Fuel Guage Wont Read on Drivers Tank

[86 Silver]

Last year I installed a new fuel tank selector valve as the old one ceased to function. with the new valve came a new pigtail. also i replaced the switch in the cab. anyway, about a month ago, my drivers side tank ceased to read it simply went to way beyone full (about the 3 o'clock position). i thought sending unit so i replaced that on that particular tank. no change. i checked my splicing on the new pigtail and all was solid. however, in messing around down there it would start to work, and then when i shoved all the wires up and way, then it stopped. i couldnt get it to start working again no matter what wires i messed with.

today i followed the wire from the fuel sending unit to the plug on the selector valve and up to the engine. no breaks in the wires. the passenger side tank has never stopped working or reading correctly. the selector valve functions as it should.

someone help!!! why wont my guage read on that drivers side tank!?! it works right on the other tank and as i said all wires are good and connections are solid.

i ran my test light on the pigtail when it was unplugged. the third wire would light up wether i was on switched to the driver or passenger tank. however, when i put the tester in that port it would make the passenger tank guage move, but when i would switch to the driver tank the guage would not move. i have no idea what the hell that means, but i am just sharing what i found.

HELP!!!!

[hatzie]

03:00 means the sender circuit is open. Your splice is crimped on the insulation not the copper. Moving the wire caused it to make intermittent contact.
Or...
Frame to sender ground wire is missing or fubar.

[86 Silver]

if the ground is bad it would be way past empty (ground is rock solid and clean). i will look at the wiring crimps.

[wilkin250r]

It's a bad connection somewhere. Your gas gauge is really just reading a resistance, with Zero resistance (basically a direct path to ground) would read as an empty tank. Any higher resistance reads as full, with an infinitely high resistance (like a broken wire or bad connection) reading way past full. The gauge would be expecting a maximum of 90 ohms, and an open circuit is like a million ohms, so it's pegged as full full full.

Very rarely would you ever have a situation that would create a better path to ground, the only situations I can think are when the insulation gets burnt or rubbed off the wire, or if it's installed incorrectly. So rarely would you ever read past empty.

Most often, you create HIGHER resistance, like a corroded ground connection, bad connection, or broken wire. Like I said, the gauge is expecting 90 ohms, and you've got a million ohms.

I'm sorry I can't help you find WHERE it is, but I know without a doubt WHAT it is. You've got a bad connection somewhere.

[hatzie]

Originally Posted by 86 Silver

if the ground is bad it would be way past empty (ground is rock solid and clean). i will look at the wiring crimps.

I have tested 73-89 GM gauges on the bench using a 50 Amp variable DC supply & General Radio 1432B Calibrated resistance decade box for a simulated sender.

GM square fuel gauges both small and large regardless of year...
Good gauges work as follows.

Switch the decade box to 0 Ohms (simulating an empty tank or sender wire with a dead short to ground) will make the gauge read right at E. They don't read lower unless the gauge calibration resistor is screwed up.

Switch the decade box to 88-93 Ohms and a good gauge will read right at F.

Unplug the decade box from the sender stud and the gauge reads at 3:00. This is the same as removing the tank/sender ground or unplugging the sender wire.

Have you tested the gauge switch portion of the tank selector valve with a meter?
The tank valve chooses which sender feeds the gauge using an internal SPDT switch. Check that gauge switch for continuity when the selector valve is switched to the offending tank. See attached wiring diagram for where to attach the meter probes.

NAPA carries the WeatherPak tower terminals... Your local GM dealer is another possible source. I keep a handful bare WeatherPak and various bare MetriPak terminals with my meter probes. They keep various sensor connectors from getting botched by directly jamming in meter probes. My wallet has thanked me several times.

[86 Silver]

SOLVED. that wiring diagram was a HUGE help thank you! the light pink wire reporting from the guage to the new fuel sending unit did not have a solid connection. i cut out all butt connections on that pigtail, restripped wires and soldered everything back together. works perfectly now...or so it seems, guess ill have to fill the tank up to know for sure, but i know its empty and its reading as such so thats a step in the right direction i hope.