[JLMounce]

Recently I started to get some wonky behavior from the gauge cluster in my 1969 Firebird. Erratic fuel level, or non-erratic, but reading lower than actual fuel level. Additionally the lights were dimmer than typical and exhibited a willingness to get brighter if the car hit a bump or otherwise jerked around.

This seemed obvious as a ground issue back there. I removed the dash bezel and did notice that the screw that attaches the gauge housing to the ground straps on the '69 dash bezel could no longer adequately tighten the ground strap to the gauge housing.

I cleaned up the ground straps, added a new fastener and an auxiliary ground that is now mounted to the dash, as well as the OEM ground straps. Upon reinstallation of the dash bezel, the lights are working great now, but the fuel gauge is still reading between 1/8 to 1/4 low. It's not pegged full or empty and does swing fully, so I know the gauge is working, it's just inaccurate.

I knew that I had around 1/2 a tank, or there about. Testing the sender itself, it's showing 35ohms. Near where I was expecting actual fuel level, so I'm fairly certain I don't have a sender issue (2.5 year old Tanks Inc floatless sender).

I then measured at the dash connector from the intermediate harness and I'm seeing 31ohms at that connector on the intermediate harness. This and the rear lamp harness are the only wiring original to the car. Everything else, including the dash harness is about 7 years old.

Am I wrong in thinking my issue here is likely that the old wiring has simply oxidized and is adding resistance in line between the tank and the gauge?

Visually the printed circuit board is in good condition without any obvious breaks in the circuits, but it is of course 56 years old.

I can't get under the dash in a way to check ohms are the gauge itself to know for sure if the circuit board is bad.

Before I load up some cash in the parts canon for a printed circuit, intermediate and rear lamp harness, is there anything else I should be checking?

[Shiny]

If you read less resistance at the dash through the harness (31 ohm) than directly at the float/sensor (35 ohm) it seems unlikely there's an issue in the harness.

Maybe I'm missing something but seems the measurement method or the stability of the sending unit (ie fuel level) are contributing to this difference?

If the harness had degraded, wouldn't you expect to see higher resistance at the dash than at the sensor?

If the grounds on the circuit board were tired, what about the hot connection to the fuel gauge? How is that connection made physically?

Can you verify the gauge before assuming the worst?

If you can disconnect the sensor/harness at the gauge, can you use a resistor to check the gauge at the expected high and low resistance values (is it 0 to 90 ohms?)? As I understand, the gauge is measuring current so a resistor between the gauge and ground should simulate the gauge, correct? If it's easier to reach at the sending unit you can do the same there.

If you're on your back reaching up to the gauge, I don't envy you. Not fun.

Mike

[JLMounce]

Yeah, it's the ohm leakage that is weird. I should expect to be in the opposite situation if the harness was corroded.

I suppose I could have issues with the wire sheathing and I could have a minor short?

The fuel gauge is functioning as expected, it's just between an 1/8th to 1/4 tank lower than actual fuel level, which makes sense considering I'm leaking 4-5 ohms of resistance across the circuit.

[Shiny]

That's a good guess at the cause for that discrepancy. Now I get why you are questioning the harness integrity.

Is there any way for you to access both ends of the harness and verify?

I suspect access is a bear, but is there any way to try and directly verify a short to ground before you dive into replacing all the harnesses?

Your meter is reading the difference so it should also be capable of reading the short.

What's the chance of disconnecting at the sending unit? If the sending unit is disconnected and the fuel gauge wiggles at all, this would confirm a short somewhere.

Does power go through the PCB before reaching the fuel gauge? If you find a short, you still need to decide if it's in the board or in a wire/harness...

[JLMounce]

When the sender is disconnected, the gauge pegs full. Most of this is factory type wiring, either original or replaced with reproduction.

Only thing different is the tank and floatless sender that needs it's own dedicated ground.

As is typical for a 1969 Firebird, there is a sender connection in the trunk that is part of the rear lamp harness. it shares a double pin with the license plate light harness and the rear lamp harness grounds to the trunk latch support. From the sender I am using that factory pin and using the same common ground as the lamp harness.

The rear lamp harness connects to the intermediate harness near the trunk divider, then the intermediate harness connects to the dash harness near the parking brake cable. It's then connected to the gauge backing plate via the 11 pin gauge connector to the PCB,

The rear lamp harness is easy to get to, but the intermediate harness requires a bunch of interior disassembly, including up into the dome light. I wouldn't even bother trying to test this as the work required to do that would be the same work required to replace it.

