Questions about gas gauge issues

[SCOTI]

Hey guys, I'm trying to sort through my gas gauge issue & I need some info. RECAP: My gauge used to read slightly past full when the tank was filled, but it went down as the tank was emptied. Now, the gauge needle sits @ about the 3-o'clock position all the time (full or empty).

I dropped the tank to check the wiring to the sending unit. Both wires appear solid & had good connections. What is the function of each wire (1-blk & 1-tan)?

For testing, I'm running a hot/12v wire from the batt to my multi-meter. Grounding the meter to the frame shows 12.43v. Thinking the tan wire was the pwr source, I ran the 12v power directly to the terminal the tan wire was connected to. I went to verifiy the voltage on the meter, & it shows zero volts even w/the hot/12v wire connected.

Since I was pretty sure the tan wire was for pwr, I didn't know what to think about the zero volts so I decided to verify the grounds. Grounding the meter to the frame shows 12.43v. When I ground the meter to the black wire of the sending unit, it shows 12+v. I then tried grounding the meter to the terminal the tan wire connects to & it shows 12+v.

One wire appears to be the ground (blk), what is the other (tan) wire? Is it the switched 12v power source or should it be grounded as well?

Please advise quickly because it's getting dark & I need to put this thing back together to drive to work @ 2:30am. Thanks.... Scot

[SCOTI]

Ok... apparently both the black & tan wire are grounds. The black is the chassis ground & the tan is the ground for the gauge & sending unit.

I removed the tank completely & removed the sending unit from the tank. The float arm swings w/o binding. I tried grounding both the chassis ground (blk) & gauge ground (tan) to the frame & the gauge still reads past the full mark.

There has always been a very slight movement of the gauge needle when the ignition switch is turned on so that's led me to think the gauge itself wasn't the problem.

Next, I'm going to disconnect the lead for the gauge ground (it's up by the starter) & see if there's something wrong w/that wire from there to the tank. I'll try running a temp ground to that connection & see if there's a difference.

[NeCrOmAnCeR]

because you were getting a 12 volt reading on the black wire, this indicates a bad ground. The tan wire is the "sense" wire from the guage to the sending unit in the tank, the black wire is the ground for the unit. If properly grounded this wire should not read more than ~0.1 volt. So check the continuity of the black wire, and check to see that the connection point on the frame is clean and secure. Good luck.

[SCOTI]

Originally Posted by NeCrOmAnCeR

because you were getting a 12 volt reading on the black wire, this indicates a bad ground. The tan wire is the "sense" wire from the guage to the sending unit in the tank, the black wire is the ground for the unit. If properly grounded this wire should not read more than ~0.1 volt. So check the continuity of the black wire, and check to see that the connection point on the frame is clean and secure. Good luck.

I doubled checked this this morning when I got home from work.

With 12v from the battery to my test meter +, I grounded the meter to the black wire & it reads 12+v so it should be grounded properly. I hooked the sending unit back up by itself w/a second ground to the frame & the gas gauge still sits in the same position.

I then switched the meter to measure ohms. With the sending unit 'float' in the lowest position it reads 0~10 ohms. When I rotate the float to the highest position, it reads up to 130 ohms. Since the ohms change as I rotate the float throughout the arc of travel, wouldn't that indicate the sending unit is functioning properly?

Is there a way to test the gauge itself?

[Captkaos]

I don't know if this is the case, but is the "arm" wearing any? I have seen cases where it still reads but due to not enough "connection" on the sending unit it reads incorrectly. 0Ohms should be empty and 90Ohms should be full.

[SCOTI]

Originally Posted by Captkaos

I don't know if this is the case, but is the "arm" wearing any? I have seen cases where it still reads but due to not enough "connection" on the sending unit it reads incorrectly. 0Ohms should be empty and 90Ohms should be full.

It's difficult to get an accurrate reading on the ohm meter when you just have things 'pig-tailed' together. I do know that when the arm is @ the bottom of its travel it reads low ohms & as you swing the arm in its arc the ohms increase between 90~130. After talking w/my buddy Roger @ Stoked Out, he figures it's reading +90 because I'm actually raising the arm farther than it would swing when mounted in the tank.

