Temp gauge issues

[RustyPile]

As stated in my introduction, I am a retired automobile mechanic.. I worked in the field for over 30 years.. During the last 10 - 12 years I worked exclusively with electrical issues and drivability problems. Even with all my experience, I'm still perplexed by this particular problem..

Before starting this thread, I searched through about 4 pages of old threads, searching to see if my problem had been addressed in the past.. I found a few similar issues, but none that were exactly like mine.. Forgive me if I've brought up a past issue that has been covered and solved..

I recently bought a 1983 GMC Sierra Classic 1500.. Full gauge instrumentation -- no tach and no idiot lights.. Temp, fuel, and oil pressure gauges didn't work when I bought the truck.. Oil pressure sensor wasn't connected and the sender wire for the fuel gauge was severed.. Proper operation of these 2 gauges was restored when the connections were repaired. The temp gauge would respond to grounding the lead, but the temp sending unit was one for a light.. I installed the proper sending unit and the gauge would never move from any position where it "parked", no matter the coolant temperature.. I obtained a new reproduction gauge spec'd for '73 - 87 models. Although the gauge has a different face with "H", "C", and hash marks instead of temperature marks, the vendor assured me the gauge would interchange with the original. When installed, the gauge would move to the "C" and never move with coolant temperature changes.. Suspecting either a defect or not interchangeable, I returned the gauge.. I obtained another reproduction gauge from a different vendor.. It gave the same response -- no matter the coolant temperature, the gauge never moved off the "C". I checked the resistance response of 3 different sending units -- all ranged from approx. 1400 ohms "cold" to approx. 250 ohms boiling point. The resistance response range of the 2 gauges ranged from 150 ohms needed to just lift the needle off the "cold peg", 65 ohms for mid scale, and 30 ohms for a full scale hot reading.. I've thoroughly checked the wiring, PC board, terminals, and contacts for the instrument cluster and found no problem areas.

What am I missing here?? do I have gauges and/or sending units from a bad batch from the factory?? wrongly applicated (is that a word?) gauges and/or sending units?? I'm wondering if these reproduction gauges require the older sending unit??? Anyone else experienced this?? Should I place a gauge order with a totally different vendor hoping to get a gauge from a different batch?? What gives? Any and all help will be appreciated..

Thanks, RustyPile

 

[Camar068]

curious if there is a thermostat in there? Maybe you have a faulty one and it's not getting up to temp? You got me curious so I went out and ran a few tests.

Mine is an 86, not sure of the differences. I have a decade box (adjustable resistor box.....no guessing, you know what you have it set to). Below, I hooked one end to negative on the batt, and the other to the sender wire. The decade box is basically simulating the sender.

-wire at sender to battery ground = gauge pegs Hi temp....over 260° (if it doesn't peg, I would suspect a bad wire or gauge)
-100 ohms between that lead and ground at the battery = 220°
-50 ohms between that lead and ground at the battery = 270°
-150 ohms between that lead and ground at the battery = 180°

So the lower the ohms, the higher the gauge goes.

I also measured the voltage at that wire while disconnected. Mine was 11.6V. My battery voltage across both terminals (not running) is 12.8V. So one thing you can do is pull the sender wire and measure the voltage between it and battery negative. If it's not near battery voltage, it could be a wire in the engine bay or the dash. If it is, I would suspect a bad sender and/or a bad ground to the engine.

Hope this helps.

 

[RustyPile]

Camero68,
thanks for taking the time to write a detailed response. Although the thermostat was my first suspect,it's not the problem. A thermometer in the radiator tells me the engine does come up to operating temperature... The reason for the lower than battery voltage reading at the sender end of the wire is due to the internal resistance of the gauge.. Removing the gauge and installing a jumper wire across the "batt" terminal and "send" terminal will eliminate the voltage drop if all is well in the gauge wiring.