[Shiny]

If there was a resistive short causing you to measure 4 ohms low at the gauge when the tank is half full, the short is a resistance in parallel that brings 35 ohms down to 31 ohms.

1/Rt = 1/R1 + 1/R2, or 1/31 = 1/35 + 1/Rshort

Rshort = 271 ohms

Unfortunately, the resistance of a short like this is still higher than the "full" value of ~90 ohms. That dashes hopes of using the gauge to detect the short by unplugging the sending unit.

What other circuits are shared by the fuse for the gauge? If the gauge was the only circuit on the fuse, you could measure current at the fuse box with and without the sending unit.... but I suspect there will be other loads on that same fuse to make this fruitless.

This is a tough one..

I guess if the short was stable, you could add 5 ohms (or whatever) in series with the sending unit and get your gauge to read correctly. The chances of a "stable" short like this sounds pretty slim though...

You need some smarter help diagnosing this one!

Mike

[JLMounce]

Where I am measuring the ohm drop, is pre fuse box. I'm measuring it at the pinned connector from the intermediate harness to the dash harness, before the fuse. So, whatever is happening, is certainly happening between the pin in the lamp harness and the intermediate harness.

[Bill Hanlon]

Originally Posted by Shiny

I guess if the short was stable, you could add 5 ohms (or whatever) in series with the sending unit and get your gauge to read correctly.

Wouldn't work across the whole sweep of the gauge because the 5 ohms (or whatever) is a fixed amount while the float resistance varies. It would only fix the reading at one point on the gauge.

If the point that it "fixed" were at or near empty maybe that would be good enough. Worrying about the full reading usually isn't as nerve wracking as worrying about do I have enough gas to get to the next gas station.

[JLMounce]

Originally Posted by Bill Hanlon

Wouldn't work across the whole sweep of the gauge because the 5 ohms (or whatever) is a fixed amount while the float resistance varies. It would only fix the reading at one point on the gauge.

If the point that it "fixed" were at or near empty maybe that would be good enough. Worrying about the full reading usually isn't as nerve wracking as worrying about do I have enough gas to get to the next gas station.

Yeah that makes sense. My current situation is the best possible without accuracy. My gauge will show empty while I still have fuel to consume. I'm not in danger of running out as it is. It's annoying though for sure.

[Shiny]

Originally Posted by JLMounce

Where I am measuring the ohm drop, is pre fuse box. I'm measuring it at the pinned connector from the intermediate harness to the dash harness, before the fuse. So, whatever is happening, is certainly happening between the pin in the lamp harness and the intermediate harness.

Would it be worth it to load the gauge at this connection? I know the sender measures too low here but you could verify the gauge calibration after you fixed the ground and put it all back together. There's always a chance the issue is up there. Honestly, you said the gauge reads 1/8 to 1/4 low. That is 11 to 22 ohms off if the reading is directly proportional. This is inconsistent with the 4 ohm error you found. Maybe there's more to it?

A dead short should be empty and full should happen with a load of around 90 ohms, right? A few resistors in between wouldn't hurt and if it all tracks, you will at least confirm your fears about the wiring.

Mike

[JLMounce]

Originally Posted by Shiny

Would it be worth it to load the gauge at this connection? I know the sender measures too low here but you could verify the gauge calibration after you fixed the ground and put it all back together. There's always a chance the issue is up there. Honestly, you said the gauge reads 1/8 to 1/4 low. That is 11 to 22 ohms off if the reading is directly proportional. This is inconsistent with the 4 ohm error you found. Maybe there's more to it?

A dead short should be empty and full should happen with a load of around 90 ohms, right? A few resistors in between wouldn't hurt and if it all tracks, you will at least confirm your fears about the wiring.

Mike

My old man mentioned doing the same thing. Get a cheap gauge to test at that location and see what it's reading comparatively.

On a 0-90 ohm gauge, each 1/4 tank on a linear scale has 22.5 ohms of range, so yeah being off an 1/8 of a tank would be more like 11 ohms of drop. I'm certainly not measuring that much leakage. But again, because of space constraints, I can't really measure this without taking the dash apart...again.

[Shiny]

I probably have some low-value resistors..let me know if you want them and I'll look. If so, I can drop them in the mail. One lead to ground, the other to the pin on your connector and you can know if the gauge is right.

[JLMounce]

I have some resistors laying around, no need to send anything, but I appreciate the offer.