Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[bill99]

I have a 1999 XJ Cherokee 4 litre with a (check engine) light on but also a bit of a twist.

As mentioned the check engine light is now permanently on and the car will start and run at idle just fine but as we drive away and the engine revs pick up suddenly bam everything cuts out, no stutter or lack of power from petrol starvation or something like that just everything off as if I had just turned off the key.

I can then start it up again just fine and run along only using idle revs or just above idle, basically in first gear, I think that’s a bit beyond limp mode.

The fact that it starts first turn of the key suggests the basics are all there, fuel / spark etc.

Any help to point me to the right area appreciated, I don’t have an error code reader and not much luck getting someone to look at it.

Cheers

[General Confusion]

You can read your error codes on your odometer.
Turn the ignition key on, off, on, off and on. The codes will be displayed in sequence. 55 means all codes have been displayed.
Sounds like a sensor, possibly TPS (Throttle Position Sensor)

[bigjeepzj]

The tps won't show up as a fault code.
But it sounds like it could be a duff battery as the jeep been left standing?
Where are you based

[Mike Pavelin]

Does it rev freely when standing still in Neutral? If it does I would check for loose/damaged wiring.

[bill99]

Thanks for all the replies.

The jeep is a daily driver and has been great for the last 5 years with just with basic servicing and one incident with the crank position sensor just needing the connector cleaned. I live in West Lothian near Edinburgh.

Today I got into it and did a general fluids check and removed the air intake to get access to the cable loom and checked all around the engine bay but nothing obvious. The battery is good and spins the engine over fine.

Anyway I cleaned the crank sensor again just because I had good access to it.

Also disconnected the battery to re-set the CPU and took it for a run.

Now it will run better I can drive it almost normal but the engine does cut out sometimes just breaking for a junction. The check engine light is still on.

I cant seem to get the on off /on off / on to work and display the error codes.

I will take it out for a longer run later to see if the CPU can learn its setting and deal with the cut outs.


I have got a guy in the motor trade asking around if someone can do a plug in diagnosis locally.  There was one place that could do it but they did not have enough demand for it to justify renewing the software licence.  I find Jeep main dealers around here are not interested in old cars and make you feel like a leper, not to mention the costs.

[General Confusion]

I cant seem to get the on off /on off / on to work and display the error codes.

Have a look here http://www.jeep-xj.info/HowtoECUcodes.htm  it may explain it better than I did.

[Mike Pavelin]

It has OBDII, any garage will have a scanner capable of reading it.

[bill99]

Thanks Mike and thanks General confusion for that link to the error code list.

Time to buy myself a scanner.

By the way I have tried every combination of key on/off and trip meter switching as per that link and all I ever get is a system check of the instrument panel but no codes.

Shame you cant just connect up a serial port on a laptop and run a bit of software, or can you?

Cheers

[Dave69]

you can but its supplied by jeep and an obvious licence cost to go with it

[Brynjaminjones]

Just for the record, having tried for ages myself in my '98 XJ, I researched some more and found that the key on-off trick doesn't seem to work after '97.

[bigjeepzj]

if you have an adriod phone
have a look at this
https://play.google.com/store/apps/details?id=org.prowl.torque&hl=en

and get one of these
http://www.ebay.co.uk/itm/ELM327-ODB2-ODB-II-Wireless-Bluetooth-Car-Auto-Diagnostic-Scan-Tool-Scanner-BF-/351361606924?hash=item51cec8a50c:g:aLIAAOSwFnFV952x

[bill99]

yeah found some of those on ebay last night and i think i am going to order one that works on a usb adapter cable with any windows pc from XP onwards.

In the mean time i see a basic hand held Maxi Scan unit in a local auto shop for £35 that will do for now just to get started and get the error codes off and move forward, i will try to get there and pick one up tomorrow.

[bill99]

Got a cheap £35 scan tool and it shows generic OBD-2 codes as opposed to what to expect from the jeep onboard codes.


Generic code is P0123, which is listed as  (TPS/Pedal Position sensor A Circuit High Input).

So I had a clean and re-seat of all the connecters around the throttle body but no luck.

Guess I will have to track down a circuit diagram and check out the cables.

Question before I get into that are these position sensors known to go faulty or is a cable fault more likely.

Also just wondering when the fault is finally corrected will the check engine light go out or will it need to be re-set?

[JamesH]

TPS do go. Cheap replacements are not always great. Before stumping out for a genuine Mopar one you can test the sensor with a multimeter. Connect to the signal wire and move the throttle position from 0 - full throttle and see the voltage steadily change. Alternatively borrow a good genuine one or get a genuine one from a good scrapper.

