Strange engine behavior

[Akula69]

In reference to the christmas jeep...

 

I6 4.0L Renix 1989.

 

Initally, it starts and runs fine. Will sit and idle for hours (really!) without apparent difficulty except the idle becomes slightly rougher after the first 20 minutes.

 

When I took it on the road yesterday, I had travelled about 8 miles at average speed when the engine just quit. No sounds, no warning lights, just quit. I coasted to the side of the road and waited for about 10 minutes and then attempted re-start, which it immediately did start. I drove for approximately 2 more miles and parked, shutting it down. After about 25 minutes I came back to the vehicle, it immediately re-started, and I drove away. I travelled about two miles and it quit, again without warning.

 

No matter what I tried, it would not start. It would crank fine, and I could clearly hear the fuel pump, but no start. It (and I by default) was towed back home. Later that evening (approximately 3 hours later) I tried to re-start it and it momentarily started but died immediately, and would not re-start.

 

Today, (15 hours later) I went out and the damn thing cranks immediately. I shut it down and put a guage on the fuel rail, getting 40 PSI at crank and 32 PSI while idling. I drove it up and down the street in an approximation of the mileage I travelled yesterday, and it has not faltered.

 

New plugs, new ECM, new coil, new wires, checked ballast, have fuel filter but have not replaced it.

 

The previous owner indicated it would stall like this and not re-start for approximately 10 minutes, then start normally.

 

Does anyone have a suggestion on this one? it really has me stumped.

[brigarpeon]

Replace that fuel filter. Sometimes they will accumulate crud water etc... that sloshes around when you hit bumps. This will temporarily interupt fuel flow and drop the pressure. As you crank it you are forcing the fuel under pressure from the tank to make a new path when it does it will start. Water can also get into the fuel filter, when picked up it will pack the rail, until the injectors get fresh fuel it will not start.

 

2nd thing I do is check the harness coil plugs on the firewall and make sure they are fully snapped into place, corrosion free and packed with dielectric grease.

[BLHTAZ]

CPS... Crank Position Sensor ;)

[Akula69]

CPS... Crank Position Sensor ;)

 

Umm. Sorry I forgot to add that as a recently replaced item (by the PO). Is there a way to test it for functionality?

 

The wiring harness mod (to reroute the CPS wires) was also done at the same time, and the plugs are tight.

[aemsee]

Double check the wiring repair to make sure the wires are soldered together.

[Eagle]

That's what mine is doing now, except that I have to sit about 20 or 30 minutes before it'll restart.

 

I believe the fuel pump ballast resistor is bad and faltering after it has become "heat soaked." I now have a short jumper wire with flat spade terminals on each end, to plug in as a jumper the next time it quits on me.

 

Naturally, last weekend in the course of my usual weekend warrior tasks, it didn't quit.

[Akula69]

Well, after a little research (and a phone call to Blataz - thanks sir!) I have dicovered two things:

 

You can re-set the CPS temporarily by unpluging it and then repluging it. Does it work? I have no idea but will let you know :D

 

I need a wire like Eagle's, and I need to ask him how to test a CPS with a multimeter.

 

Otherwise, I drove the beast around the neighborhood (didn't want to get too far away in case I had to push it home...)for over 30 miles today and it would not misbehave. I let it idle for 90 minutes, and although it was somewhat warm it (a) did not stall, and (B) did not overheat.

 

I am still stumped.... :nuts:

[Eagle]

Unplugging doesn't "reset" the CPS. It scrapes a bit of possible corrosion off the contacts in the plug, hopefully improving the signal strength enough to tell the ECU what the CPS is saying. That's all. It's a temporary fix at best, and may not work at all.

 

Somebody awhile ago posted a link to a tech article put together by Eric Giordano (a.k.a. on-line as Jabbathehut) that explains how to test the sensors -- including the CPS. The method described in the FSM is useless -- it calls for testing the resistance at normal operating temperature, but how do you get to normal operating temperature if the engine won't start? And I've tested the resistance of known bad units against brand new ones, and they ALL tested within spec, and within about 5 ohms of each other.

[maddzz1]

Link to test sensors:

 

http://www.lunghd.com/Tech_Articles/Eng ... ostics.htm

[joedrives2fast]

It sounds to me like maybe an ignition module. If you did have a bad coil at one point it can cause problems in the module. Heat can make these arise. I wouldn't suggest running out and buying one but if you could find a buddy with one who would let you run it for a while it may help.

 

It would be nice if you could find some Noid lights to make sure your injectors are getting a signal to fire. They are like little light bulbs you plug into the injector harness so you can see the signal.

 

As far as fuel filter it's an easy swap and well worth the time and piece of mind.

 

One last sparing possibility is the sock on the bottom of the fuel pump. Luckly you don't have to drop the tank to take the pump out but make sure your under a quarter tank.

 

The sock on the bottom of the pump will only get real nasty and cause problems if the truck sat for a while and built up some rust. Or if you got sand in your tank.

[highlandjeeper]

I hate gremlins :mad:

[Mountianrider]

Ok, this is perplexing. I'd check the connection to the CPS to see if it appears in good shape. I'd be tempted to make up a kind of piggy back wire set up, splicing in to the old CPS connector and running the new wire to the next connector and splicing back into the line at that joint. I am begining to think those little wires corrode inside and cause these failures?

Any one else want to comment on this idea?

 

Just last week mine refused to start when I went to move it after sitting in the hot sun all day. In the evening it fired right up. I replaced the CPS which had been it since new (1990) and had 300K miles on it. No problems yet, and I swear the milage has improved quite a bit.

 

I replaced the fuel filter when I had the engine rebuilt last year, I also replaced the fuel pump. Yes, the filter screen was virtually varnished shut!

Don't really see how she was getting enough fuel before I replaced the pump. So, it wouldn't hurt to replace that as well.

 

Sounds like it might be a good idea to make up a jumper for the fuel pump resistor as well and just tape it to the fender by the resistor as insurance measure.

[aerocorey]

http://autorepair.about.com/library/faqs/bl484c.htm

 

This link isn't nearly as good as the one I had before. I can't remember where I got it, but I used to have the actual service bulletin. The idea is that there's too much resistance in the stock wiring between the CPS and ECU. The CPS becomes more resistant as it warms up and the signal becomes too weak. The solution is to drill a hole in the firewall above the gas pedal and run a short patch harness from the CPS through the firewall straight into the back of the ECU. The patch kit is set up so you pass the harness through the firewall, de-pin the 2 sockets on the ECU plug for the CPS wires, then re-pin the ECU plug with the wires in the new harness. To get my truck back on the road I cut the plug off the CPS wire harness, drilled a hole in the firewall, stuck the wires through, and used butt terminals to splice the CPS into the ECU harness 1 inch away from the plug. The wires (IIRC) are white and white with a black tracer. Works like a champ. Before I patched my CPS my truck was leaving me stranded ALL OVER THE PLACE. I replaced the plugs, wires, coil, module, CPS, cam position sensor, fuel filter, etc. After spending hundreds of dollars on parts, it ended up being fixed with 2 butt connectors my father in law had rolling around in his toolbox.