Crank Sensor bad or not?

[LewisE]

Sorry if this seems long winded. I have a 1991 4.0, AW4, 2WD all stock with 65k miles on it. It’s in really good shape and has no rust…not even on the floors.
Truck starts and runs fine for about 15 minutes and then starts dying and coming back like you would expect a failing crank position sensor to cause. Dies for like a half second at a time, tach drops then it comes back to life getting progressively worse as the engine warms. The other gauges and lights are unaffected. Eventually it’s dying and bucking so much I have to stop. If I let it cool for 30 minutes or so I can get another 10 or 15 minutes runtime. Always starts and runs fine when cold. I think my runtime is getting less as time goes by and eventually something will fail altogether. My fuel pressure is about 34 when running.
So, anyway, I’m thinking it’s the crank sensor so I set out to get a MOPAR one….ha, not for a 1991. They can still be found for 97+ so, thinking they are most likely the same electrically, I get a 97+ MOPAR one and a new plug for my harness. I put the new one on and fire it up….within 30 seconds I have the same symptoms but with a cold engine now….WTF. Well, as the new MOPAR sensor was not U.S. made, I figured I either had a dud or the 97+ ones are not the same electrically after all. I went out and got myself a new “Standard” one for my specific year and made an adapter to plug into my new connector on my harness. Now I can use either type. Put my new one with the round connector for my model year on and fired it up. Same symptoms again with a cold engine. Since I was doing this work in a friend’s garage I thought, crap, I gotta get home somehow, so I threw the original, good for 15 minutes, one back on using my adapter. What do you know, it’s running good for 15 minutes at a time again. WTF. Couple of days later I get yet another two new sensors one old style and one 97+ one and they both seem bad too even when cold. The old, original one still runs good for 15 minutes. This basket full of new crank sensors I have can’t all be bad.
What in the heck am I missing here? There is obviously something else bad but why does it run good with my old CPS for 15 minutes but not good at all with any of the new ones? Could there be some other piece of the puzzle that for some odd reason is more happy with my old CPS? Anyone seen anything like this?

 

[old_man]

Being a test engineer by profession, I would look elsewhere, like putting a fuel pressure gauge on, and a noid light. Other culprits are a clogged/broken CAT or a temp sensitive coil.

I would double check, using an oscilloscope, but not everyone has one in their tool box.

 

[LewisE]

I have a fuel pressure gauge installed on the fuel rail. It shows 34 when running. Is that ok? The pump and filter are new. The cat and O2 sensor are new and the coil is about a year old. It's an MSD coil which I would expect to be high quality but you never know. Would having a strobe timing light on one of the plugs when the problem is there be a good indication whether it's losing spark?
I didn't know what a noid light was so I Googled it. Might have to get one of those.
Still baffled as to why it likes the original CPS better than the new ones.
I noticed this as well when I hook my hand held scanner up.......no codes but the RPMs are all over the place. It goes from 0 up to 3000 or so...not the engine, just the display on the scanner. I don't know what to make of that.

 

[old_man]

That is low.

 

[Runnin'OnEmpty]

........the coil is about a year old. It's an MSD coil which I would expect to be high quality but you never know..........
I noticed this as well when I hook my hand held scanner up.......no codes but the RPMs are all over the place. It goes from 0 up to 3000 or so...not the engine, just the display on the scanner. I don't know what to make of that.

RPM values are ultimately sourced from the coil.

I'd suspect the MSD coil being bad. There are resistance tests
you can do to test a coil, but the best test would be to replace
it with a known good one.

 

[8Mud]

The first thing I'd try is to jump the ballast resistor. My best guess.

You can check the CPS and Cam position sensors. orange wire, for low voltage. Check the CPS wires and see if they didn't flop onto the exhaust manifold. I always zip tie them to the speedo cable. Just a loop it doesn't have to be tight, a four-inch loop will keep it off the exhaust manifold but still allow some movement.

The orange wire voltage to the CPS changes from year to year, yours is likely 6 volts. if it is like three volts you have an issue. The easiest way to find the bad sensor is to unplug the CPS, then the Cam position sensor, and then the speed sensor, look for the voltage to come up when you unplug the bad sensor. Sensors can fail three ways, internal short, wrong resistance, and open circuit.

