Mr. mare, welcome to Just Answer!.
A couple of things fit your description, but I think low coil spark output is our best bet at this point. Since the entire secondary ignition system (besides the coil) has been replaced recently, it doesn't seem likely that you've burned out a coil wire or blasted through a distributor rotor.
Pull any spark plug wire off and set up a grounded test light (or jumper wire) so that you can coax spark to leap to its maximum output. You should see a minimum of 1/2" of spark, 3/4" is more like what I'd expect. Anything less than this will mean your coil is a hurtin' unit.
I've been seeing instances with our Magnum coils where the individual stamped-steel core sections that are exposed to the elements become rusted and swell, breaking the coil internally and allowing secondary voltage to jump back to primary 12v. Once rolled over, you might notice cracking in the epoxy base where the primary connector is located...
Write back if this comes up short, knight. We'll continue from there!
No problem bud... we'll get to the bottom of this.
I'm surprised (as you might imagine) to see that the coil didn't solve your problem. Check your spark output at a few locations, starting at the coil.
Then move to the distributor side of the coil wire.
Then to the spark plug boot ends.
Watch for a dropoff of spark along the way, something that is significant. You'll naturally lose SOME spark due to circuit resistance, gap "jumpage" at the rotor... but it should remain fairly hot all along the way. I'm looking for a spark loss that drops it below 1/2"... probably quite a bit below that.
Try to compare the spark you see while it's running OK to when it's... not.
Just wondering. Did you see the rust expansion and cracking of the original coil? I'll be standing by!
Misfire should set a code of some sort, but the Misfire Montior plays each cylinder against the others to determine whether reduced cylinder performance has been found. When they ALL run poorly there's no actual reference and you might escape a misfire code.
Very good observation!
A restricted cat can cause some oddball problems, no doubt about it. I'd expect your problem to get worse with added throttle or engine load, something that gets progressively worse as you pour on more demand. It somewhat resembles what you've described, but I heavily discount exhaust restriction at this point. For one thing, it just doesn't happen in the 2003 units, unlike earlier years. You could have someone rev the engine in neutral while you check exhaust flow at the tailpipe. It will normally kick your palm backward on a quick throttle stab. If you see something other than this, let me know!
Last (probably should've been first), let's check for codes.
Your Dakota is equipped with a factory code-read feature that is tripped by rolling the ignition key from off to on three times... leaving it ON.
Watch the odometer window as any powertrain-relevant codes will come streaming out. You'll see "P-done" at the end of the test or if no codes are stored, but it confirms that you've triggered the code-read.
Check it out and I'll stick around for at least another hour to see what you find.
Dang! Just be advised that the code read you just did is the same thing as most low-end scan tools. You can't pull information other than codes, but you just did the same thing without the added expense. Pretty cool feature!
Ok, totally understand the late-night thing. I'll get back with you later on. Try to avoid disconnecting your battery or running it down if all possible to retain info and Monitor functions if at all possible.
And by the way... that wasn't a crack. I'm confident that we'll get through this based on your observations, something that's rare and very welcome!
A slow loss of power (and intake backfire) like you saw most recently sounds quite a bit like a fuel pressure issue. I have another customer who had a similar complaint as yours, doing a very good job of simulating a fuel-starved engine and it turned out to be a PCM... something that took us a good amount of time to sort out. As I recall, it was a 2003 Dakota 3.9...
One of the things that helped us sort it out was setting a code on purpose in PCM memory, then checking to see if it was still there after a stalling event. When the code disappeared, it meant (in this case at least) that PCM memory had been erased because of intermittent controller failure. With no codes stored right now, I think that might be the next step.
I like to set a throttle position sensor (TPS) code because it can be done easily and instantly by just disconecting the sensor with the key on. You'll see a CHECK ENGINE lamp illumination after that and you should come up with a P0123 code when you do the code-read fire drill.
Check to be sure the code is still in memory if the CHECK ENGINE lamp goes out after a few days (should be retained) and then check again immediately after it acts up on you.
Your original observation that you had weak or no spark at all supports the possibility of the PCM being a player in this... maybe this test will help to sort it out.
While it doesn't happen on every truck (don't know why), the ABS may show a warning lamp illumination after one of these events if the PCM went nite-nite. When ABS can't confirm a VIN match between it and the engine controller, it sets a VIN MISMATCH code and turns the yellow and red warning lamps on. Let me know if you've seen something like this.
PCM failure isn't the only thing that can cause a loss of consciousness tho... lost power or ground feed to the controller or a short-to-ground of the 5v power feed to any of the major engine sensors (MAP, TPS, cam, crank, trans governor pressure sensor, A/C pressure transducer) will knock it out in a heartbeat, leaving no memory of the event. Your power ground connection for the PCM in this year is very near the controller itself, just to the (vehicle) rear and below the air cleaner body on the right wheel well. Give it (and the PCM) connectors a wiggle to see that they're in good shape. The PCM is just to the right of the air cleaner (3 connectors).
Last... information only. The PCM and catalytic converter are both covered under an extended federal emission warranty up to 80,000 miles and 8 years from its in-service date. If you're still under on both ranges, the PCM will be covered 100% if it's found to be bad. Try to avoid breaking the 80K barrier if you're on the edge right now because the Factory won't budge an inch on the miles.
Sorry it took so long to get back to you. Something happened in my mail that didn't prompt me to your last post so I just stumbled across it on the JA site. I'll try to keep an eye out from here on out.
Knight.... you crack me up! I do feel your pain however and have another suggestion.
