2000 XJ with Check Gauges light on

[MoFo]

2000 XJ 4.0, AW4, NV242

I just took a road trip to visit a friend 900 miles away. After leavine his place, I stopped for gas. When I restarted the engine, I had a Check Gauges light and the voltmeter read zero volts. The engine was running fine, so I figured it was eroneous and I drove on. I stopped at a rest area and measured the battery voltage at 13.05V. When I restarted the car, the CG light was out and the voltage was back to normal. I drove on to 300 miles and stopped for gas. When I restarted, the oil pressure was at 0 and the voltmeter was normal. At 400 miles, I stopped and everything under the hood was normal. When I restarted the car all was well again. I stopped at 600 miles for gas. On restart, the voltmeter had dropped to zero and oil pressure was back up. I stopped again at 800 miles for dinner. Everything was fine from there home.

• At no time did I loose the readings on my water temp., gas level, tach or speedo.
• Everytime I checked the battery voltage it was between 13.0 and 13.15V.
• The battery is 3 months old. All connections were cleaned and greased when the battery was installed.
• I have since cleaned and lubed all engine grounds.

I have owned this XJ since 2002 and have never seen the CG light before.

Any help you folks can offer is greatly appreciated.

 

[Matthew Currie]

If the oil pressure gauge had not been involved, I'd suspect the alternator, but if more than one gauge was involved, I'd suspect the connector for the instrument cluster first, and try cleaning and reseating that.

If you're having an alternator issue, it will also drop the voltmeter to zero, but not immediately. There should be a few minutes during which you'll see lower voltage before it goes to zero and sets the check gauges light. If it was at zero when you started, it suggests a gauge problem, since obviously you could not start with zero voltage.

 

[MoFo]

I would have suspected a connector to the Instrument Cluster as well, because of the 2 gauges (only one at a time go to zero). However, the FSM would indicate all the gauge data comes through a twisted pair serial connection from the PCM. If that connection failed, I should loose the water temp, gas, tach and speedo. I don't know what connector I can clean and reseat. The +12V, ground and serial data cable are common to all these gauges.

 

[Matthew Currie]

Although it's never happened on my 99 I've heard many reports of connector problems with 97-up, and as I understand it there's a big connector on the back of the cluster that can get glitchy. It seems worth trying, since it should be pretty simple to unplug and replug it a few times and see if it helps.

 

[98_XJ_Classic]

its the connector on the cluster, mine did this a few times, took it apart cleaned it worked fine until the whole think took a dump and had to replace it

 

[UNCC_99XJ]

Another vote for the connector. My 99 is starting to have problems with it as well, and I got this cool little message while driving down the road a couple of weeks ago:

Even though my speedometer indicates 0, I was actually doing about 30-40mph driving down the road. The cluster can do all kinds of weird things when the connector gets dirty enough.

 

[Jeepguy03]

Another vote for the connector. My 99 is starting to have problems with it as well, and I got this cool little message while driving down the road a couple of weeks ago:

Even though my speedometer indicates 0, I was actually doing about 30-40mph driving down the road. The cluster can do all kinds of weird things when the connector gets dirty enough.

:roll: :laugh: In all seriousness that message means the cluster isn't registering with the computer. My friend had the same issue on his '97 and we cleaned up the connections in the back. Haven't had the problem again.

 

[jbmoose]

It's the bus connection. Every gauge dies on mine, and as soon as i step on the brake, they all come back to life. But the Sentry Key light still stays on.

 

[UNCC_99XJ]

:roll: :laugh: In all seriousness that message means the cluster isn't registering with the computer. My friend had the same issue on his '97 and we cleaned up the connections in the back. Haven't had the problem again.

Guess my only other option is to take the train? lol!

Yeah, some day I'll get around to cleaning it. It's not a DD, so i'm not in a huge rush to fix it!

 

[MoFo]

I am not sure what connector you guys are talking about. The FSM shows two Self-docking Wire Harness Connectors where the Instrument Cluster plugs into the Instrument Panel. Are these the connectors I need to clean and reseat?

How much time does this take?

 

[MAY POP]

Takes about 1-2 hours to pull the dash apart and put back together. After the first time it gets faster. First pull the plastic and steel covers under steering wheel. Next the cover over radio. It is held on by clips only. Next the screws and cover over speedo. The knob on the headlights has to come off. Look on the side of the switch there is a small knob or pinhead that is spring loaded. push it in and pull out the headlight knob. After the speedo cover is off a few screws and the speedo assy comes out.
Ron

 

[UNCC_99XJ]

Just took mine apart today, I was able to do it w/o removing the headlight knob. It will fit through the hole on the trim piece you're trying to remove. It's a tight fit, but it will fit through.