 

[bruceb]

Cheap advice from a long time Jeep owner. I think you have two separate problems. The first is electric. Check your coil as stated in another post. Then check your distributor for shaft wobble, bad wire to cam sensor, inside contact points to distributor cap and rotor. I had similar, though not as bad, symptoms and it was a cracked coil warming up and arcing to the metal around it. Your second issue is fuel. If your fuel pressure doesn't come up to spec you may have a bad fuel pump, clogged fuel filter, bad fuel regulator. I do agree that you can't have that many bad CPS in a row.... hope this helps

 

[markaboo929]

Basic checks here. First find out if it's fuel or spark your loosing.That will help to determine where to start looking. Possible ASD relay shutting down. Easy to swap with say the A/C clutch relay.The asd relay should power the coil and injectors fyi.

 

[markaboo929]

A side note on the CPS and the CMP or cam position sensor is that they are magnetuc and produce an AC voltage .
If you take a multimeter and go from the center itself even in your hand between the signal wire and the ground and if you take a screwdriver and pass it by the magnet quickly you will see an ac voltage at those.

 

[LewisE]

The coil is one part that I don't have a spare for so I'll keep checking the other suggestions to maybe eliminate them. I like the thought of checking voltage on the orange wire and looking for changes as I disconnect the other sensors powered from the same source. The combined knowledge here is pretty awesome. I rechecked my fuel pressure and this time disconnected the vacuum like I was supposed to. It's 41 without vacuum and 34 with. The pressure regulator is newish and a MOPAR one at that. Pump and filter are about 8 months old. Tank was clean inside. I used a hand vacuum pump to look for leaks as well....can't find any.
The distributor is almost a year old. I replaced the original because it had some wobble. The replacement is not OE so I guess it could be suspect. It's a later model one and I have a new cam sensor that I will try in there. I agree that it would help a lot to know if I am losing spark or fuel. Let me ask you guys this: Would hanging my strobe timing light on one of the plug wires provide a good enough indication of it getting spark? It has new MOPAR wires, an Accel cap and rotor and new Champion copper plugs.
I'm stuck working on it on the street and I'm so far north that it gets dark at about 4:30. Plus I'm always worried that some old lady will send the Polizei out to give me a ticket because I'm not supposed to work on my car on the street. I hope to get to the voltage checks tomorrow or Saturday. It'll probably snow and rain the way my luck is going.

 

[8Mud]

I always pull a plug or two and hold them against a ground and crank the engine over. A sharp blue-white spark that you can hear go crack
is a good spark. If you open an old plug up to around 60 thousandths
and get a good spark your system is likely healthy.
You can use the timing light on each individual cylinder, point it at something dark and you can see a miss if there is one.
Jump the ballast resistor, it is a coil inside of a ceramic shell with resistance and the resistance changes with amp flow. If it is screwing up it can make the fuel pump current unreliable. What I'm trying to say is it may not be good or bad, but can get erratic.

 

[markaboo929]

The best way I found to check spark is pull the coil wire off the cap stick a long screwdriver under it and have someone crank the engine while you hold the screwdriver about quarter of an inch or so away from any metal object to see if it has spark. This way you're not messing with the tiny spark plug that hard to see it holding it against ground .A long screwdriver I find is the best tool for that.Just my personal 2 cents on that.Any method will work.

 

[LewisE]

So, I took the truck to a Jeep garage here in the area. they have some guys that have been there since the 80's so I felt like they might have seen some stuff. They went through the same thinking as me and some of you guys but they had the tools to test the ideas. They suspected the crank sensor like me but hung a scope on it and could see steady output even when the problem was occurring. Next they suspected the coil....put on a new Bosch unit but it didn't fix it. Next it seems they isolated it to a loss of fuel....swapped out the filter which they said was pretty dirty. It was only 8 months old. They dropped the tank and looked for contamination and checked out my new fuel pump and sending unit. They finally started looking at the pressure which they said was low by their gauge. They replaced my nearly new fuel pressure regulator and it seems to have fixed it. They tell me the symptoms were really odd and they were surprised at it being the pressure regulator. Hmmm.....me too so it will be a while before I really trust it. Just thought I'd post to maybe wrap this up. Thanks for everyone's input.

 

[jeep ride]

Thanks for the follow up.