Temp & fuel gauges - what am I overlooking?

[67Dart273]

If it came with resistors, what are they? Look at the post 3, bottom left, in the yellow. Those are the test resistors for low, 1/2, and full scale. All you do is connect them from the sender terminal of the gauge to ground. you of course must have at least 12.6 (fully charged battery) connected to the cluster and a cluster ground point.

That IS the definitive test, AKA, "end to end." Power to the cluster, all assembled, working IVR, and test resistors. if that does not give correct results, then WITH THAT connected in place, check the input voltage to the IVR by pobing the trace up to the IVR, maybe pull the IVR out a little and probe right on the IVR power terminal directly. Then make sure it is grounded. Same deal. Check voltage at the battery NEG, and voltage at the IVR ground terminal. Should be zero. it is difficult to check the IVR output for accuracy, unless RTE give you destructions. It is difficult to check the gauges, but I bet Redfish has some ideas. If you have a known good gauge, you can "rig" it in series with a questionable one, and use 1 or 2 D cells and the two gauges should rise to the same reading.

https://www.google.com/url?sa=t&source=web&rct=j&opi=89978449&url=https://www.forabodiesonly.com/mopar/threads/bench-testing-a-gauge-cluster.379879/&ved=2ahUKEwjQ6Z742vSOAxV8N0QIHZhzBYAQh-wKegQIFBAD&usg=AOvVaw1wPCCUIXGVPVpClRryNDGb

67dart273, gauge tester - Google Search

 

[67Dart273]

Just so you know, I've had a house fire. What I had for spare gauges, a home brew test box, and other items are buried somewhere in the storage unit. There is no chance of digging those out for a few demo photos. I don't even know where the cluster for the car is at present

Otherwise, I'd be happy to dig some of it out and snap some photos. There used to be some on here, but if they are still online, I cannot find them

 

[Super-Cuda]

It's on the bench and I powered it up. The light started flashing on the IVR, there was fluctuating voltage at the 2nd and 3rd studs, and at the 1st and 2nd pins past 12 o'clock. So there is continuity from power to IVR, from ground to IVR, from IVR to both gauges, and from both gauges to the connector pins.

The IVR came with a couple of resistors and sort-of-instructions on using them to test the gauges.

FWIW I saw something today saying to check the resistance across the gauge and it should be 15 Ohms. I don't know if that's right or not but figured it can't hurt so I put the DVM in Ohms mode and touched the probes to the studs. Both the fuel and temp gauges read open.

I've got a few things to do here but I will check back often to see if there are more ideas.

Ok.

lets keep going step by step.

First, put your meter in dc volts and measure the voltage using the cluster chassis as ground and connect the other meter probe to the second and third gauge posts as indicated in the pic of the cluster board in post 10.

You should get 5 volts.
Check back with your reading when you get a minute.

Edit. My reasoning here is to first establish that you have the proper voltage to the gauges. You have already established via your first post that the gauge pointers are deflecting when the sender wire is grounded.

 

[RedFish]

There were 2 type gauges. Majority measure 20 to 20.5 ohms across. That's approx' 13.5 inches of ni-chrome that measures 1.5 ohm per inch at 68 degrees. Later fuel gauge were changed to a 1.0 ohm per in wire, thus 13 to 13.5 ohm across. Less accurate but lives longer.

 

[Uncle Bob]

Just so you know, I've had a house fire. What I had for spare gauges, a home brew test box, and other items are buried somewhere in the storage unit. There is no chance of digging those out for a few demo photos. I don't even know where the cluster for the car is at present

Otherwise, I'd be happy to dig some of it out and snap some photos. There used to be some on here, but if they are still online, I cannot find them

Understood. Sorry for your bad luck.

If it came with resistors, what are they? Look at the post 3, bottom left, in the yellow. Those are the test resistors for low, 1/2, and full scale. All you do is connect them from the sender terminal of the gauge to ground. you of course must have at least 12.6 (fully charged battery) connected to the cluster and a cluster ground point.

The paperwork doesn't say what the rating is, just that the resistor limits it to full scale. It has brown, brown, black and gold stripes on it. According to the converter at digikey.com that means it's 11 Ohm 5% tolerance, and per post 3, 10 Ohms gives you full scale high. With it hooked in, the fuel gauge went right to full, not past, and the temp gauge went to almost all the way hot, but not quite. So once again, the test passed.

