• Welcome to the new NAXJA Forum! If your password does not work, please use "Forgot your password?" link on the log-in page. Please feel free to reach out to [email protected] if we can provide any assistance.

Jeep cuts out, then dies, then runs fine again after cool down.

T

thechief86

NAXJA Forum User
Location
White House, TN
Title states symptoms. 91 Xj, 4.0, ax15, np231,172k miles.
Only acts up after an hour or more on the interstate. Around town and wheeling it has been fine. (Never running more than 30 minutes at a time)
Schrader valve on fuel rail squirts gas several feet when depressed, replaced the coil assuming it had to be spark. Still does the same thing. When it shuts down it will start back and die immediately. After 10 minute cool down I can drive another hour or so before it screws up again.
What do you guys think?
 
Already try replacing the CPS?
 
That was fast, Ken!!
No, I haven't. I thought the CPS only gave initial signal to fire on cranking, then went dormant? In the past, CPS problems have caused a no-start that rendered no spark ever. This condition still has spark, although week.
 
haha, saw it pop up in "new posts".

Nope, the CPS is absolutely required at all times for the engine to keep running. The cam position sensor (the one under the distributor rotor) is only required at startup on some years.

CPS sensors sometimes like to fail in high temps right before they fail completely - yours is probably at that point where it works until it gets hot, then fails, then when it cools down, works again.
 
My 96 did something similar, the CPS was failing slowly. Starting got progressively harder (though not extreme), longer crank times, it would periodically crap out randomly after it warmed up. In fact it happened twice, once was a failing CPS and the other time the CPS got covered in oil form a leaky valve cover. I sure don't know why hot oil would mess with the CPS. but as soon as I cleaned it, it worked fine, been working since (years).

First I'd jump the ballast resistor, just because I tend to troubleshoot from simple and easy, to exotic and complicated..

I got stuck in my Renix with a hot CPS that refused to work well enough to give me a start. Just on a hunch, I dumped a bottle of water on the CPS to cool it down and it fired right up. I realize the 91 is different (different CPS) but there may be some parallels.

It needs a CPS pulse continuously or the PCM opens the ASD relay, but best guess, it can miss a few beats and the PCM will compensate some.
 
So you guys think the CPS is a good place to look next? The fuel pump resistor is a lot cheaper, so I'll try that one soon as well. Does anybody know a good part # for one? It doesn't offer one in Oreilly's computer, but I know we have a few on the shelf, different resistances, but none say what they are...
 
Here is the Crown number for the, Resistor Ballast 8933000682. You can just jump it for a test, unlikely to hurt anything in the short term. My 87 never had one and it never caused any issues I noticed.

If you have no spark the next time it refuses to start, the CPS becomes more likely. It will shut off fuel and spark.

Intermittent problems are harder to nail down. If you always have good spark when it is cranking but failing to start, fuel becomes a lot more likely. Ballast resistors can screw up and slow down the fuel pump, you may get good pressure initially but the volume falls off. When I suspect fuel, I always jump the ballast resistor, do a volt test at the fuel pump connector (you can test voltage when it primes), ohm the fuel pump ground. Then do a pressure and volume test. For me it has always been something other than the pump itself, either motor management, or electrical.
 
Last edited:
I've been driving my 91 with no ballast resistor (shorted it out in like november 09, when the previous owner still had it) for three years on the factory fuel pump. Over 250k miles now.

I also agree that it's the first step in debugging here, I don't think it'll fix it but it can't hurt anything, costs nothing, and takes 20 seconds so why not?
 
A digital multimeter will tell you if the CPS is bad.

Heat it up until it quits and then test the CPS with the meter set on ohms. EDIT--set meter to 1k to 10k, and the reading should be OPEN--if there is resistance it is bad.
 
Last edited:
IIRC that only works on RENIX CPSes... the later ones are a 3 pin hall effect not a 2 pin variable reluctance, so there's no winding to check for resistance. Might work if a wire breaks internally on either the VCC or ground pin, but won't test anything on the signal output pin.

A better test is possible but involves either an oscilloscope or a logic analyzer probe from, say, radio shack. Not something most people have or need in 99% of jeep diagnostics.
 
Well, my friend has an oscilloscope, but I have no idea how to use it. It sure looks cool, though! I'll try the ballast resistor first, and if I have another issue I'll swap the CPS. The thing that makes it a pain to nail down is that I'm an OTR truck driver, and rarely have the opportunity to drive my jeep long enough for it to shut down. It has done this twice in 2 months.
 
Here's the settings I would use for checking the CPS:
if it's a dual trace, set it for channel 1 only.
Ground coupling (probably on a switch marked AC GND DC) and adjust brightness/focus/vertical offset until you get a flat dim line across the screen around the bottom, say 1 division up from the bottom, then set to DC coupling
auto trigger, rising edge
2v per division, x1 scale
set probe to x1, or even just use a piece of bell wire. This isn't critical, you just want a general idea of what the signal looks like.
timebase 10mS/div (this should get a bit more than one rotation of the crankshaft onto the screen)

You should see around 12 pulses, maybe a few more, across the screen, approx 2.5-3.5 divisions tall. When it stops working you will see few to none.

If you post a pic of the control panel of the oscilloscope I can give more detailed settings.
 
kastein said:
IIRC that only works on RENIX CPSes... the later ones are a 3 pin hall effect not a 2 pin variable reluctance, so there's no winding to check for resistance. Might work if a wire breaks internally on either the VCC or ground pin, but won't test anything on the signal output pin.

A better test is possible but involves either an oscilloscope or a logic analyzer probe from, say, radio shack. Not something most people have or need in 99% of jeep diagnostics.
Click to expand...

Agree to disagree:

FSM 95--all OBDI

"Disconnect sensor pigtail harness connector from main wiring harness.

Place an ohmmeter across terminals B and C. Ohmmeter should be set to 1k to 10k scale for this test. The meter reading should be OPEN (no indicated resistance). Replace sensor if a low resistance is indicated."

This test WORKS, and I have used this procedure numerous times at the shop and on trails.
 
Oh, my mistake. That's rather strange...

Apparently my recollection was wrong :eyes: another thing learned.
 
kastein said:
Oh, my mistake. That's rather strange...

Apparently my recollection was wrong :eyes: another thing learned.
Click to expand...

LOL!

Half the time I don't remember if I took my meds or not. Not a big problem, except for the "blue" pills. :doh:
 
Well dang. Now I can't get my heep to screw up on me. Does that mean it's fixed, or that I haven't driven it long enough? I guess I'll figger it out next time I try to go wheelin 3 hours away...
 
You must log in or register to reply here.

Similar threads

B
2001 XJ dies at highway speed
2
Replies
20
Views
2K
N8N_99xj
N
8
Engine dies after reaching operating temp.
Replies
10
Views
1K
Saudade
Saudade
S
Renix 4.0L cranks and dies
Replies
0
Views
526
specauto
S
tjmotter
How to test/repair Renix era Cruise Control
Replies
15
Views
2K
RCP Phx
R
L
Brake pressure issue when easing up on brakes slightly, then pressing down further
Replies
10
Views
2K
lordoeuf
L
Back
Top