I have a decade box in my tool arsenal. The original gauge was marked in degrees, but the reproduction gauges I have aren't marked in degrees. Instead, they have "C" and "H" designations and hash marks in between.. My decade box results are a bit different from yours, but still very similar. At 150 ohms, the needle just lifts off the "peg" at the "C" mark. Around 60 ohms or so, the needle is at mid scale. 30 - 35 ohms puts the needle on the "H" mark.. Manufacturing tolerances could explain the difference in our results. The grounds have been checked -- I even added a couple more using braided cable for added flexibility..

I've been at this for a month or more, been over everything.. The reproduction gauges are spec'd to fit '73 - '87 models.. The sending units for models prior to around '79 or '80 have a "button" type terminal, while the later ones have a spade terminal.. I haven't gotten my hands on an earlier sender to check the resistance range, but I'm wondering if the resistance range is different on the two sending units.. If it is, then the reproduction gauges have to be paired with the earlier sending unit..

 

[chengny]

For Truck Years: 1974 - 1978

The Temp Gauge Needle Should Point to:
Left Line (Cold) when sender resistance = 350 Ω
Middle Line when sender resistance = 68 Ω
Right Line (Hot) when sender resistance = 46 Ω

For Truck Years: 1979 - 1990

The Temp Gauge Needle Should Point to:
Left Line (Cold) when sender resistance = 1,365 Ω
Middle Line when sender resistance = 96 Ω
Right Line (Hot) when sender resistance = 55 Ω

I would first check the sender's resistance under actual operating conditions. Remove the sensing lead from the sender and - with the engine cold - get an initial resistance reading. One probe on the sender's terminal and the other to a clean nearby spot on the cylinder head.

As the coolant warms up, keep checking to confirm that the resistance across the sender begins to fall. Since you are getting absolutely no reaction out of the indicator, don't be overly concerned with the actual ohmic values. Just confirm that resistance does decrease as a function of coolant temperature rise.

If you are interested, here is the temp/resistance matrix for 79-91 GM coolant senders:

You must be registered for see images attach

If the resistance tracks downward with an increase in coolant temperature (even if it's a bit off), I would say you could eliminate the sender end of the circuit as the problem.

As you can see from the chart, the indicator would like to see a resistance of about 1600 ohms to bump the needle off "C" (100 F). And, when the engine reaches normal operating temperature (coolant @ 195 F) resistance across the sender should be about 240 ohms - which should drive the needle to just left of the midway mark.

 

[RustyPile]

Chengny;
Thanks for the chart, it's much more detailed than the chart I've developed while checking the operation of 3 different sending units spec'd for my truck. I am thoroughly convinced that I don't have a bad sending unit..

Your opening statement and last paragraph provides a clue to my problem.. To reiterate -- I have determined that the original gauge was defective. I purchased two reproduction gauges from two different suppliers. Both are spec'd for '73 - '87 model trucks.. Neither gauge would indicate anything other than slightly "off the cold peg" readings.. One was returned as defective or not for my application. I was allowed to keep the other one and refunds were issued in both cases. No indication for the sending unit needed to make these gauges function properly in the spec'd model range was noted with the gauges. The resistance response range needed by both gauges is below that produced by the sending unit spec'd for my '83.. I haven't tried an earlier model sending unit, but I suspect that might be the solution...

 

[RustyPile]

Just to bring closure to this thread.. I had family business that needed my attention most of last week so the GMC work was placed on hold until the 15th.. I managed to devote considerable time to the temp gauge problem the past 4 - 5 days and have the problem solved..

I abandoned the reproduction gauge and got after the bad original gauge.. I checked the gauge resistor on the back (something I should have done in the first place) and found it to be open.. A few phone calls to friends in low places and they confirmed the resistor value to be 90 ohms +/- 8 - 10 ohms.. As a ham radio operator, I have quite a stash of misc. parts. I came up with an 87 ohm resistor, put it on the gauge, put the gauge back in the cluster, and warmed the engine up.. With a 180 degree thermostat, the gauge reads about 1/3 scale, a little below the 210 mark, and 326 ohms across the temp sending unit...

