Fuel Gauge jumping

[lamberthkp]

I have not had a fuel gauge since starting this project but figured the wire was touching metal somewhere since it was pegged out on full.

It's a 1978 K5 and I found the pink fuel wire hanging from the frame after Trailering it back from GA in Nov.

I finally got most of the other electrical issues that all the critters have chewed to pieces fixed and I crawled under the truck and searched for the start of the pink wire to fix the fuel gauge.
I located it up above the starter wires trying to hide. I cleaned everything up reconnected everything and when I turned the key it showed about 3/4 tank.
I thought I was good..... but it starts jumping while moving or just at idle. It will go from 3/4 to 1/2 to almost full at times....... The grounds seem to be great, iv already checked that.

Any ideas?

Sending unit going bad? Or is bad?

 

[fastazz81]

Probably the printed circuit board on the back of the cluster

 

[dyeager535]

How fast are we talking? Like instantaneous? All the below is based on the assumption the needle is rapidly jumping between readings.

I have never seen an explanation of how the fuel gauge "damping" actually works. Somewhere in the circuit there has to be something slowing down how fast the needle changes. As far as I can tell it's not at the sending unit...resistance changes immediately as the float moves. I can only surmise however that is accomplished, is done in the gauge.

Based on what I recall when filling the truck up, with key off, gauge reading low (correctly), fill up, then turn key to run, the gauge will slowly climb to full. No quick movement of the needle (unlike oil pressure or voltmeter which jumps to the proper position on the gauge immediately) which leads me to think that even if the power/ground to the circuit is flakey it still shouldn't move rapidly. On top of that, the fuel gauge almost certainly shares the power/ground with at least one other gauge, if they all act properly, another indicator its not the IP connector. Although it is possible the connection on the back of the gauge to the flex circuit is iffy.

If you have another gauge, I'd swap it just for the heck of it. About as difficult as pulling the cluster to check the flex circuit connector (IP) on the back. The gauges are often not the issue, but the way the fuel gauge seems to operate, does not leave much room for other causes of the issue.

When the truck isn't running, but key is in run, does the gauge stay at the same level? If it does, hard to move these trucks much, but if someone can get the rear suspension to move a bit to try and slosh the fuel, and it's the gauge, that should do it.

 

[lamberthkp]

I will take a closer look when I make it back to the shop in a little while. I know yesterday, when I got in the K5 the needle was at rest and after turning the key it made a small jump.... About an 1/8. I cranked the truck it did steadily climbed... maybe another 1/4 inch on the gauge.

This is when it started bouncing everywhere about every 1-2 seconds (truck has not moved). Sometimes it seemed that it was going to stabilize but then right back. I can shack the vehicle and it gets even worse. I don't believe it is just fuel slosh because at times just setting it might jump from 1/2 to nearly full.

I've checked all wiring on the frame so today's agenda is to swap an extra gauge cluster and see if anything changes.

 

[campfire]

I have a related issue. The fuel gauge in my dual-tank '83 reads instantaneously. Rock the truck with your hand and the gauge will precisely track the sloshing in the tank. This is true for both tanks. Tank valve and both sending units are new. I've had these results with two different gauges. There's not much left in the circuit, but the gremlins persist.

IIRC, this gauge is two magnetic coils pushing against each other. One with 12V and the other with whatever comes through the potentiometer in the tank. We "solved" the problem by installing a capacitor across the sending unit wires to integrate the voltage peaks and provide a steadier input value. However, we found out that this only dampens the input signal, not the gauge behavior. Now my gauge will instantly drop, but rises slowly over a few seconds. I'm not sure why, as I would have expected the cap to charge and discharge equally slowly. But it's close enough to normal that I no longer fear breaking the needle.

I'm curious to see what you smart people think.

 

[lamberthkp]

Probably the printed circuit board on the back of the cluster

There's no telling how long it had been driven before I got it with the fuel gauge wire touching the frame. Definitely testing this today, hope that's it.

 

[diesel4me]

CHEVYTRUCKS.ORG - Troubleshooting Your Gas Gauge