My temp gauge doesn't work. Or I didn't think it did...

Grounding the temp sender wire in the engine bay yielded no response in the cluster. So I resigned myself to pulling the gauge cluster. When I finally got the face off, I found the needle was merely wedged against the bevel cover. Result! The needle was just jammed and it bounced back up to the lowest mark on the gauge. Plugged the cluster back in and called my other half over and asked her to ground the sender connector in the engine bay. As soon as she did, the needle pinged down hard and wedged itself back under the bevel! It went completely the wrong way!

So, what would cause that to happen? The wiring under the dash looks pretty new, so I guess the previous owner wired something up wrong. Everything else works, so did he just swap the connections?

I found this diagram of the reverse of the cluster in this thread:
https://www.thirdgen.org/forums/electronics/718393-88-gauge-cluster-pinout.html
Has the PO simply wired connector 7 and 9 (ground?) the wrong way round?

I just bought a shop manual to try to make these things easier, but the damn thing arrived with the electrical troubleshooting section missing!! Gaaahh!!

Thanks for any help or pointers!

 

First some info!
Info
Assuming your gauge looks like this. . .



Notice the curvy lines (coils) on both sides of the pointer for the temp gauge

The two coils generate a magnetic field that pulls the pointer towards one or the other. The coil for the min side has a fixed resistance and the side for max has a resistance that changes with temperature (the actual temperature sender). As the temperature sender heats up it lowers resistance, until finally it has less resistance than the min side and starts pulling towards max. Since the grounding test did not cause the gauge to show max here are possible causes:

Causes
1) The pins are reversed. If they are, the gauge should work but in the opposite direction, the gauge would start at Max and move towards Min as temperature increased.

2) There is a bad ground at the temperature sender (teflon tape or something), if that were the case when grounding the wire the gauge would of shown Max.

3) The temp sender wire is not connected (wrong pin, cut wire, etc), this would mean there is only one path for the current to take ( the min).

4) The gauge is out.

5) The wrong temp sender is installed or it is faulty. If the sender has a high a resistance, it will never go below the fixed resistance on the min side, thereby showing min ( maybe moving a little, maybe not). Quick check is to get an ohm meter and check the resistance of the sender as you submerge the sensing part of it in boiling water.

That's all I could of think of at the moment, if I put some bad info feel free to correct me! Hopefully that gets the ball rolling. I can't even find my temperature sender wire, likely got cut by PO because it melted or something.

Good luck!

 

Thanks a lot for that information - that diagram is just what I need!

The info on how the gauge works is really handy, and is helping me try to get my head around how the gauge is behaving.

The sender's definitely connected to SOMETHING as the needle does respond when the connector's grounded in the engine bay.

Staring at that diagram, it's like the sender wire from connection 7 in C2 is connected to the cold side of the dial instead of the hot side, but the hot side must be disconnected (otherwise the needle would peg to hot when the key is turned, and peg to cold when the sender connector's grounded.... right?!)

But how can that be possible? The gauge and the PCB in the cluster are all fixed, so it must be an issue with the wiring to the connectors behind the cluster. However, the only way the sender wire could be connected to the cold side of the dial, would be via connection 9 in C2, which is a shared ground... so how can that be possible??

That said, the ground shares connections with the indicators and the tachometer. Both of which work rather sketchily. The tach over-reads, and the indicators blink slowly... but would they work at all if connection 7 and connection 9 were swapped over, or would they still find a ground (with higher resistance) via a different route?

On a slightly separate note, could that be why my tach over-reads? I took it off to solder the resistor, but the pcb looks in really good nick... I need to check it with an ohmmeter tonight.

Finally, that diagram is from a 1989 - 1991, would it be very different from my 1985?

Thanks! Sorry for the stream-of-consciousness response, I'm trying to figure things out as I go!

Ah, this might be setup:
http://www.austinthirdgen.org/mkport...uges_part2.gif

My car does have a tach filter (I think....!)

