• 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.

Hard starting when cold

alfman

NAXJA Forum User
NAXJA Member
Location
Pike Co, PA
My buddies 1992 XJ with 4.0 fuel injected is having a starting problem when cold. When hot it starts right up. When cold it just cranks but won't fire up. Things that have been replaced cap, rotor, spark plugs, wires, ballast resistor, fuel filter & air filter. Fuel pump was checked & is OK. When cold it is acting like a carbureted engine with no choke. Anybody have any ideas?
 
My 94 had the same issue. Ended up being the temp sensor. The engine would not adjust the air/fuel ratio properly to the temp. Other possibilities would be Idle Air Control (IAC) solenoid. Throttle Position Sensor or like the other person posted "dirty fuel injectors" Good Luck.
 
Any check engine light? If so, pull codes. I would suspect the coolant temp sensor here. They can be tested for resistance change when hot and cold. Best way is to remove and test the sensor but you can do some basic testing on the engine. I don't think they are real expensive to replace.
 
Which temp sensor the one on the back of the head or the intake manifold. My 89 is doing about the same thing. I have to wedge a stick inbetween the gas pedal and the seat to keep it running untill it gets warmed up.


Parakeet
 
Hallo. A '92 has a MAT sensor (Manifold Air temp.). This sensor can act in the same way as the CTS. AKA choke. :laugh:
But the condition of the Battery and the cables are the most important.

'92 XJ
 
Last edited:
We checked everything again & it is not geting spark when cold. If it is warm, like in a heated garage, it starts fine. If outside & it's cold out it has no spark. It was doing this before changing all the ignition parts & is still doing it. Any ideas? Maybe starter relay?
 
If it's cranking, it's not the starter relay. You're problem is weird, because normally temperature related electrical problems occur when things get hot. You've got the opposite issue. You have to get it cold, where it won't start, and start checking for current to the coil and backtrack from there.
 
Crankshaft position sensor I think does not allow spark when although it will still turn over. I may be wrong, someone want to add to this? I'm thinking it may be that or the sensor beneath the rotor of the distributor. Sorry my brain is fried right now been working with numbers and binary coding for 14 hours straight today.
 
Crankshaft position sensor is possible, but it would be unusual for it to work when warm and not when cold.

Well, let's throw out this crazy possibility........What if????? The crankshaft position sensor was in a position where it only was close enough when it was hot, but when it was cold, the metal contracting of the flywheel results in no signal (or somthing like that. Some years are adjustable for that gap.
 
Hallo, I renewed my Ignition Coil last week in my '92 and that will make also a big difference in starting. The CPS is also a problem with a bad connector or a bad harness.
Did you check the MAT sensor? By a cold start is the fuel/air mixture very sensitive. The ECU is monitoring this sensor also in the start-up mode and the warm-up mode. :spin1:

'92 XJ
 
Last edited:
Further investigation, no spark in run or crank. crank position sensor was changed (cracked in half), no power at coil input, coil was also changed with a known good one. I do not think it is ignition switch as I have power at the ignition fuses in the underhood fuse block, infact I have power on both sides of all the fuses in that block. I also changed the ignition relay with a known good one. Where is the ECM on the 92 XJ? I thought it was the black box between the air intake box & drivers side inner fender, I have someone else told me it is the silver box behind the dash panel under the glove box. This other person said he jiggled the wiring under the dash on passenger side at one point & it started. There is no power at the ballast resistor on driver side inner fender near firewall. Is this ballast resistor ignition or fuel pump? Any other ideas? Thanks for all the help so far.
 
Last edited:
Ballast resistor powers fuel pump for sure and coil also (I think). Ballast resistors were originally designed for the coil, but fuel pumps added later. They can be jumped with no bad effect. It's not the crank positon sensor if you don't have power to the coil. If you don't have power to a component, it means that the problem is upstream. So, if you don't have power to the to the ballast resistor, don't replace it. If you have power coming in to it, but not leaving it, then you either replace it or jump it. So the question is: What is upstream from the ballast resistor? (Sorry, don't know for your year.) Look in wiring diagram for your year.
 
The only wire diagrams I have are in a Chilton & a Haynes manual. There is no power on either side of the ballast. I need to know where the ECM is, I don't see the location in either manual I have as I need to find out if I have power there. In my diagrams it appears that the ECM controls the ignition & fuel pump relays.
 
The ASD (auto shut down) relay won't close (supply power) unless the PCM gets a pulse (the motor is turning over and the CPS is good). The fuel pump relay will close for a second (just enough time to prime the rail) and will then open and won't close again unless the PCM detects a pulse (the motor is turning over and the CPS is good).
 
We exhausted everything we could think of, so it's off to the dealers repair service. Will let you know what we find.

Thanks to all that have help us so far
 
Well the verdict is in, the problem was the ECM itself. There were no codes present but it appears that something internal like a solder joint let go. When it was warm it expanded & made contact, when it was cold it contracted & broke contact.
Thank you for all the help we received.
 
Thank You for posting the repair. Diagnosing a bad solder joint can be a P.I.T.A. Glad to hear you are up and running again. :patriot:
 
Well the verdict is in, the problem was the ECM itself. There were no codes present but it appears that something internal like a solder joint let go. When it was warm it expanded & made contact, when it was cold it contracted & broke contact.
Thank you for all the help we received.

Well alfman . Great news. But can you explain please. Is the ECU easy to open?
Are the solder joint easy to reach? Was the repair easy to do? Can you make a little -How To-? Mayby a pic or diagram.?
Because this is a very important item and very unknown.


'92 XJ
 
Back
Top