Fuel Gauge Mystery Continues...

[RickTheEngineer]

Howdy all,

I have been dealing with a fuel gauge problem for some time now and I have a bunch of details to share, so please hear me out.

The problem:

- Fuel gauge floats around, sometimes reading correctly, sometimes not, never pegged full or empty

- Took cluster apart and removed all gauges to read the resistance directly from the connectors behind the fuel gauge. It reads close to the correct value when the vehicle is OFF, but becomes about half as many OHMS when vehicle is ON. Then, drops by another quarter or so when headlights are on

My findings:

- When measured DIRECTLY from the tank sender, it goes from 0 OHMS to 90 OHMS as it should when filling up

- Cleaned all grounds, particularly the one for the sender

- Removed the cluster entirely and measured right from the pigtail, results are similar

- Got under the vehicle and CAREFULLY traced the pink sending wire all the way down the frame, up to the engine bay, through the connector in the fire wall, and to the cluster pigtail. All was in perfect shape, no damage whatsoever

-Once again confirmed with vehicle off, the resistance value was very close to what was on the tank sender directly

-Getting solid 12V from the 12V circuit for the gauge itself

Results:

- Definitely not the cluster traces, problem stems right from the pigtail

- Sending wire is in mint condition without any shorts there



Anyone know what could be causing this?? Thank you all!

 

[porkysplace]

Corrosion inside the wires.

 

[RickTheEngineer]

I thought about that, but shouldn't the resistance wire be completely independent of all voltage sources? If it reads correctly with the vehicle off, it should also read correctly with it on right?

 

[Tinstar]

Maybe it’s just enough corrosion to cause the problem.

I’ve replaced many a wire that looked perfect and for the most part handled the load requirement.
Once replaced, peeled back insulation on old wire and entire length had corrosion.
The new wire solved issues.

Just a thought

 

[RickTheEngineer]

Well here is an update. The sending wire from the pigtail from behind the gauge cluster more or less is right on when everything is off. However I realized that when you measure voltage, it actually receives 1V from somewhere and that's what's messing it up. Still no idea how though... I peeled back every piece of conduit left and it seems perfect all around. At least I know it's not the wire being corroded, but rather receiving voltage somewhere... Hmmm....

 

[cucvrus]

Did you remove and clean the ground wire from the sending unit to the left rear frame rail? I don't see any mention of cleaning the ground.

 

[RickTheEngineer]

Did you remove and clean the ground wire from the sending unit to the left rear frame rail? I don't see any mention of cleaning the ground.

Yes I did, I said it in the original post. All those grounds are nice and shiny

 

[RickTheEngineer]

Well the main firewall terminal block has a bunch of shmoo in there. I am breaking out the contact cleaner and seeing where I get with it

 

[dependable]

That shmoo is dielectric grease. If you clean it all up, might want to consider adding new on re assembly. I have seen contact problems in that area once though, it was on a high mileage civi K-30 that had probably already been taken apart there.

The grease is there to keep moisture out of harness connection.

 

[RickTheEngineer]

I should have been more descriptive. The shmoo was mostly dirt and debris along with the grease. I applied a generous amount afterwards. Still no change. When I measured for resistance at the terminal on the firewall directly, it's stayed consistent despite keying on the ignition and whatnot. What I can say though is that I have isolated the problem to somewhere behind the fuse block. Is it just 2 bolts to get it off?

 

[dependable]

Two bolts and the two plastic hooks you have to pry back a little to separate.

Not much room to work or see even then, unless you pull the harness. Good luck, got to say I have not ever seen any problems there except in trucks that had a serious short out or something. Let us know what you find.

 

[RickTheEngineer]

Before I start really ripping it all apart. I want to make sure I'm just not missing something. So on the gauge cluster there are 3 terminals for the fuel gauge. The bottom one is a ground. And the top left one is 12V source. And if I am not mistaken, the top right terminal is solely the resistance value sent from the sender right? So in an instance where the tank is full, and has a 90 OHM value. This means that if you read from the top right terminal to the bottom terminal it should read 90, regardless if the key is on etc right? Like, that top right signal wire should never become energized right? If I am correct in my hunch, then there definitely is a problem somewhere...