 

I was checking my tachometer last night with an aim to repairing it as per this thread:

https://www.thirdgen.org/forums/electronics/640522-82-89-tachomoeter-fix.html

with the intention of applying 200k ohms across resistor R4 on the white chip. My tach doesn't peg, just seems to read considerably above what it should (idling at 3k etc.) However I found R4 to be an open circuit, which should peg the tach on startup... but mine doesn't. I checked, double-checked and triple-checked and I was definitely testing the correct terminals!

Looking at the diagram for the 1985 cluster, the only way the temp gauge could be wired the wrong way (I think) is for the tan CCC connection (31) to be crossed over with the dark green coolant sensor connection (35), which would also affect the ground to the tach and oil gauge. I've not observed strange behaviour in these when the temp sensor was grounded... but then I wasn't looking for it! Any ideas? Would this wrong ground to the tach account for the tach NOT pegging to max on key-on as it should with R4 open?

Does anyone have pin out diagrams for the cluster connections for 1985? I can find 1986 online,

https://www.thirdgen.org/forums/attachments/electronics/232562d1326709897-86-camaro-gauge-cluster-1986-camaro-cluster-pinouts.jpg

but I think this year was different? Perhaps the PO reading a diagram for the wrong year explains what's happening!

 

Assuming I'm correct lol
For now let's just focus on the temp gauge for now lol good enthusiasm though!

It's possible that the gauge is dying, perhaps the coil for the temp sender path is wearing out causing a very weak magnetic field? That could explain how there is some change in the pointer when you ground it.

It's also possible that the clear plastic PCB on the back of the gauge pod is incorrect? They all do seem too look the same, and if the wrong year was swapped that could cause some of the issues you discussed about swapped pins.

For me it's 2 am lol, I was going to respond earlier with more detail, but it would of been through my phone and it was just a PITA.

As for the tach, typically it's a capacitor (as those wear out) issue, however I just recently fixed mine by soldering a by replacing certain pins in the Resistor Tree IC with a 300 Ohm (or maybe KOhm?) to the PCB for the tach. My tach isn't precise down to the single digits, but it reads correctly (or maybe within 50-150 RPMs). However it'll likely degrade over time since I didn't replace the capacitor, but by that time I'd likely replace the gauges with that sweet all digital one lol

Any luck so far?

 

Haha, yeah I know! It's getting complicated!

I brought the tach into it, as the wiring diagrams suggest the tach, oil pressure gauge and temp gauge share the same ground. So if there is an issue with the wiring, I'd expect to see some reaction with the other gauges... if I understand how they work!

There could still be a problem with the temp gauge itself, but I'm still thinking it's a wiring connection. The needle didn't just move weakly when the sender was connected - it pegged hard the wrong way and wedged itself in place. Makes me think the gauge might be fine and it's the wiring that's wrong.

Not sure how likely it is that the PCB is wrong - the car's been in the UK since 1989, so it's unlikely the PCB's been changed. There's no legion of abandoned F-bodies in the junkyards here!

I'll have to admit I'm nervous of trying to fix the tach. I'm no expert at electronics or soldering, and really need to be 100% confident of what I'm doing before I begin. Plus, I'm wondering if the behaviour of the tach and temp gauge are somehow related.

I won't see the car 'til next weekend now, sadly. So I'm going to have to sit on my hands and do as much research as I can!

Cheers,

Neil

 

Have you tried opening the cluster, removing the temp gauge and connecting a digital volt meter in it's place?

The thin plastic PCBs can delaminate with age, and the 82-89 gauges can fail internally in a variety of ways, but you need to eliminate possibilities to figure out which part of the car is actually malfunctioning. You can eliminate the gauge by testing with a digital volt ohm meter, and you can eliminate the PCB by removing the cluster and checking the dash harness pins with a DVOM. It goes without saying, but the electrical section of the service manual will have a diagnostic chart for temp gauge issues, and 90% of it is going to be eliminating possibilities.

