Hmm....Hot or Cold.... (Temp Gage Q)

[Lunatic]

Hey all,


I have a temp gage which isnt working. Thats the simple conclusion.

I have searched the forums and found some info - But, although helpful, nothing that I found actually provided a solution.


So, here's what I know thus far:

- "Its the green wire"

- The temp sensor is on the manifold, sending unit in the cylinder head.

- Chilton schematic available, circuits traced to printed board behind instrument cluster - PINK/BLACK for power, BLACK for ground, GREEN for signal


Here's what Ive tried so far:

- Jumped from GREEN connector point (at the cluster) to anywhere in engine bay, searching (ringing) for the missing green circuit. One was found, but Im not certain where it tails to. The wire I found comes off the back of the fuse block, and connects to a small orange connector that has two green wires (together), drops behind the distributor, comes up to a branch in the harness above the RH cylinder head, but does not follow through with the wiring which runs along the cylinder head for engine accessories - Its as if the green wires (which ring) branch back into the dash, behind the glove box, through the firewall.

- New sending unit installed.

- New sensor on shop bench to be installed tomorrow (Although I dont think anything will change).

- Jumped sensor connector terminals to test for a sign of life from the gage - Nothing. I can only assume this is due to the circuit being incomplete.

- The gage does have power, verified, and twitches with the turn of a key.



What am I missing here ? It doesnt help to have half-stepped Chilton schematics for in depth circuit tracing, but it is currently all I have.

Perhaps some of you might have some insight, or a suggestion or question of your own which will trigger another step which stumbles across a solution. What would be most helpful (IMO) would be standing next to another rig which has an operable temp gage - But I wanted to pick everyone's brain before sending that particular PM.


Thanks for reading, and TIA for any help, insight, ideas, etc provided. Questions certainly welcome, Im sure theres more which Im not thinking to provide at this moment.




Troy B
Ft Hood, TX

 

[4X4HIGH]

Green wire goes from sending unit on drivers side head into the cab and hooks directly to the gauge. If you touch said green wire to ground the gauge should peg HOT.

 

[Lunatic]

I thought about doing this.... What you have said.

I considered running a single wire to the gage, perhaps splicing the green wire at the cluster board connector, to see what happens. It makes sense, but would really ......annoy..... me if something started smoking from my curious idea - Which is when I chose to thread up


Chilton schems show GREEN from sensor to connector, tapped via ignition. It is a simple circuit to say the least but it'd be nice to actually find the physical OEM wire prior to re-installing the engine side of the harness in a custom manner.


Thanks for the reply 4X4HIGH



Troy B
Ft Hood, TX

 

[bp71k5]

I think Scott was trying to help you determine if the guage was broken or the sender. If you remove the green wire from the sender on the cylinder head and touch it to ground, a "good" guage will read very high temperature. If it doesn't read high, then the guage or the wiring to it is bad.

If the guage reads hot (like it should) when you ground the green wire, then the sender is bad.

 

[4X4HIGH]

I think Scott was trying to help you determine if the guage was broken or the sender. If you remove the green wire from the sender on the cylinder head and touch it to ground, a "good" guage will read very high temperature. If it doesn't read high, then the guage or the wiring to it is bad.

If the guage reads hot (like it should) when you ground the green wire, then the sender is bad.

Yep, i should have made myself a little more clear. But still the green wire goes to the gauge and i think the power wire goes through the ignition switch so that the gauge only reads when the key is on.

 

[British Slave]

Try this:

General Motors 0-90 Ohm Gauge Diagnosis Procedure​

Especially as cars age, it is often somewhat difficult to pinpoint the root cause of instrument cluster gauge failure, thanks to questionable wiring, normal wear and tear, and sometimes even incorrect diagnosis procedures. The following diagnosis procedure is the method that GM publishes in it's shop manuals, will work on oil pressure, temperature, and fuel gauges, and should apply to all but the most recent GM-made vehicles with computerized gauges.
Before You Begin
You will need either Kent-Moore tool J-24538-A, if available, or tool JG-GM-1, a homemade GM gauge testing jumper. To build tool JG-GM-1, you will need the following materials:

• 1 Jumper wire with alligator clips (about $2 US from Wal-Mart)
• 1 220 ohm, 1/2W resistor (Radio Shack #271-1109, pack of 5, $0.99)
• 1 150 ohm, 1/2W resistor (Radio Shack #271-1111, pack of 5, $0.99)
• about 1" of 1/4" heat-shrink tubing, or vinyl electrical tape

Construct the jumper as follows:

1. Cut the jumper wire in half, and strip about 1/8" of the insulation off of the cut end of each wire. Tin the bare wires using a soldering iron.
2. Take both resistors and twist their leads together on each end, so that the resistors are joined in parallel. Twist the leads tightly, cut off the excess, and tin the twisted leads. If using heat-shrink tubing, make sure that the resistor assembly is about half the length of the tubing, or less.
3. If using heat-shrink tubing, slip the tubing over one of the jumper wire halves.
4. Solder each half of the jumper wire onto the opposite ends of the resistor assembly.
5. If using heat-shrink tubing, slide the tubing over the resistors and solder joints and shrink using a low-temperature flame or hair dryer. If using electrical tape, tightly wrap the tape around the resistors and solder joints, ensuring that the solder joints are completely covered.