Whole thing took me about 30 minutes start to finish.

 

[MoFo]

Verified Fix: 2000 XJ with Check Gauges light on

I cleaned the connectors on the back of the Instrument Cluster. This did not change the symptom. The Check Gauges idiot light (and low voltmeter) problem continued.

I started thinking about what a new or rebuilt alternator would do to solve this, if the alternator was charging the battery just fine. I called the local rebuilder. He told me often a diode junction starts leaking, letting some AC into the PCM and they don't like that. I did not buy this, thinking the battery would filter out any AC at the alternator output. Just to make sure, I got out the old oscilloscope to check. The output at the alternator battery terminal could not have been cleaner; No AC there.

While I had the scope out, I connected it across the 2 small wires that lead from the PCM to the brushes. I saw over 30 volts of electrical noise, in frequencies between 1kHz and 7kHz. This certainly meant the brushes, and maybe slip rings were shot. This is my DD and I needed this thing fixed, so I could not wait for an online parts order.

I took it to the rebuilder for their free bench test service. I explained my symptom with the CG idiot light and my suspicion of electrical noise, leaving out the detail of the oscilloscope. He spun it up and said "it is good, charging well and no AC on the output". I told him I expected higher frequency "trash", not AC. He pulled it apart and looked at the brushes and declared them completly worn out. His offer was either exchange it for a 2 year warranty rebuild for $120 or replace the brushes and clean the slip rings for $48 with a 90 day warranty, which could be applied to the cost of a rebuilt unit any time during that 90 days. Five minutes later he handed me back my alternator with new brushes and polished slip rings. The alternator has 161K miles, but the bearings are still good.

Since this problem started, I could get a maximum of 2 start-run cycles before the symptom came back. Since I replaced the brushes, I have counted a total of 26 cycles without the CG idiot light, so I am calling it fixed.

Summary
My problem was caused by worn brushes and slip rings, which generated electrical noise that fed back into the PCM through the rotor control (regulator) circuit. A new set of brushes, which probably cost $10 to $15, and cleaning of the slip rings fixed the problem. This eliminated the need to replace the entire alternator with a $150 to $200 rebuilt AZ or NAPA one.

If you replace your brushes and cure a Check Gauges problem, please post your results to this thread, including part numbers and your parts source.

Check Gauges | Idiot light | volt meter | voltmeter | alternator

 

[Matthew Currie]

That's interesting. I've never seen worn brushes affect the PCM that much, which is not to say I don't believe it, just that it's a useful thing to know.

e.t.a. as I've mentioned, on this system, if the voltmeter drops for long enough, it will drop to zero and set the check gauges light, and this is indeed what my 99 did when the brushes went bad, but it never affected any of the other gauges.

 

[MoFo]

In my case I was getting a solid charge, with 13.1 to 13.2 volts at the battery. I suspect the poor brush contact was too short of a duration to be seen at the alternator output.

Where did you buy your replacement brushes and what did the cost you?

 

[Matthew Currie]

In my case I was getting a solid charge, with 13.1 to 13.2 volts at the battery. I suspect the poor brush contact was too short of a duration to be seen at the alternator output.

Where did you buy your replacement brushes and what did the cost you?

I went to a local auto electric guy, and asked if he had the brushes, and he gave them to me! Just so you know, unless you get the whole holder assembly they must be soldered in, which is not difficult but makes it not just a plug and play job either. In my case, the slip rings were deeply scored, so as soon as I could, I replaced the alternator altogether.

You can check on line for Nippondenso brushes. I don't think they're very expensive, and might even be a pretty good deal with holder included.

If your slip rings are just scorched or a little worn, you can clean them up and go, but be warned not to sand them or use abrasives or they'll eat the brushes up. Crocus cloth is ok.

The local auto electric guy is, obviously, both a nice fellow and a savvy businessman. When my starter went bad a couple of months later, you can guess where I took it.

 

[MoFo]

I noticed several vendors on ebay have these brushes for sale. Look for a brush assembly for a Denso 90 to 120 A alternator with External Regulator and Internal Fan (ER/IF). They can be had for 12 bucks, including shipping.