I got a tip on ChevyTalk on how to check the gauge itself w/a couple of 'D' cell batteries. With the batteries hooked up in series (& providing 3v), I ran wires from the + & - to the circuit board & could get the needle to swing from less than empty to more than full. When I switched the leads from the batt, the direction the needle swung went opposite. According to the tip, the needle should remain @ 1/2 tank w/3v of power. IMO the test was inconclusive because I tried the same test on another set of 67-68 gauges & it did the same thing.

I rigged up some jumpers & tried the 67-68 fuel gauge hooked to the 74's wiring. When jumpering it out this way, I can get the gauge to read empty when the sending-unit arm is @ the bottom of the arc. As I swing the arm through the arc, the gauge swings over to full (it does not stay between full or empty.... it's one or the other). If I drop the arm down again, the gauge goes back to empty.

So I hooked up the 74's gauges to the same jumpers & got the same results. If the arm is @ the bottom of the arc, the gauge reads empty. As soon as you swing it up in the arc, the gauge goes to full. This was promising since because I couldn't get it to do anything before.

I hooked the gauge cluster back into the factory wiring circuit board bulk-head & it was reading empty. As soon as I hit the key the needle shot over to the 3-o'clock position again & woudn't reset.

My hair is turning gray.

[1969 C/K CST]

Haven't looked into mine, but at 1/4 tank if you add 5 gal the needle reads maxed out. Eventually it will come down and at E its empty. Falls fast at 1/4 tank. '84 model. Any clues?

[SCOTI]

Originally Posted by 1969 C/K CST

Haven't looked into mine, but at 1/4 tank if you add 5 gal the needle reads maxed out. Eventually it will come down and at E its empty. Falls fast at 1/4 tank. '84 model. Any clues?

When I first got the 74 it did the same thing.

If it was @ a 1/4 tank & you added a couple of gallons it would read full. If you filled the tank it would stay on full for the longest time (long enough to make you think its getting great milage your first trip down the highway).

Then, once the needle starts dropping, if you're not driving/looking for a gas station . . .... you'll be walking/looking for a gas station. That's also about the same time I learned that once the needle got to the empty hash mark, it was to late (well, I learned that after the third time it ran out of gas ).

[boxrodz]

Scoti-

I looked at the wiring schemetic for my year truck which has an Aux tank, but I think the concept should be the same.

Your sending unit should have a lead (probably the Black wire) for chassis ground and the 'sensor' lead (probably the Tan wire) wired to your gauge. So, like you tested, you should see 12+V between the 'Black' lead and the positive terminal of the battery. I wouldn't consider the Tan wire as a ground, but as a 'sensor' connection to the gauge. It sounds like the sending unit tested ok, if you are able to observe resistance readings between 0 and 90 ohms. I'm not sure of the full range of the sending unit, but I agree with your buddy's assessment that it is possible to have readings above 90 ohms. If I understand your write-up, it sounds like you have 'stable' readings through the full arc of travel. If so, it sounds like your sending unit is ok.

I haven't heard of the 2-DD batteries gauge test. I tried the test with a variable power supply and another guage that I have, and could not produce the 1/2 reading. I'm thinking the guage acts like an amp meter, so it just senses the amount (and direction) of current based on the voltage applied across a specific resistance. So, I don't think it would register a value if no resistance is present. That also would explain why you would notice the gauge needle move in a clockwise direction with 'correct' hookup and in the opposite direction if the hookup is reversed (i.e. sensing the direction of current).

If you have access to the Fuel gauge terminals, you could check your gauge another way. You can use the truck battery or a 2-DD battery setup. Connect the ground terminal of the gauge to ground. Connect the positive terminal of the gauge to positive side of the battery. Then, connect a resistor (90 ohms or less) between the 'sensor' terminal of the gauge and the ground. If you use a 45 ohm resistor, you should see a '1/2 tank' reading. Other resistor values should give you a proportional reading. If you really wanted to get fancy, you could stick a 100 ohm pot(entiometer) and vary the resistance to check the reading of the gauge. But, usually with meters it either works at all values or not at all.

From your description, it almost sounds as if you have a bad (open)connection between the guage 'sensor' terminal and the sending unit. When the ignition is swiched on and power is applied to the gauge, if a bad 'sensor' connection exists, then the gauge would read past full at the 3 o'clock position. You could check the connection between the connector on the gauge's 'sensor' terminal and the connector at the sending unit using a long wire lead and your meter (ohms). Connect one end of the long wire to the gauge 'sensor' connector and the other end to one lead of the meter. Connect the other lead of the meter to the sending unit 'sensor' connector. In the ohms setting, the meter should read 0 ohms for a good connection. If it reads infinite, then you have a bad connection. You can use the same process to pinpoint the location of the bad connection.