It's like there's nothing wrong, except that the gauges hate me LOL.

 

[Uncle Bob]

Ok.

lets keep going step by step.

First, put your meter in dc volts and measure the voltage using the cluster chassis as ground and connect the other meter probe to the second and third gauge posts as indicated in the pic of the cluster board in post 10.

You should get 5 volts.
Check back with your reading when you get a minute.

OK I did that. I don't get 5 solid volts, I get a range of voltages. I get it at the pins and at the gauge studs.

I kind of expected the RTE IVR to output a steady voltage, but it doesn't. As a black box, it behaves just like the factory part.

 

[Uncle Bob]

There were 2 type gauges. Majority measure 20 to 20.5 ohms across. That's approx' 13.5 inches of ni-chrome that measures 1.5 ohm per inch at 68 degrees. Later fuel gauge were changed to a 1.0 ohm per in wire, thus 13 to 13.5 ohm across. Less accurate but lives longer.

OK, to be clear, I can check that without power to the gauge by putting the DVM in Ohms mode and reading across the two studs, right? On this cluster, doing that shows the gauges are open.

 

[Uncle Bob]

Understood. Sorry for your bad luck.


The paperwork doesn't say what the rating is, just that the resistor limits it to full scale. It has brown, brown, black and gold stripes on it. According to the converter at digikey.com that means it's 11 Ohm 5% tolerance, and per post 3, 10 Ohms gives you full scale high. With it hooked in, the fuel gauge went right to full, not past, and the temp gauge went to almost all the way hot, but not quite. So once again, the test passed.

It's like there's nothing wrong, except that the gauges hate me LOL.

Also, I have two resistors because I bought two IVRs. They are both the same.

 

[RedFish]

Have you tried putting 10 gal. gas in? Do we know we have hot water? The needles need a bit of reason to move a bit.

 

[Super-Cuda]

OK I did that. I don't get 5 solid volts, I get a range of voltages. I get it at the pins and at the gauge studs.

I kind of expected the RTE IVR to output a steady voltage, but it doesn't. As a black box, it behaves just like the factory part.

I think we have found your problem. The whole point of the IVR or, the fuel gauge mechanical regulator in the rallye dash fuel gauge is to drop the 12 volts from the battery to 5 volts for the gauges. Believe it or not, this is a throwback to the original 6 volt system that used to power gauges in older cars.

I think you have either a defective IVR or, its incorrect for a 76 gauge cluster.

 

[Uncle Bob]

Have you tried putting 10 gal. gas in? Do we know we have hot water? The needles need a bit of reason to move a bit.

Oh it's got gas in it, I top it off every time I take the car out. Last time was Sunday.

The temp gauge stays on stone cold when the external gauge shows there is heat in the engine.

 

[Super-Cuda]

Also, I have two resistors because I bought two IVRs. They are both the same.

Did you buy the IRV 3 or the IRV 4 ?

 

[Uncle Bob]

I think we have found your problem. The whole point of the IVR or, the fuel gauge mechanical regulator in the rallye dash fuel gauge is to drop the 12 volts from the battery to 5 volts for the gauges. Believe it or not, this is a throwback to the original 6 volt system that used to power gauges in older cars.

I think you have either a defective IVR or, its incorrect for a 76 gauge cluster.

I have been led to believe that the voltage averages out to 5 because of the IVR so the DVM should show a range of voltage. Which it's doing.

The IVR is brand spanking new. It's an RTE digital job, and the light is flashing so it thinks it's working.

Did you buy the IRV 3 or the IRV 4 ?

They are both IVR4. One is for a Duster, the other is for a road runner.

 

[Super-Cuda]

I have been led to believe that the voltage averages out to 5 because of the IVR so the DVM should show a range of voltage. Which it's doing.

The IVR is brand spanking new. It's an RTE digital job, and the light is flashing so it thinks it's working.


They are both IVR4. One is for a Duster, the other is for a road runner.

Ok.

Here' s what I think , maybe 67Dart273 can help us out here as well.

