For the 96-98 model year PATS can be reprogrammed by the end user. The procedure takes 45 minutes and the steps must be followed exactly. The procedure is documented in the Ford service manual.

An option is to use the extended license Forscan product, a compatible dongle, and a laptop. The Forscan product is well worth the effort to set up even with the free version. It's a real game changer.

Amazon product ASIN B01F0GVBWYView: https://www.amazon.com/OHP-ELMconfig-compatible-interface-diagnostics/dp/B01F0GVBWY/ref=pd_sbs_263_3?_encoding=UTF8&pd_rd_i=B01F0GVBWY&pd_rd_r=BB7HDHN8NHH411NPMJWN&pd_rd_w=Rl3DH&pd_rd_wg=PcWX7&psc=1&refRID=BB7HDHN8NHH411NPMJWN

Forscan: http://www.forscan.org/

Forscan extended license needed for PATS programming: http://forscan.org/forum/viewtopic.php?f=5&t=836

There is the board level repair. The driver components are in plain sight on the board. YMMV. I don't have any additional help beyond that.

Or try United Radio.
Well, I did some probing with a multimeter and discovered one coil transistor had different ohm readings than the others, so I desoldered it and replaced it with the most similar one I had available. (Kinda hating RadioShack for imploding) it worked for a little while, then got a P0353 code and the coil fuse blew. But it did run, which was more than the dead transistor did. I went ahead and ordered an exact transistor since I know that has the potential to fix it. I like spending $5 more than spending $200.

Turns out PCMs can be repaired after all.
 
  • Like
Reactions: 1 user