tachometer reading high

First off 96 XJ, AW4, np231.

A few weeks ago I jumped on the gas a bit and quickly revved over 4000rpms to avoid a wreck.
Ever since that my tach is reading 500 higher than it should. The idle and rpms are normal it's the gauge that isn't correct now.
Occasionally it'll read normal for an entire trip only to be 500 high when I start the Jeep up the next time.
This has happened before a few years ago under the same conditions, but it eventually returned to normal as I hadn't revved it again till now.

I've tried smacking the dash and even removing it and cleaning the whole back of the cluster and ground.
It's still reading high. Yesterday it read normal after dropping off the throttle from highway speed to a stop light. 500 high after the store.
Today it read normal from start up to the gas station in the morning.
500 high on startup from there to the mountain.
Normal from startup from there to the lake.
500 high from startup from lake to the City.
500 high from startup from city to city to home.

What gives? Any ideas?

what are you considering "normal"? What are you using to verify that the gauge its self is what is messing up?

Several years of having it operate normally right along with a scan gauge that I kept in for a while..
I've lost the gauge but I know my car. I could even tell you the exact note it makes at 2000 rpms. It's the same key one of my favorite songs is in. got another song that matches idle perfectly.

I've also had it scanned at the mech when it's actin' up and it is indeed 375-500 high.

Torque converter not locked up?

It is not a TC issue. The problem lies in the gauge itself or wiring associated.

Everything is running at the same values as always (verified by a friewnd's scanner last night) only the actual gauge itself is reading higher than what is actually occurring.

Normal: 825+/- rpms idle (reads 11-1250 when gauge is weird), 1600 flat ground bottom of TC lockup 43 mph (reads 2100 when gauge is weird).
shifts around 2200 from 1st medium throttle....

I could go on, but the point I am making is it has NOTHING to do with the car acting up mechanically. The Jeep is doing and behaving as it always has untillI floored it a while back. last time after a few months it went bak to reading normally.
It's the GAUGE. It's the gauge that is reading higher than it should be.
It's reading normally this morning btw.
I know the difference between what my rig sounds like at 1600 vs 2100 rpms.
It's at 16 but the gauge says 21.

I'm trying to figure out Why it seems to have almost physically "skipped up a notch" after jumping higher than it does normally,why it "falls down to normal" later, and how to fix it.

Well you seem adimate that its the gauge and annoyed we are trying to get other info so here it is in simple black and white. Replace the gauge..... That is what you wanted to hear right, conformation that it can not be anything else but what you are saying it has to be. You asked and we tried. You are stuck in your decision that it "is the gauge" that is acting up. I'm not trying to be rude here and if I come across that way I am sorry but you seem to not want others ideas for consideration unless it falls into the "has to be the gauge." mentality. Knowing how your XJ "sounds" is great but considering others thoughts and opinions in indespensable in fixing your issue. The tach on most of newer rigs are computer controlled meaning they get there readings from the ecu and not directly from the distributed like they use to in the old days. It could be as simple as a bad ground or a loose wire. Could be the IAC, MAP sensor, even a leaking injector can cause a surge that the computer will see and send that info about the rpm's to the tack. It could be the printed curcit board o. The back of the gauge cluster or to much resistance in a ground. Your feeling of the issue as far as what is normal doesn't help us as a group as we can not feel what is happening through the computer screen.....not yet anyways....lol. That is why we ask what we do. Alot of us have been doing this a long long time, myself included and have alot of ideas but to many variables to type them all out.

But like I said if this sounds rude it's not ment to be. I am just stating my point.

I could be wrong, but it seems to me that 96 was the 1st year of OBD II.
If that is accurate, then the tach in that year is just a electronic meter. The PCM sends commands to the cluster and all you see is a readout.

Maybe the PCM hickuped? Try removing the positive terminal, waiting 5 minutes, and putting it back on. That should reset any weird the PCM is up to.

Didn't mean to sound obstinate. I'm not "stuck in my decision"....it is a fact. The mechanic has agreed that it is indeed only the gauge or wiring associated is acting up not the rig. I made that clear.
If I was sure it was only the gauge I would've been asking how to replace it not inquiring about wiring and such. I'm not being adimant, I'm stating what has already been diagnosed by mechanics and 3 different scanners, so please don't react to me as if I'm being stubborn. I have also responded to each idea others have had. Sorry if I haven't agreed with them. doesn't make me difficult, I'm communicating what is going on.
Didn't mean to come off as annoyed either, but when I describe it reading high at idle in P not just in drive, I don't see how the torque converter could be involved (but appreciate the ideas and input alwaysold man!)
Just thought I decribed very specifically what the issue was. I know what the issue is, I need help knowing how to diagnose the cause of the faulty gauge reading and fix it. Pathology.

I'm not being resistant to others' opinions, I'm trying to state as clearly as possible, what the mechanics and I have determined the issue is and what I have confirmed it isn't with my "feelings" and with their scanners that verified those feelings exactly, and asking how to diagnose the prob and fix it!!!1. Can I replace just the tach on a seemingly 1 piece gauge cluster? How are they put together? How are they wired? etc.

I have verified that everything is working as it should with a scan gauge and the mechanic's very expensive scan station. Ran it for an hour around town with a very expensive portable scaner the mech let me use. No surges no nothing, the scanner says everything isjust perfect.
At idle for instance the scanner says 800rpms+/- however the gauge says 1200-1250+/- rpms at idle. This discrepancy is consistent through all gears and conditions which makes me think it is either a physo=ical problem with the gauge or an electrical issue between ECU and gauge.

There are no anomalies of any sort, the Jeep is running and operating perfectly aside from an inaccurate gauge.
When I floored it, it's like the gauge skips a tooth or something then finds it's right spot months later after normal driving..
Yes 96 is the first year of OBD2 but I can do the on off on off on thing.

IS there a way to replace just the gauge?
I'm also wondering why after a while last time it just eventually returned to reading normally?
I guess my real inquiry would be:
"Why did it go up 500 rpms permanently after flooring it?" Does the gauge have actual teeth/gears that could've skipped?
And
"How would I go about diagnosing where from the gauge to I guess the ECU the problem lies? I know nothing about electronics.
Thanks for everyone's input BTW.
Hopefully someone has maybe a step by step how to diagnose/trace the issue.
Thanks again guys!
Hopefully someone can help with this lol.

The gauge is an electromechanical device and utilizes a magnet spinning inside a metalic hub. If metalic particles get inside the gauge, it can cause it to bounce around or simply read too high. You could pull the gauge and flush it with an electronics cleaner/lube and see if that fixes it. The problem is that a can of the cleaner/lube will cost you somewhere north of $10. You could probably grab a tach from a cluster in a pull-n-save yard for about the same amount. I have worked on several of the tachs and modified them from 4 cyl to 6 cyl and vice versa, so I know the electronics. Nothing in the electronics should be capable of causing what you are seeing, unless it is a one in a million failure.

I have a speedo on my Ford Exploder that I am going to have to do this to this weekend. It keeps sticking.

old_man said:

You could pull the gauge and flush it with an electronics cleaner/lube and see if that fixes it.
Nothing in the electronics should be capable of causing what you are seeing, unless it is a one in a million failure.

Click to expand...

Thanks old man!!! I wondered how it could be electrical if all is functioning right.
So just the tach can be replaced eh? I thought the whole cluster was a 1 piece deal. Guess I'll be heading to the JY today.

old_man said:

The gauge is an electromechanical device and utilizes a magnet spinning inside a metalic hub.\

Click to expand...

Seems to me that 96 was the 1st OBD II year and All readouts come from the PCM.