[bigjeepzj]

 
if its the original then it's not water proofed and they get damp inside
the new mopar ones are sealed better

[LairdScooby]

Have a look at this :



It's the Auto Shut Down device kicking in from the sound of it, mine has been doing the same for a week now, since getting it back from the MoT station where they welded the drivers side sill/floorpan area. Obviously they had to lift the carpets so the seat, sill trim etc had to come out including the kick panel trim. I can get the kickpanel out without disturbing the lower dash which holds the fusebox but obviously they couldn't. This means the lower dash was left hanging with the fusebox pulling on the wiring. It's pulled several connectors "out" - not all the way out so it's obvious but partially out so they no longer make proper contact. I've been pushing the fuses back in one by one while pushing the connectors back in with a thin screwdriver.
Obviously most of the info on the ASD shows the underbonnet fusebox (PDC) and i'm referring to the underdash fusebox but there are some feeds from it to the PCM and other things. As that was what was disturbed in mine that's where i'm concentrating my efforts. However, yours could just be a dirty fuse in the PDC or something of that sort.

Worth checking out maybe?

[bill99]

Hi Lairdscooby, yeah mine was like that at first but after taking the battery off the CPU for a while and starting over it seemed to learn to deal with the bad TPS and actually drives not too bad now and only cuts out when slowing for junctions.

But I must say when you are doing a bit of speed and everything cuts off including all the power-assisted brakes and steering it is a bit of a shock.

I’ve got a cheap TPS on order just to eliminate it so I can move another bit forward.

Cheers

[LairdScooby]

Funny you should mention disconnecting the battery effecting a cure Bill - i've also had a similar experience. Although that almost sounds like a contradiction in what i said earlier, it does make sense when you know the sequence of events on mine.
Firstly i should say mine has LPG - a single point mixer system.

Four weeks ago i put it in for MoT and it failed on welding and emissions. The emissions i knew about, the stepper motor for the LPG had siezed in the en position so it was running rich. At some point someone has fitted a power valve to the mixer so i was able to tweak it down to 2% (a pass on LPG) which is still richer than ideal but a compromise to ensure it doesn't go lean at high revs. However, this would often cause the engine to die on a trailing throttle approaching a bend or junction - very embarassing and quite a shock the first time the steering goes heavy!
Anyway, when i got the car back, it wouldn't run on LPG as in the ECU wouldn't change over. Basically the LPG ECU had died while at the MoT station. I thought it was simply a wiring problem because of them pulling the fusebox down but no. Luckily i had another (different make) LPG ECU which i had bought to convert one of my Rovers. So i fitted this, it now runs better but still cuts out occasionally due to the over-rich idle but i haven't got the programming software working yet - another story, long and boring one!
Anyhow, the system was showing a Lambda fault because of the old LPG ECU so i disconnected the battery to reset the PCM. This has cured the ASD coming in. All because of an apparent Lambda fault which is no more because it was down to a FUBAR LPG ECU!

Today i finally found (and thought i had fixed) the last (i hope!) fault caused by the dash hanging on the fusebox - the heater switch! AMazingly the weight of the dash hanging had pulled so much on the loom that it had yanked on the cables that go to the heater control panel, in particular the switch that operates the fan and air con, breaking the mounting legs! This meant the switch "fell off" the temperature control part leaving no fan at all or direction control. I superglued it back together but the glue was past its best so broke again in short order. Fortunately i made sure it didn't go to the "OFF" position at any point so i still have the fan and heat coming through the screen and footwell with a little out of the vent.

There is a point to all this waffle - now it's running right, when the new TPS arrives don't fit it until the fault comes back or you could be chasing your tail as i was for a few days simply because there were several faults with similar symptoms. Hopefully you only have the one fault and have fixed it - if yours had a dodgy lambda and the PCM wasn't reset after renewing it, this could have caused your problems.

[bill99]

Well I got my new TPS and fitted it today and nothing changed so out with the code reader and cleared the error codes but it comes straight back on as 0123 the A circuit is high.

OK so battery off and reset codes again but it comes straight back on, so I took it for a drive to see if it would learn and sort itself out but no luck.

So out with the voltmeter and splice in a test cable to check the TPS in circuit and it seems to be working fine to me. It varies smoothly from 1.3volts to 4,8volts and the earth cable checks out fine and the 5-volt supply is fine.

