Tachometer gauge not reading properly

The needle of the tachometer just jumps around. I just did a cluster switch to get the tachometer, but now the tacho is behaving like crazy. I am wondering if it's an issue of the gauge or if the signal might not be 100% correct. Since I never had a tacho before, it's difficult to know.

What should I look for? Is there any way to read the signal with a multimeter or is it a pulse which cannot be read with a normal multimeter? Where does it come from so I can clean connectors or what has to be cleaned.

Regards

Well, I doubt this well be much help. I don't know what year your Jeep is, but if it was after Chrysler bought out Jeep, this might be relevent. I used to have a 93 Dodge Daytona that had a very similar problem, below 2 or 3000 RPM the needle would jump around like crazy, However it always went away after a few start-stop cycles (I'd go somewhere, turn off the car, start it up again after I was done). It would take anywhere from 1-3 of these cycles before it would fix itself. It generally only did this after it had sat without being run for and extended period of time.

Like I said, it's not much help in finding the actual cause of the problem, But you can try driving it around some (Or just starting it up, rev it a bit, and turn it off a few times), and see if it "fixes" the problem.

Pat

Well, I tried. By the way, it's a 1988 XJ. It just doesn't go away, but the gauge I swapped in might not have been used in a long time. I tried moving the needle carefully and it moves up freely and returns just as it should.

The tach gets it's signal from the ICM located under/attached to the igniton coil. I'd try unplugging the 2 plugs going to it and clean the connections with WD-40 or something similiar. Might simply be a corroded terminal.

I believe the tach feed is the green wire (w/tracer) coming from the ICM.

OK, I've never moved anything with the ICM, in fact this is my first XJ / Jeep (I had a Toyota previously), so I'm not familiar with it. The ICM should be between the pressure tank and the battery, just over the passenger side front tire right?

schmiedel said:

OK, I've never moved anything with the ICM, in fact this is my first XJ / Jeep (I had a Toyota previously), so I'm not familiar with it. The ICM should be between the pressure tank and the battery, just over the passenger side front tire right?

Click to expand...

No it's under the coil on a I6 4.0

OK, then one more question. As far as I've seen, the tachometer is a 12V pulse signal, am I right? How can I test it without an osciloscope? Does anyone have a trick to do it?

I've never tested one, but I don't see why you couldn't run a wire from the coil negative terminal, to the tach signal input. That's basically the way the tach is wired anyway. I'd check the connector first though. As Langer says, the ICM is attached to the ignition coil. Look for a connector with the green wire, which should be the tach signal wire.

schmiedel said:

OK, then one more question. As far as I've seen, the tachometer is a 12V pulse signal, am I right? How can I test it without an oscilloscope? Does anyone have a trick to do it?

Click to expand...

I would think it's an AC swine wave.

Runnin'OnEmpty said:

The tach gets it's signal from the ICM located under/attached to the igniton coil. I'd try unplugging the 2 plugs going to it and clean the connections with WD-40 or something similiar. Might simply be a corroded terminal.

I believe the tach feed is the green wire (w/tracer) coming from the ICM.

Click to expand...

No, no, NO! Clean the terminals if you must, but never use WD-40 as an electrical contact cleaner!

WD-40 isn't even a proper lubricant - it's a "Water Displacer" (that's what the "WD" is for) that carries some mild oils and solvents. These will get under whatever water is on the surface, and "displace" it from the metal.

Catch is, it leaves a slight film of itself on whatever you spray it on - to continue displacing water. This means increased resistance in an electrical connection...

If you can't get something like Berryman's Contact Cleaner or CRC Electric Motor Cleaner, you can use some sort of spray Brake cleaner (prefereably not that non-chlorinated rubbish - I can use urine and get better results...) or even some acetone/MEK/denatured alcohol, following with a CLEAN WATER rinse to remove the solvent.

One more time - WD-40 is not RPT NOT an electrical contact cleaner - don't use it like it is! I've actually made a decent amount of money fixing electrical "problems" that were caused by something like this. No failures - just crapping up the connections.

The best part about it was that the bill was all labour - which all went into my own pocket...

5-90

I have never herd using WD-40 to clean shit. Its more of a lube and penetrator to me. I used rubbing alcohol to clean a few electrical connections on my old C10. Seemed to work, weather it was the right thing to use or not. I would take 5-90's advice. He seems to know his stuff quite well. He also has a PM.

Well, the coil is just where I remembered it was on my XJ. It's between the battery and the pressure tank, above the passenger side wheel.

I found the connector, it has three wires, one green, yellow and black, so I think the yellow should be the ignition signal and the green one the tachometer one.

I found that the green wire is split a few inches from the connector, the connection is only made with some tape. One end goes to a panel that's just aside of the pressure tank, and the other one I couldn't as I am short of time these days.

Now the million dollar question. What is this signal in the green wire used for? If it's only the tachometer, why is it split? It sounds to me as if this was the problem I have. Think the best way to find out is to separate the connection and turn on my rig, see how it behaves. Any thoughts of this?

Regards

Now the million dollar question. What is this signal in the green wire used for? If it's only the tachometer, why is it split?

Click to expand...

From the split, one wire goes to the tach, and the other goes to the diagnostic connector (probably the panel you mentioned - yellow with a snap on cover?).

Exactly, then that's all normal? I was worried about the split covered with normal insulation tape.

I want to test that signal, is there some way to do it without an osciloscope? I will try to clean the terminals, the whole area looks quite messy so I wouldn't doubt it is not clean, but I moved the connector and I see no difference in the gauge, so I want to be able to test the signal.

The second wire goes to Diagnostic connector D1-1; the smaller, 6 pin connector.

but I moved the connector and I see no difference in the gauge, so I want to be able to test the signal.

Click to expand...

Do you mean you removed and re-connected it, and the tach was still erratic?

You might try removing the wire from the diagnostic connector, and see if that helps. It may be shorting out in the connector, although that would probably make the engine missfire.

The tach signal pulses from an open to a ground, every time the coil fires, which is very fast. Too fast for an ohmmeter to pick up, I'd think, although it might give some sort of resistance reading. I've never checked it that way so don't know for sure.

If you have access to another tach, you could attach it to the green wire or D-1, and see if it works correctly......sounds more and more likely that your tach is faulty.

Check the plug on the back of the dash panel and don't forget the feedthrough connector on the firewall.

OK,

I checked everything and it seems OK to me, I haven't found a good connector cleaner, but I connected a multimeter to the D1-1 (to the tachometer signal) and measuered AC voltage; if I accelerate the voltage reading will increase. This is not a good measurement as a frequency change will deliver unexpected results in a multimeter reading, however the results seem more or less constant, so I'm thinking it's not the signal (at least not at the ICM output).

I also tried hitting / knocking on the cluster where the tachometer is and the needle starts to jump in a different way; if I accelerate it'll jump kind of higher (not sure if limited by the RPM value it should stop at), but it behaves differently when knocking / hitting on the cluster.

I'm starting to suspect it's the cluster rather than the signal, but as I said before, I need a way to be 100% sure. What would you do?