Sometime in the future when curiosity overtakes me, I'll rig up a test circuit and test the reproduction gauge using an earlier sending unit.. At this point, I'm not wanting to divulge the names of the dealers that supplied the two reproduction gauges as I don't think they were at fault in this..

 

[olnick]

In your first post above you answered your own question when you stated " the temperature sender was for a light not a gauge" at some time it got hot enough to close the contacts then the sensor wire at the gauge went to ground through the 90 ohm resistor and blew it out! Glad that you finally did find it. Why GM made changes to the gauges and sensors is beyond me However it does cause a ton of confusion when we are working on the things and the parts look and fit in the original location!

Olnick

 

[Backfoot100]

I wanted to revive this because there's some really good info in this thread and I have a lot to add. As an FYI, have the later style gauge and sender for an '86 350.

I had a perfectly good working temp gauge til about three weeks ago. Then it just pegged high one day. I disconnected the sender wire at the sender and it was still pegged high.
I read the ohms of the sender with the sender lead disconnected and witnessed the ohms decreasing as the engine warmed up. At this point I'm not concerned what the ohms read, just that they decrease as engine temp increases. Sender is good.
I then tested the sender wire from the sender to the gauge and it ohmed out good. Nothing was grounded either.
Hooked everything back up but I did not have the gauge installed. I verified that the sender lead at the dash panel where the gauge would plug into it does indeed read the same ohms as the sender and again, its not grounded. Everything checked out fine so the issue has to be the gauge.

I found that the resistor on the back of the gauge was open as others have stated in this thread. Reading here or on another thread that the resistor might be 85 or 90 ohms. I also have read speculation somewhere that the gauge might be setup to read a little cooler from the factory so that it doesn't run above the 210 mark (which is the halfway mark on the gauge). My factory gauge always seemed to run a little on the cool side. By that, I mean that it rarely went much above that mark between 100 and 210. I just assume that mark is 180. At least that's what I call it for argument sake. The only precise reading you can get with the factory gauge is 100, 210 or 260. Everything in-between is purely guessing.

I have a good supply of various resistors so I played around a bit.
A 47 ohm resistor on the back of the gauge will never reach the 180 mark of the gauge. A 120 ohm resistor makes the gauge run right up to just touch 210 on the gauge.
I ended up using a 100 ohm resistor which makes the gauge run right at 190-200 or right about halfway between that 180 mark and the 210 mark. Running various temp checks with my IR thermometer confirmed that engine temp was indeed 195-200 depending where you shot it at.

Either way, the 100 ohm resistor gets MY gauge to read how I like it for MY engine. I gotta assume each engine, sender and gauge combination will read slightly different.

The point here is if you take some time, you can actually tune your gauge to read how you want it to read. Kinda like taking the time to actually tune your carb, which very few ever do but they're the first ones who insist on getting EFI.....Sorry, I digress but that's a particularly sore subject with me.

Anywho.....just some neat information for anyone who wants it.

 

[Old60Driver]

I wanted to revive this because there's some really good info in this thread and I have a lot to add. As an FYI, have the later style gauge and sender for an '86 350.

I had a perfectly good working temp gauge til about three weeks ago. Then it just pegged high one day. I disconnected the sender wire at the sender and it was still pegged high.
I read the ohms of the sender with the sender lead disconnected and witnessed the ohms decreasing as the engine warmed up. At this point I'm not concerned what the ohms read, just that they decrease as engine temp increases. Sender is good.
I then tested the sender wire from the sender to the gauge and it ohmed out good. Nothing was grounded either.
Hooked everything back up but I did not have the gauge installed. I verified that the sender lead at the dash panel where the gauge would plug into it does indeed read the same ohms as the sender and again, its not grounded. Everything checked out fine so the issue has to be the gauge.