Interestingly while I had the 3-pin cable off the TPS to measure the voltage off load the CPU detected a 0122 error, which is for a low circuit, which yes it is due to the cable being off.  So the CPU must be getting the signal feed ok because it can detect both high and low errors so that should eliminate any cable short or open circuit from the TPS to the CPU.

So ok I checked the old TPS on a bench rig with 5 volts supply and it works fine but because it is on the bench and the throttle mechanical connection is not there it goes right down to 0.1 volt and up to 4.8 volts. So it looks like that was always ok to start with.

So now I am stuffed I really don’t know where to go next because it detects a TPS error and shows the symptoms of a TPS error but I cant see anything wrong here.

Am I missing something here, does the TPS just get fitted and go or is there some set-up procedure.

I’m grasping at straws here but could it be a mechanical issue with the throttle mechanism holding it slightly open and giving 1.3 volts instead of allowing it to fully turn right down to 0.1volts. As far as I can see the operating blade is in the correct position.

[LairdScooby]

Check the earth point for the PCM. I think it's somewhere near the L/H headlamp and you'll probably have to remove the PCM (and before that, the battery) to do it. I had something similar on mine where it was telling me the Lambda sensor was U/S and it was running high emissions on both LPG and petrol and it turned out to be the earth point near the oil filter on the engine block - removed the nut, cleaned the ring terminals and the block/nut etc, refitted and problem solved! That is the earth for the Lambda ssensor strangely as the sensor is the other side of the engine.
Also check the earth for the TPS just to be on the safe side.

Basically the PCM will see an extra low voltage from most sensors if it has a bad earth and if the sensor has a bad earth, the PCM will see an extra high voltage from the sensor. This could only be something like 0.1 or 0.2 volts but that might be high enough to throw things into confusion.
There probably is a set-up procedure for the TPS but it might pay to make sure the throttle butterfly is shutting properly first. There is an electronic bypass for the idle speed/fast idle and the butterfly only supplies enough air for the basic idle speed which is usually lower than the correct idle speed.
Things that might stop the throttle shutting - over-tight throttle or kickdown (or cruise control) cable, someone has been messing with the throttle stop screw, gunge from blow by from the PCV system build up on the butterfly.

All basic stuff that's often overlooked.

[Jonny Jeep]

Typical values for TJ/XJ TPS are .6-.8V at idle and 3.6-3.9 at WOT. Yours seem a bit off from that and more like what would be seen from the TPS in a ZJ.

[bill99]

Thanks for all the help so far guys.

I have swapped back to my old TPS and checked the output to see if I have anything different from the replacement and it is pretty much identical.

The TPS signal is 1,2-volt / 4-volts.

I am coming round to the idea that this is somehow mechanical, when I fit the TPS and slide the bolts in it has to twist just enough to turn it up to 1,2 volts even though the throttle butterfly is completely closed and you can hear it physically close.  If the CPU is looking for 0.1 volts then 1.2 volts could be considered as high and produce the error C0123 which is TPS A circuit high.

The thing is you can’t fit it any other way so I can’t adjust it. Can anyone who has fitted a TPS in the past confirm that it is normal to have to give it a bit of a biased on twist to get the bolts in?

Anyway I have also cleaned and checked all the earth points around the CPU and checked all the pins on the CPU for corrosion but all is well there. For good measure I have also run an additional earth wire back to the battery.

That’s me for tonight will get back to it tomorrow.

[JamesH]

I went through this and it eventually got really bad and finally undriveable. Turned out to be the MAP sensor and fitting a known good one, genuine one off a scrapper fixed it instantly. Handy if you know a garage with this option near you.

And yes TPS needs a bit of a twist to line up and fit screws.

Where are you?

[bill99]

hi james thanks for that confirmation about the twist of the TPS when fitting,that was bugging me.

I’m in West Lothian near Edinburgh.

I went to the big scrappy in the area (Sports car Breakers) and not a single jeep, I had planned to basically take off all the sensors.

did your MAP sensor not get a error report?

cheers

[LairdScooby]

I went through this and it eventually got really bad and finally undriveable. Turned out to be the MAP sensor and fitting a known good one, genuine one off a scrapper fixed it instantly. Handy if you know a garage with this option near you.

And yes TPS needs a bit of a twist to line up and fit screws.

Where are you?

That's really interesting - although i thought mine was cured of cutting out because of the fusebox troubles and the ASD kicking in, it seems to cut out every so often but fortunately for me, when it does the LPG takes over. I can still hear the clicking of the ASD relay every so often but it keeps going on gas.
That said i do want it completely sorted so i might have to look at the MAP sensor as well.