 

Hi,

that sounds like a really good idea - unfortunately I don't know what I'd be looking for. I do have a service manual but, sod's law, it arrived with the entire electrical troubleshooting section missing!!

(I'm trying to get a partial refund from the eBay seller - they really should have mentioned an entire section was missing. Unfortunately half the cost for me was postage to the UK, so sending it back for a refund isn't going to work...)

That means I don't have the pin out diagrams (I can't find one for 1985 online) so the DVOM approach is a bit of a stab in the dark until I can get the info. I've just purchased the service manual on CD, so hopefully that'll be complete when it arrives sometime next week.

Thanks for the pointer though - I can keep an eye out, and try to research that method. Those cluster pin out diagrams would sure be useful though...

Would I check the gauge itself by simply checking for continuity between the three pins on the back? (one should have continuity with both, right?)

Thanks for the advice!

 

Something wrong with that schematic. It shows the "Hot" side of the temp gauge going to ground through the ignition switch when in "RUN" jumping the temp sensor.


This is more like what it should be, where the ground is in the "Bulb Test/Start". This schematic is from 1988 and may be different for earlier models. Your printed circuit may have some problems giving you the tach problems. Could the temp gauge have been replaced with a newer unit that is terminal wired differently? It appears that although it is a dark green wire that goes to the connector for the printed circuit, the terminals are different numbers (locations) in the connectors.

 

I think this diagram should be closer:



My tach seems to have a filter in the engine bay, so I think my setup predates the solid-state tachs. As it's proving so difficult to find diagrams, I might just pull the cluster back out and use my multimeter to find continuity and work out the connections that way.

Still not sure quite what misconnection could cause the needle to ping the wrong way like it did... Unless the sender signal really has been swapped with the shared ground? Pity that diagram doesn't have pinout numbers!

 

You don't really need the pin out diagrams to eliminate the gauge as a possibility. Remove the black plastic bezel around the gauges, then the clear bezel, and the trim plate. You'll see several silver screws that hold each individual gauge into the cluster. Remove the 2 or 3 that retain the temp gauge and carefully pull the gauge out of the cluster. You'll see three brass terminals. One is going to be a ground, one will be hot in run, and the last will be terminal to the sending unit.

Process of elimination. A test light connected to any ground and probed to each terminal, it'll light up when you touch the terminal to hot in run. From there you can measure the resistance between the other two and ground to determine which is your ground wire. Then you can measure what's happening at the gauge with your voltmeter. If you can verify that the sending unit for an 85 Camaro is the same part number as the sending unit in another year Camaro you can find a diagram for, you can refer to that diagram for the resistance you should be getting from your sending unit.

 

Thanks for the advice Drew, that all sounds perfectly logical - definitely something I can tackle come the weekend. Once I've established which contact should be the ground, I can then trace from the front to the back of the cluster housing to work out which connector pin should be the sender, and then I can look at my connectors and hopefully what's going on should become apparent!

Ok, newbie question coming up….

Can you suggest a good ground I can use in the car? I know in the engine bay almost anything is game, but I'm not so sure in the interior!

Cheers for the help, this is feeling like progress!

ETA: Rockauto lists the same part for temp gauges from '82, '85 and '88, so that should make things easier!

 

1. Easy way to think of this is that the temp sensor is a variable resistor that changes its resistance depending on the temperature.
   
2. The circuit is based on 5 volts(IIRC), and the entire metal body(and engine block) is ground.
   
3. The gauge is a volt meter which varies depending on the voltage the temp sensor produces.
   
4. So from that, as the resistance of the temp sensor changes, so does the voltage on the powered side of temp sensor, then the gauge changes in accordance to the sensor.
   

For instance, when you ground the lead from the gauge, it reads zero resistance. If you were to add resistance between the lead to ground, the meter should respond accordingly. At full resistance, the voltage would be maxed out and the gauge pegged the other way.