Both the fuel gauge and the temp gauge need 5 volts to operate properly.

The fuel tank sending unit and the temp sending unit operate by changing the resistance in the sending units wiring back to ground.

As the float rises, the resistance of the sending unit drops , allowing more current to travel through the gauge , deflecting the pointer . As you saw, by shorting the sending unit wire to ground, the gauge pointer went to full deflection indicating full.

The temp gauge works the same, but, uses an electrical component that also offers less resistance as it gets hotter allowing more current to flow and thus causing the gauge pointer to deflect to hot.

The constant here has to be the 5 volt supply to the gauge. Without it there is no operation of the gauges which is exactly where you find yourself.

 

[Super-Cuda]

Ok.

Here' s what I think , maybe 67Dart273 can help us out here as well.

Both the fuel gauge and the temp gauge need 5 volts to operate properly.

The fuel tank sending unit and the temp sending unit operate by changing the resistance in the sending units wiring back to ground.

As the float rises, the resistance of the sending unit drops , allowing more current to travel through the gauge , deflecting the pointer . As you saw, by shorting the sending unit wire to ground, the gauge pointer went to full deflection indicating full.

The temp gauge works the same, but, uses an electrical component that also offers less resistance as it gets hotter allowing more current to flow and thus causing the gauge pointer to deflect to hot.

The constant here has to be the 5 volt supply to the gauge. Without it there is no operation of the gauges which is exactly where you find yourself.

A couple of questions.

1. Have the gauges worked since you changed out the cluster to the 76 one presently in your car?

2. What was wrong with the original cluster?

3. Do you still have the old cluster?

 

[RedFish]

The wire count and their locations in the round harness connector did change from time to time.

 

[67Dart273]

Understood. Sorry for your bad luck.


The paperwork doesn't say what the rating is, just that the resistor limits it to full scale. It has brown, brown, black and gold stripes on it. According to the converter at digikey.com that means it's 11 Ohm 5% tolerance, and per post 3, 10 Ohms gives you full scale high. With it hooked in, the fuel gauge went right to full, not past, and the temp gauge went to almost all the way hot, but not quite. So once again, the test passed.

It's like there's nothing wrong, except that the gauges hate me LOL.

Well if the cluster/ gauges pass the test resistor, then it HAS to be either in the body wiring or the sending units, if you tested it "in the car." You can eliminate the body wiring by putting the resistor right at the sender connector to ground. Should get the same result.

If you tested the cluster out of the car, you have the additional possibility of a cluster grounding problem and a possible poor connection feeding 12V from the key to the cluster

 

[harrisonm]

I have tested these gauges before with two AA batteries. The IVR puts out a pulsing current of about 5 volts. I tested them by placing two AA batteries end to end for 3 volts and then running two wires (one from the + end and one from the - end of the batteries) to the back of the instrument. 3 volts should move a gauge that needs 5 volts to go all the way up to about 2/3s of the way. I assume that this test method was safe. I have done it many times with no problem.

 

[RedFish]

I have tested these gauges before with two AA batteries. The IVR puts out a pulsing current of about 5 volts. I tested them by placing two AA batteries end to end for 3 volts and then running two wires (one from the + end and one from the - end of the batteries) to the back of the instrument. 3 volts should move a gauge that needs 5 volts to go all the way up to about 2/3s of the way. I assume that this test method was safe. I have done it many times with no problem.

That has to be the simplest and safest way to test a stand alone gauge. Still mounted attached to a circuit board wouldn't be proper.

 

[Uncle Bob]

A couple of questions.

1. Have the gauges worked since you changed out the cluster to the 76 one presently in your car?

2. What was wrong with the original cluster?

3. Do you still have the old cluster?

The car came with this cluster and the gauges have never worked.

 

[Uncle Bob]

Well if the cluster/ gauges pass the test resistor, then it HAS to be either in the body wiring or the sending units, if you tested it "in the car." You can eliminate the body wiring by putting the resistor right at the sender connector to ground. Should get the same result.

If you tested the cluster out of the car, you have the additional possibility of a cluster grounding problem and a possible poor connection feeding 12V from the key to the cluster

But the body wiring must be good because grounding the sender wire causes the gauge to go toward full. Ditto grounding and 12v to the cluster. If there was a problem in the wiring, ground, or 12v feed the gauge wouldn't work at all.