I found that the resistor on the back of the gauge was open as others have stated in this thread. Reading here or on another thread that the resistor might be 85 or 90 ohms. I also have read speculation somewhere that the gauge might be setup to read a little cooler from the factory so that it doesn't run above the 210 mark (which is the halfway mark on the gauge). My factory gauge always seemed to run a little on the cool side. By that, I mean that it rarely went much above that mark between 100 and 210. I just assume that mark is 180. At least that's what I call it for argument sake. The only precise reading you can get with the factory gauge is 100, 210 or 260. Everything in-between is purely guessing.

I have a good supply of various resistors so I played around a bit.
A 47 ohm resistor on the back of the gauge will never reach the 180 mark of the gauge. A 120 ohm resistor makes the gauge run right up to just touch 210 on the gauge.
I ended up using a 100 ohm resistor which makes the gauge run right at 190-200 or right about halfway between that 180 mark and the 210 mark. Running various temp checks with my IR thermometer confirmed that engine temp was indeed 195-200 depending where you shot it at.

Either way, the 100 ohm resistor gets MY gauge to read how I like it for MY engine. I gotta assume each engine, sender and gauge combination will read slightly different.

The point here is if you take some time, you can actually tune your gauge to read how you want it to read. Kinda like taking the time to actually tune your carb, which very few ever do but they're the first ones who insist on getting EFI.....Sorry, I digress but that's a particularly sore subject with me.

Anywho.....just some neat information for anyone who wants it.

Backfoot, thank you for this post, and the information found within.

I've run into a situation similar to yours. Both my voltmeter and temperature gauge are inop. The voltmeter never moves off of 0, my temp gauge reads maxed out and fluctuates very slightly after engine heats up. I've determined that both of the resistors on back are open. From what I've read one should be about 90 ohms, and the other should be about 135 ohms.

Not to hijack your thread, but can I just build my own resistor set (ceramic stunts are hard to find), and if so, what wattage should I be looking for.

Thanks again!

 

[Backfoot100]

Not a hijack at all my friend. Glad to help out.
I just laid the new resistor right on top of the blown ceramic one, hooked the wire leads around the terminals and screwed the wire nuts down.

I bought this kit of assorted 1/4 watt resistors from Amazon. It has 1200+ total count of 66 different resistor values for $10 or $11.

You must be registered for see images attach

They have smaller and cheaper kits but this one has enough different resistor values you don't have to use multiple resistors wired in series or parallel to get the overall resistor value you're looking for.

 

[kuhnzoo]

Been trying to resolve a temp gauge / sending unit problem lately. I believe I have the data I needed after reading through this.

Just wanted to say "Thank You" to everyone that contributed to this thread. Lot's of helpful info.

 

[Kyle K]

Hey @Backfoot100 , quick question on this post. I'm fighting the same deal as mentioned above. Boiled more sending units than I can count. Going to try the resistor you mentioned. I haven't done it before, so curious, do you remove the current ceramic one, and then just wrap the ends around each top/bottom post? that's the orientation on my '73 for the ceramic resistor (top/bottom). I saw another forum about the left/right posts, but believe that as for a different year. Thoughts? Thanks!

 

[Backfoot100]

Hey @Backfoot100 , quick question on this post. I'm fighting the same deal as mentioned above. Boiled more sending units than I can count. Going to try the resistor you mentioned. I haven't done it before, so curious, do you remove the current ceramic one, and then just wrap the ends around each top/bottom post? that's the orientation on my '73 for the ceramic resistor (top/bottom). I saw another forum about the left/right posts, but believe that as for a different year. Thoughts? Thanks!

If the ceramic resistor is open, just leave it on and lay the new resistor over the top and wrap the wire ends around the posts. Then spin the speed nuts down on the threaded posts to secure the new resistor in place.
Easy peasy.