• Another way of looking at it is with the temp sensor with no resistance, it would also have no voltage drop between its two leads. (engine block to output of sensor)
• At full resistance the full 5 volts should be present across the same leads.

Draw it out... it will be perfectly clear once you do.



If you look on the back of the cluster, the ground circuits are typically wider and goes to one side of every component on the circuit. (It carries the most current by default.)

Easy, simple, no schematics needed.

 

Thanks Scorpner, with a little frowning I think I've got my head around that!

Good tip on how to recognise the ground wires, that'll be very useful when I've got the cluster back out again.

 

Ok!

A little bit of sleuthing pulled up this image from eBay (I hope the seller doesn't mind!)



I now feel a bit of a plonker for suggesting I'll need a multimeter, as the connections are flippin' obvious! This pin-out diagram is from 1986 but, at least as far the temp gauge is concerned, it looks correct.



Right. I'm guessing the PO wired the No. 7 connection (sender wire) to No.5 (gauge power blk/pnk - coolant temp) hence pulling the needle the wrong way. As the needle doesn't max, I'm guessing he hasn't simply swapped 5 and 7, but the gauge power (blk/pnk connection) is missing, or shoved in one of the redundant pins...

So, I'm now desperate to get back to the car and find out exactly WHAT is there! Worst case scenario (I think) is a missing blk/pnk power wire. From those wiring diagrams I'm not entirely sure how that's connected up with the other blk/pnk gauge power wires. If I can't find a spare one, will I simply have to splice into one of the other blk/pnk wires, or take a wire directly from the fuse block?

Getting nearer.........

 

Good to know! I hope it helps.

 

Oh, I didn't see the latest reply. I'm not sure where #5 routes to, schematics might help with that actually. I’m guessing eventually to the fuse block for the cluster. Looks like you have the hardest part figured out now.

 

Ok, I'm guessing it's the temp gauge itself.

Armed with a week's-worth of research, I pulled the cluster back out and checked everything. All connections seemed correct, and everything responded to the DVOM about as I expected (although I did pull nearly 12V from pin 9... too high?). I also checked the connections on the back of the cluster, and registered virtually zero resistance between the connector pins and the gauge connections.

So, I plugged the cluster back and repeated grounding the sender wire. Switching the car on, the needle pinged to the lowest mark on the gauge (yay!), but grounding the wire saw the needle snap to the top of the gauge (as in, right off the gauge). Three out of four times, it would return to the lowest mark, every one in four times it would flick off the bottom of the scale and get stuck. Occasionally it would also wedge itself at the top of the gauge.

For now, I've just put it back together with the needle free and I'll make a mental note NOT to ground the temperature sender. When the car finally hits the road (should be next month!) I can see if the gauge is doing something normal.

Does pinging off the scale sound like a dodgy temp gauge? New replacements from Rockauto aren't too expensive...

 

No way. The factory temp gauge NEVER fails! -/sarcasm-

Actually the factory temp gauge in the Camaro frequently fails in this manner. That's why I asked you if you had tried bypassing the gauge.

If yours is anything like mine, once the needle is carefully bumped back to the free position it'll read correctly. If that's the case, I just made a couple 'stops' for the needle so it can't swing far enough to get stuck. I used a couple pieces of thin stainless steel wire about 1/4" long, and glued them to the rivets that retain the gauge face. After some very careful application of black paint with a fine brush they look factory. Not that it matters anyway since that part of the gauge isn't visible when the cluster is assembled. With the added stops, the gauge still flings the needle around violently during the gauge test, but it just hits the stop and returns to the normal position.

 

Ah, ok - glad to have some reassurance that the gauge will probably work as it should.

Next time I'm at the car (hopefully putting the tach back in…) I'll hunt out something suitable to stick on as needle-stops for the temp gauge.

Seems like a long journey just to end with gluing little thumb tacks on the gauge, but I think I learnt an awful lot in the process!

Many thanks to everyone who contributed - this really is a great forum!