It's all confusing. That's why I tried everything I knew to try including replacing a good IVR to try to get them working before posting for help.

I have tested these gauges before with two AA batteries. The IVR puts out a pulsing current of about 5 volts. I tested them by placing two AA batteries end to end for 3 volts and then running two wires (one from the + end and one from the - end of the batteries) to the back of the instrument. 3 volts should move a gauge that needs 5 volts to go all the way up to about 2/3s of the way. I assume that this test method was safe. I have done it many times with no problem.

I'm going to give this a try.

 

[Super-Cuda]

But the body wiring must be good because grounding the sender wire causes the gauge to go toward full. Ditto grounding and 12v to the cluster. If there was a problem in the wiring, ground, or 12v feed the gauge wouldn't work at all.

It's all confusing. That's why I tried everything I knew to try including replacing a good IVR to try to get them working before posting for help.


I'm going to give this a try.

Uncle Bob,

I did a little more research on the RTE IVR last night and found something very interesting. That unit is designed to shut down if it detects a short which seems to be a symptom of what you are experiencing.

As this cluster has never worked for you, Here's what I would do in this case.

1. Completely disassemble the cluster and remove the circuit board.

2. Examine it on both sides for damage/repairs.

3. Clean every point of electric contact to shiny copper finish.

4. Replace the bulbs. They can short internally.

Reassemble, power up and test.

Remember , you have to have 5 volts at the output of the IVR , measured between the cluster housing ground and the 2nd and third gauge studs in the pic looking left to right.

No 5 volts, no gauges .

The measurements with the resisters suggested are a great way to check accuracy of the gauge but you have already determined that they work by temporarily shorting the sending wire to ground.

I have done this many times , built my own IVR's from scratch and I think you are really close to solving your problem.

Keep us posted.

 

[Super-Cuda]

The wire count and their locations in the round harness connector did change from time to time.

Great point Red Fish but Uncle Bob has grounded the sensor wire for the fuel gauge and the Temp gauge and got full deflection. That would prove the harness is wired correctly.

What I don't understand is that in order for that to happen, there must be 5 volts at the gauge lugs in the circuit board.

This seems to be a loss of ground or a short once the cluster is installed.

 

[Uncle Bob]

That unit is designed to shut down if it detects a short which seems to be a symptom of what you are experiencing.

The RTE IVR has a heartbeat light. If it shuts down, the light goes off.

Also don't forget I had the same problem with a factory IVR.

As this cluster has never worked for you, Here's what I would do in this case.

Will do all the things you listed plus open the cluster up and have a look at the gauges as much as I can. Those are sort of the only thing I have little faith in because all the tests involving wiring and circuit board traces have passed.

This seems to be a loss of ground or a short once the cluster is installed

But, the behavior was the same when the cluster was installed, so I don't think that is it either.

I'm leaning toward something that happened in the past killed both gauges. Maybe I will be able to tell once I've opened up the cluster and can see the mechanical part of them. I would just order an new OER fuel gauge if they cost half or less of what they do and I don't have an old one around here. (Don't care about the temp gauge TBH it's just part of the symptoms).

 

[Super-Cuda]

The RTE IVR has a heartbeat light. If it shuts down, the light goes off.

Also don't forget I had the same problem with a factory IVR.



Will do all the things you listed plus open the cluster up and have a look at the gauges as much as I can. Those are sort of the only thing I have little faith in because all the tests involving wiring and circuit board traces have passed.


But, the behavior was the same when the cluster was installed, so I don't think that is it either.

I'm leaning toward something that happened in the past killed both gauges. Maybe I will be able to tell once I've opened up the cluster and can see the mechanical part of them. I would just order an new OER fuel gauge if they cost half or less of what they do and I don't have an old one around here. (Don't care about the temp gauge TBH it's just part of the symptoms).

Easy to check the gauges.

Just connect one 12 volt lead to one gauge terminal post and momentarily touch the other lead off the other post.

Gauge pointer will deflect immediately if gauge is good.

Caution!!

A momentary touch is all you need.

Do not leave wire connected or you will burn out the gauge!!