Since all this started after the plug wires were replaced, track the coil wire all the way from the coil to the distributor. If it doesn't make a turn and follow the INNER valve cover area, it may be coming in contact with the exhaust manifold, accounting for the wet weather problems. It doesn't take long for the coil wire to burn through on direct or intermittent contact and it might be the answer we've been looking for.
The loss of fuel system repair (fuel pump) seems to be an inconvenient coincidence. Even with fuel system flushing (performed at the injector rail) there should be no connection between the two. If the fuel pump was holding pressure as it should, no cleaner would find its way back to the pump. At worst, you had some leakage from the fuel system check valve (or regulator) in the pump module and the cleaner... cleaned it, making it worse. Bad part, regardless.
So check your coil wire or even better yet, load up a spray bottle with water and a tablespoon of dishwashing detergent and spray the entire secondary (high-tension) ignition system starting at the lower areas, leading to the top with the engine idling. If you hit a "leak" area, the spark will immediately follow the easier path the water/ detergent has provided and you'll have an instant answer. The ignition coil wire was originally routed inside the valve cover in this year so I didn't think it would be worth checking. But with parts replacement... who knows?
Hey, thanks for typing all that service info in! It's a pain but it sure helps me to know the vehicle (and you) a little better. Sorry this is taking so long and I'm having email problems it seems... your post STILL didn't come up on my alerts. Keep up the good work and we'll talk later.
Well... good! At least the time wasn't wasted. :o)
That's the trouble with a catastrophic failure of the PCM that fixes itself... not much left behind. The complete absence of previous memory is the first thing I look for, a little easier with the factory scan tool (DRB3). It records a restart count that keeps track all the way to its max of 255, so you can tell how long it's been around, counting the starts sometimes right back to the event.
Convincing the dealer might be an issue, like you figured. There's a certain amount of skepticism about anything a customer carries in with him, particularly something off the Internet. But once you've seen whether the code you set disappears or not, you'll have a bit more ammo to back up your story. Give them all the info, make sure it makes it to the repair order (writers are bad editors sometimes) and see if the tech can sort this out for you. I've searched for a marker that you can use to identify an intermittent failure of the PCM (besides lost memory) and can't come up with one. Sometimes the 5v feed to the major sensors drops out, sometimes not. Sometimes the 5v feed becomes a weak grounded circuit (through PCM), sometimes not. It's typically a hot engine/ been driven a while-type of problem with the JTEC though, a half-hour or better of good, hot underhood heat wash before it acts up.
Insulating the PCM from underhood heat might tell you something if it extends or "cures" stalling. Use an ice pack right against the controller, followed by layer upon layer of towels or bubble wrap, tucked in tight enough to prevent most air circulation past the unit. The JTEC seems to repond favorably to this kind of treatment, where the type used in Caravan (SBEC3) doesn't seem to care much.
Good luck and keep me updated!
No problem. Whenever you get the time is fine by me. 5.19v is just about dead-nuts right for the 5v feed on the JTEC, so you're getting good readings. If you can manage to tap into a 5v feed and drive with it with the meter inside the truck it will probably come in handy. The slow dying nature of your truck might show up in a slowly degrading 5v signal.. but maybe not.
You might do a resistance to ground check of the 5v circuit while it's healthy for comparative purposes. The 2K scale works best for this and check it with the key off.
Have a great weekend,
That stall when turning the wheel is something that can happen when your truck hasn't been retrofitted with a power steering pressure switch, a TSB update. With no advance warning to the PCM about power steering system load, it can cause idle variances and even kill it if it catches it off guard. No biggie.
You're right about the regulator, it was replaced with the pump unit. This type of fuel system is called returness since there is no return line from the fuel rail back to the tank. Return-type systems have one disadvantage over your type... they shuttle engine and exhaust heat back to the tank, something that can cause vapor lock eventually in severe circumstances. Evaporative fuel control gets a bit tougher with hot fuel trying to evaporate as well.
Your buddy might be right about the crank sensor, but there just isn't enough evidence to go one way or the other right now. A backfire like you mentioned earlier could be consistent with a random crank signal that happened at the wrong time. Crank sensors also use the 5v feed and can short the circuit to ground, taking the PCM down with it. It's not common at all on the 3.9 engine (unlike the 4.7) to see this happen, so it wouldn't be my first choice.
The resistance test I mentioned is one that would check the 5v circuit at any underhood location to ground. You're going to have some natural grounding of the circuit as the 5v passes through sensors on the way back to the PCM, so the first reading should be on a day when the truck is running fine to get a baseline reading. You'll need to use the 2K scale to get a good reading (it will be more than 200 ohms, less than 20K). I don't recall what the reading will be but it seems like somewhere in the 500 ohm is about right. Check it with the key off and the key on. I'm not sure if you'll get a reading with it on, but try it anyway.
Then when it takes a dump again, check the resistance in the same way if the 5v feed has dropped (should be steady 5.19v). If the 5v is still good, no testing of the circuit is needed. We're looking for the reason the 5v dropped, so if the circuit resistance is lower, go around and start disconnecting sensors until the 5v comes back up or you run out of sensors. The problem may be something within the PCM or out on the truck... hard telling until it lays down again.
Thanks! I surely appreciate it!
The bulletin I mentioned was TSB 18-015-04. The repair involves replacing the power steering hi-pressure line with a revised unit that carries a pressure switch, installing an overlay harness that connects it to the PCM and then a software update to the PCM that will alert it to the new sensor. These are (obviously) things that you wouldn't be able to do at home and figure something around $350 for the update. If occasional steering inputs that kill the engine don't matter THAT much to you... I'd sure understand!
Have a great trip and I'll expect that you'll bring me something! :o]
I dunno.... wet cats usually make me happy. As long as they don't stay that is...