Temp gauge bouncing from cold to hot and back to cold over and over again.

[glhx]

I had a temp gauge that wasn’t working at all. I found it to have a broken stud.

So I replaced it with a new one from classic industries.
I also replaced the printed circuit board classic from as well as the old one had some bad tracers
I went ahead and bought the RTE voltage limiter
Then went ahead and ran a dedicated ground to the instrument panel

I tested the original fuel gauge and new temp gauge with my power box. Running 12 volts into the voltage limiter to see what the gauges would do. I had to directly ground each one and they both maxed out because of the solid ground. This showed me that the gauges were at least working. It didn’t show me the variable ground side because I didn’t have resistors to ground them.

I tested the rte voltage limiter output……it runs back and forth 5-6 volts output

I installed the gauge cluster and the new ground.

The temp gauge showed nothing. I figured this was a bad ground because I tested it on the bench and it pegged out.

So I went out to the sensor and saw corroded terminal. I just put a new end on it.
The gauge started bouncing. So now I went from a non working at all…..to a bouncing gauge
It was weird because the temperature would rise……but the gauge kept bouncing while it was rising. Maybe 1/2” back and forth…….

I also have a mechanical gauge that shows correct temperature. So I know what the temp should be reading

I then put a new temp sensor in.

The gauge never rose this time. It now bounces between cold and hot… it sweeps to hot fast and then drops to cold and then back to hot and so on.

Anyone had this happen before or know what could be wrong with it?

 

[glhx]

I just took the wire off the temp sensor. Grounded it direct to the block.

The gauge still bounces between cold and hot.

I don’t know if it’s the voltage limiter……because the gas gauge doesn’t bounce.

 

[67Dart273]

What does the fuel gauge do? Are you CERTAIN that the cluster is grounded?

You REALLY need to get either some test resistors or a variable rheostat so you know what you are up against.

You have made a lot of changes, NEVER assume that "new parts" are "good parts." Could be a bad connection in the board, something wrong with the gauge, or with the RTE.

 

[glhx]

the voltage limiter also runs the fuel gauge. Which is working perfectly. That makes me think it’s not the rte limiter.
(New) never ever worked……

I don’t like making this many changes either.
I changed over to the Rte to get the fuel gauge working. It did not work before. That did fix that.

I did test the gauge with a power pack running 5 volts direct to it and a solid ground. It did max out. It didn’t pulse like it’s doing now.

I will get a rheostat and take the cluste back out for more testing.

As far as the grounds go.
I did add a dedicated ground. It should be ok but I’ll check the connections for ground again.

I did solid ground the gauge after taking it off the temp sensor. It’s stayed the same. I think that isolates the sensor out of the equation.

The fact that the fuel gauge works fine. Makes me think the Rte limiter is good.

Might be the gauge or the ground on the gauge side and not the sender side.

I’ve never seen one pulse like that.

I did also check the board on the bench to make sure all the tracers had connection. I sanded all screws on the underside where they touch the board.

 

[67Dart273]

OK I did not realize the fuel ga. was working. That eliminates the RTE. You are down to connections/ problems on the pc board, and the gauge itself. One thing you can do is pull the gauge and power it with a flashlight battery or two. I have forgotten--try to find the thread. Either 1 or 2 1.5V will bring the gauge up some

 

[RustyRatRod]

Just an FYI. Classic industries is known for selling totally inferior parts. To put it bluntly, junk. So keep that in mind.

 

[DartManPlum340]

OK I did not realize the fuel ga. was working. That eliminates the RTE. You are down to connections/ problems on the pc board, and the gauge itself. One thing you can do is pull the gauge and power it with a flashlight battery or two. I have forgotten--try to find the thread. Either 1 or 2 1.5V will bring the gauge up some

I just did the flashlight battery test 2 weeks ago. Put 4 D-cell batteries in a paper towel tube for support. Connect to the gauge in question. That will send 6 volts to the gauge. It should move up somewhat slowly to near full.

 

[RedFish]

Thermal instruments are not chassis grounded (other than the 3 post gauges with limiter inside).
The low voltage goes in on 1 stud (which is which makes no diffetence), through the gauge winding and out on other stud to variable ground signal provided by a senders path to ground (senders don't send squat). If you have a chassis ground at a 2 post guage can, you have a problem.

 

[glhx]

OK I did not realize the fuel ga. was working. That eliminates the RTE. You are down to connections/ problems on the pc board, and the gauge itself. One thing you can do is pull the gauge and power it with a flashlight battery or two. I have forgotten--try to find the thread. Either 1 or 2 1.5V will bring the gauge up some

 

[glhx]

Amazon Sign-In

I used this to check it.
It allowed me to run 12v into the limiter to test the output of the limiter.
Which is 5-6 volts.

It also allowed me to run 5 volts direct to the gauge.

The box is fully adjustable and will give low voltage power to whatever.

This was how I powered the gauges.
It was the ground I couldn’t do because I didn’t have a ground rheostat. I don’t think that box can do ground variable, but I don’t know.

So I’ll get a rheostat and test ground output as well.

What the box did help me with was dash light.

I was able to find a bad ground that would have driven me nuts inside the car.

One light would not light up. Because I used a wire wheel to clean the screw. It polished the rust and looked shiny. So I sanded it and it grounded.

The power box also showed I had some led bulbs in backward. It is a very comprehensive way to check everything except variable ground.

It showed me that everything worked as far as all the hot wires were concerned.

It is $50 and worth it

 

[RedFish]

An ohm meter should show neither post common to the can. Resistance across a good oil or temp gauge should be close to 20 ohms. 13 ohms for some fuel gauges.
Variable resistors show accuracy throughout the scale.
I hear that built in protection causes the RTE piece to fault out. That may be the on/off condition you see.