Diagnosis Procedure

1. Disconnect wire from sending unit, and connect tool JG-GM-1 from the wire connector to ground.
2. Turn ignition ON.
3. If the gauge being tested reads at the high end of the scale (i.e. full for gas, 60psi for oil, etc.), the sending unit is at fault and must be replaced. Retest after sending unit replacement.
   Else, go to step 4.
4. If the gauge responds, but not accurately, go to step 5.
   If the gauge does not respond at all, go to step 7.
5. Disconnect engine harness connector at firewall bulkhead. Connect tool JG-GM-1 from ground to bulkhead connector lead that goes to gauge. If the gauge now responds accurately, check the wiring between the sending unit connector and the engine harness connector, repair any faults, and retest.
   Else, reconnect the engine harness connector, and go to step 6.
6. Remove the gauge from the instrument cluster and check the speed nuts holding the ceramic shunt across two terminals on the back. If they are loose, tighten them and retest gauge; if tight, the gauge is at fault and must be repaired or replaced.
   NOTE:You might be able to repair the gauge with a kit from Fix-A-Gauge, and/or you might be able to manually reposition the needle where applicable. Retest after repair.
7. Remove the instrument cluster and check for bad connections at the gauge terminal clips, instrument cluster connector, and both the instrument cluster and sending unit ground. If the connections are good, inspect the wiring from instrument panel to firewall harness connector, and repair any faults if found; if bad, repair the connections, reinstall gauge, and retest. If the connections and wiring are both good, the gauge is at fault and must be replaced.
   NOTE:A gauge diagnosed as bad through this step most likely can not be repaired.

 

[Lunatic]

I think I could have been a little bit more clear myself.... It is the wire which needs to be grounded to test the gage.... That is missing entirely.

I will run a jumper and post results. I'll even take a few pics to clarify.


Thanks for the help guys !



Troy B
Ft Hood, TX


Initial post edited due to initially switching sending unit and sensor identities around. Result being "new sending unit installed, new sensor on shop bench", all other references edited accordingly

 

[Lunatic]

OKay - So I ran an outside jumper - Nothing blew up !

I watched the gage steadily go ..... all the way to 260

So, I decided to run the signal leg through the engine compartment (following the wiring harness) while waiting for the coolant to cool. I soldered a pigtail wire to the back of the gage GREEN terminal, and sent the leg from the engine compartment through the speedo cable grommet, and crimped a connector to it - To connect to the pigtail.

Clean enough on the install to hold me over til the cluster mod kicks off.

So, next I popped the water neck and replaced the thermostat (195*). The one installed looked original. I put it all back together, topped off a bit, and started the engine. Finished topping off and went to look at the gage. The gage climbed...red line.

Just for giggles (which didnt last but a second) I thought to put it in gear to move it up my driveway into the shade. When I depressed the brake pedal....The temp gage pegged.

Oh man - That was the end of my......ambition.....towards this issue for that moment.


Ive had this truck for a little over a year and it has never physically over heated. Its been on road trips in excess of three hours drive, and rolled through a safari zoo for over 6 hours last July. Im wondering if the gage (which checks as good) is accurate at this point.


Ideas ??


TIA


EDIT 8:25PM

Okay, I have now installed the sensor (new) and cleared my bench of .... related items.

A thought occured to me while sitting here thinking of the possiblities of over heating, if the gage is accurate (I actually have two - they both read the same), and what to do next as for ideas towards a solution.

I can honestly say that I didnt feel a pulsing sensation through the upper radiator hose while running earlier, which indicates a weak/dieing pump. Upon that thought I worked the procedure through my head and realized a perfect excuse to drop my belted blade for the Windstar electric.

But that doesnt answer the question of wth is going on. Is the circuit tapped by the brake pedal switch ground - Do I have something wrong. The gage sending unit signal wire is isolated completely, so the brake pedal is back feeding through the green wire in the circuit overall. And being stuck with the half-step Chilton schemo, I cant trace anything and expect it to be - What its supposed to be.

Ugh.




Troy B
Ft Hood, TX

 

[bp71k5]

It sounds like a short on that signal wire somewhere. Did you remove the old wire from the back of the guage before you attached your new one? The old wire could be shorting on something.

 

[Lunatic]

Good idea - Ill completely isolate the signal tomorrow. Hopefully that does the trick.

I post feedback then.


Thanks for the idea !



Troy B
Ft Hood, TX

 

[Lunatic]

Okay, so I isolated the signal wire completely by shrink wrapping the connection tabs on the cluster panel, and the prong on the gage itself. I soldered a connector to the prong, and then the signal tail to the connector.

Status: It seems to be good. Im running at (around).....right above 220 or so - check the pic. I cant remember what Ive read in the past that was average.

I'll DD it this week and see what happens. I think its good. Time will tell.

Thanks for all the help and ideas - I couldnt have fixed this problem as quickly without the brotherhood input !


Troy B
Ft Hood, TX