Let's get to the bottom of this...

I have, and read about others who have, gauges that read low. It's an 89 XJ. The voltmeter reads about 11v, the temp gauge reads about 160 (IR themometer shows 190 is actual), the oil gauge reads 40psi cold and 10-20 operating, and the gas gauge MAY be accurate (I've run it low and not run out).

I've been told Chrysler inst clusters are famous for having voltage regulators go bad. You need a good regulator (not the alternator one, in the cluster) as reference voltage for gauges. Has anyone pulled it apart and tried anything?

I also have been running with a Radio Shack digital temp/volt meter that plugs in the lighter. It reads (as does my DVM) 12.5-13 at the lighter socket, but 14v at the battery. Clearly, there is voltage loss in the wiring to the inside too!

I've never had a vehicle with such varying voltages! I've cleaned all the ground connections/straps, but no diff. Has anyone dug into this? I'm considering adding some radio equipment, so these drops are bad, as well as potential RF problems later on.

Since it seems a common 88-90 type problem, and they weren't like that new, maybe we can put our heads together and fix it. I'll try the suggestions and report what worked and what didn't. Or, has anyone fixed it already?? ;-)

Skip

i know the volt gauge was changed, and if i remember correctly it just seemed like the numbers were changed so it was reading a higher charge, more accurate to the actual charging voltage. so i would guess if the temp gauge was bad, maybe they would do the same.

I have the same problem -- voltage reads about 11.5 on the gauge and 13.5 at the battery. Temp reads lower than it "should" but I have no way to verify what it actually is.

It has been suggested that the instrument panel ground may be corroded. I know there's a multi-connector that snaps into the printed circuit, so I can see how some corrosion might affect the readings. I just haven't taken the time to open it up, remove the connector, and treat it to some Ox-Gard.

As to the voltage regulator: I vaguely remembered AMC cars having a voltage regulator for the gauges, so I went looking for it on the XJ. There isn't one in the cluster, and there isn't one shown in the wiring diagrams, and there is no mention of one (that I have found) in the FSM, so I'm operating under the assumption that there isn't one.

the old amc's and chrysler cars did use a voltage regulator for the gas and temp gauges, but not in the xj's. i'm not sure about the xj temp gauge but the volt gauge was replaced for the innaccurate reading problem. ive done it a couple of times years ago, but nowadays the customer just deals with it instead of spending the money as long as they're reassured its ok.

Ahh, good stuff. I did some checking tonight. I brought some heavy cables around to the inside while connected to the battery. The instrument gauge reads 11.5 volts, the lighter 12.8, and the battery cables 13.8. Measuring between cable positive and lighter ground, 13.8v, so ground isn't the issue with that. Between cable ground and lighter +, 12.8v. Whatever is on the same circuit as it is drawing down 1v.

I then connected the cables into lighter to simulate a heavy cable run directly from the battery to reduce the loss. While the voltage stayed up at the lighter. the gauges didn't change. I was hoping they would instantly become 'right' and the problem solved ;-)

I've used some deoxIT D5 on other underhood connectors and was impressed with how it restored the metal on contacts. I also looked at stabilant 22 for enhancing connections, but at $62 for 15ml, that's too much... It sounds like taking the cluster apart is next.

Are there any other ground or other points that typically corrode or are a problem while I'm taking it apart... ?

Also, can anyone send me schematics for the wiring in an 89xj, even down to the detail inside the cluster? Maybe a pdf?

Thanks,
Skip

Both Haynes and Chiltons have wiring diagrams if you don't want to spring for the FSM.

There is no schematic inside the cluster. It's a printed circuit mounted on the back (out) side of the cluster housing. Each gauge mounts by two screws, which make contact with the printed circuit. The chassis wiring harness is a multi connector that just snaps into a recess in the cluster housing. The printed circuit is flexible enough to bend into the opening so the contacts on the connector can scrape against them as the connector is inserted.

You can trace the printed circuit easily enough, but if the gauges work at all I don't think that's the problem. Clean up the contacts and Ox-Gard them, and maybe run a supplemental ground wire from the printed circuit to a known good chassis ground and let us know what happens.

skipc said:

Ahh, good stuff. I did some checking tonight. I brought some heavy cables around to the inside while connected to the battery. The instrument gauge reads 11.5 volts, the lighter 12.8, and the battery cables 13.8. Measuring between cable positive and lighter ground, 13.8v, so ground isn't the issue with that. Between cable ground and lighter +, 12.8v. Whatever is on the same circuit as it is drawing down 1v.

I then connected the cables into lighter to simulate a heavy cable run directly from the battery to reduce the loss. While the voltage stayed up at the lighter. the gauges didn't change. I was hoping they would instantly become 'right' and the problem solved ;-)

I've used some deoxIT D5 on other underhood connectors and was impressed with how it restored the metal on contacts. I also looked at stabilant 22 for enhancing connections, but at $62 for 15ml, that's too much... It sounds like taking the cluster apart is next.

Are there any other ground or other points that typically corrode or are a problem while I'm taking it apart... ?

Also, can anyone send me schematics for the wiring in an 89xj, even down to the detail inside the cluster? Maybe a pdf?

Thanks,
Skip

Click to expand...

Check the blower motor. On my 87, that drew down the voltmeter reading especially on high. I just got used to it and stopped worrying, but occasionally considered hunting down the positive lead for the voltmeter and running it through some other circuit.

My 95 has just the opposite problem, reading constant overcharge.