 

[dependable]

The place where the cluster harness plugs into the plastic circuit sheet for the instrument panel sometimes requires cleaning, especially if there were ever mice living in there. That seems more likely place for problem than at the fuse box.

 

[RickTheEngineer]

I know what you mean, but keep in mind, the problem is that I'm having the same mis readings even from the pigtail itself that goes directly behind the cluster and supplies the cluster all the information that displays on the gauges... So, the problem stems deeper than that, but definitely after the terminal block on the firewall

 

[RickTheEngineer]

Well everyone, I have another update. So I was wrong about the fuse block as was predicted by "dependable". However I saw some crazy stuff in there. Aluminum oxide grains everywhere. But not enough to create a short. Instead what I discovered was the black wire on what I think was pin 18 is the ground for the fuel gauge on the cluster as well as all the bulbs. I came to discover this was getting energized by .5V. I cut back the insulation on the harnedh and found it travels right behind the climate control, and from the factory, it gets lost in a splice between 4 other black wires, on what I am assuming you also be grounds. My plan tomorrow is to cut pin 18's wire and create my own ground for it, so it will be stable and noise free as to not interfere with the fuel gauge. If this works then it may serve as a "go-to" solution if anyone else has a similar issue. As for why it's getting energized? Well, I am not sure yet, but I will say that aside from the fuel gauge, everything else seems to work great and haven't noticed this having an effect on any other electronics in the vehicle. Not sure how far I should end up investigating right now.

PS- as a reminder, this only gets energized when the headlights are on. I traced every single wire in the headlight assembly and just couldn't find any problems there either. Even all the factory conduit was in place before I took it all off! Impressive!

 

[dependable]

A bad ground somewhere is usually the problem when things change when you turn on the lights.

 

[cucvrus]

A bad ground somewhere is usually the problem when things change when you turn on the lights.

That could be any ground connected to the harness that is robbing/pulling ground thru another source. Bad grounds will drive you crazy. Once found it will never be the same ground location on the next one. Good Luck. You seem to be sniffing it out. Most would have given up.

 

[DeadParrot]

Bad/poor headlight ground can cause voltage to feed back through the non-energized headlight filament. Same for the turn signals.

Also, the fuel gauge itself has a damper gizmo in it. If that fails, the gauge will float around fairly quickly. My 75 C10 fuel gauge started the all over the place thing. Replaced with Pull-a-Part find and fixed. Apparently, there are enough momentary variations in the signal the gauge sees that the damper thing is needed to average the reading. Don't know if the damper gizmo is a bit of magic goo that slows the needle or some kind of electronic thing that slows the gauge response.

 

[dependable]

From what I've seen, the "floating" fuel gauge is usually intermittent ground condition. Most likely the ground on top of the fuel tank where it is hard to get at.

Not sure about that damper gizmo, never encountered one...goo or electronic.

 

[RickTheEngineer]

To anyone reading this post, I am going to be giving my final update before I will have to take a new course of action. I cut the ground pin on the cluster pigtail for the fuel gauge and measured it directly as to not get any interference. Turns out it was also getting energized from the pigtail end. I shook the connector and tried to get out any grains of aluminum oxide I could, of which i did not see any come out. Upon measuring again, almost magically, it was fixed. So not I have a perfect ground for the sender (progress!). But here is where I hit my final road block. The sensing wire (pink) for the sender is getting 1.9 mV when the headlights turn on. It may not sound like much, but it is enough to completely throw of the resistance value that it should be at. Works perfect when the truck is off though. I cut it from the pigtail just to check if it was a similar situation as the ground. Still no change, it is in fact being energized from somewhere else.

Now my plan is to start cutting that pink wire further back until the ghost voltage has been eliminated, then I will run my own pink wire all the way up to the pigtail, solder, and heat shrink it all back together so it all looks and acts factory. It is coming from somewhere clearly, I just have no clue what it could be.

PS- I would also like to add, that I have cleaned every single ground, from sender to headlights and everything in between. Keep in mind, that it is the pink sender wire getting a tiny amount of voltage from somewhere. It would be noise from other wires where there is a fault, or it would be some very mysterious short that has very low conductivity. Either way, it has been proven to not be a bad ground, but rather potentially a bad wire. I plan to one day replace the bulk head connector with a deutch DRB series connector (we use these every day at work). That way the harness will forever be corrosion free.