Good luck.

[SCOTI]

Boxrodz, what you're saying is the conclusion I came to.

My best guess is there's a bad connection between the printed circuit board/gauge wiring bulkhead & the fuel gauge sending unit.

The problem w/chasing that connection is on the 74's gauges is there are no 'terminals' on the back of the gauges to connect anything to. They have some weird 'nuts-zert' style attachment that took extreme patience to hook my clips onto out of the truck w/lots of room. Trying to accomplish that task in the truck I feel is near impossible w/o me kicking the sh!t out of something from frustration.

I was trying to decipher how the power & grounds are routed through the printed circuit board to see if I could duplicate that routing w/some lengths of wire & aligator clips but haven't had any luck in that.

This is how it typically goes for me & makes me really wonder why I mess around w/cars. To recap my last 2 days:
I wanted to fix my broken fuel gauge after getting home from work on Sunday (my day off).
Day 1, I drop the tank & check the wiring.
I check more wiring.
I run temporary wiring.
I go break something, clean-up my tools & stuff because it's now pitch black outside. I go to work @ 3am (using her car). I take a personal day & get home @ 8am so I can get this wrapped up.
Day 2, I attempt to solve the issue w/info & suggestions from the web.
I check wiring.
I check the sending unit resistance.
I decide to take the gauges out to check all connections to the gauge. But first . . .....
I have to remove the stereo (cluster won't come loose because of stereo chassis).
Next, the headlight switch. The shaft wont come out. I'm hurting my finger from pressing so hard on the small release button to dislodge the shaft. I can't access it good w/o removing the a/c ducting. All of it comes out when I attempt to remove 1pc.
I finally have to pull the freaking headlight-switch shaft out w/much more force than should be necessary.
Then while removing the gauge cluster, the dash pad is about to fall off so I remove it.
I run conductivity tests on the gauges.
I compare another set of gauges w/the same tests.
I make phone calls & research more info on the net about my situation & eat some leftover KFC (finger lickin' good).
I hook up a gazillion jumper wires & check both fuel gauges through the sending unit.
Realizing I'm running out of time to complete this task, I admit defeat & re-install the tank so I have a way to get to work tomorrow.
I wrestle the tank into place.
I wrestle the tank into place some more.
I cuss the exhaust guy for welding the muffler hangers directly in front of the gas tank bolt holes because I have to fish my arm behind the muffler & blindly search for the mounting strap holes.
I get all bolts in. Half w/some daylight remaining, the other half using a droplight.
I double check all wiring, bolts, muffler clamps & the drop it down to the ground.
I begin re-installing the gauge cluster.
I struggle w/aligning the oil sending unit hardline w/the oil pressure gauge.
I cuss.
I cuss much more.
I get the gauges temporarily installed.
I re-install the headlight switch.
The headlights won't work.
I double check the headlight switch & now that shaft that was nearly impossible to remove earlier will come out w/o hitting the small release button.
I try harder to get the shaft to engage the switch & no go. The switch is now broken.

Sooooo...... I started yesterday to fix my gas gauge only to wind up a day & a half later w/not only my gas gauge still not functioning..... now the dash is still out, I have no radio, & now my freaking headlight switch is broken as well & I have to fix it before I can drive to work early tomorrow morning.

I do this stuff to relieve stress . My friends call it 'Scot-luck'.

[boxrodz]

Originally Posted by SCOTI

Boxrodz, what you're saying is the conclusion I came to.
.
.
.
This is how it typically goes for me & makes me really wonder why I mess around w/cars..
.
.
.
I do this stuff to relieve stress . My friends call it 'Scot-luck'.

Boy, I guess misery does like company.

Your experience sounds like what I go thru a lot of the times. For me, I just gotta laugh when it happens now... from repairing one thing while damaging other things to breaking expensive/hard-to-get little parts to getting oil/fluids/grease all over a clean garage floor and the list goes on and on...

Good luck on tracking down the problem.

[SCOTI]

I'm not going to mess around w/locating a stock gas gauge. I'm going to switch them all out for aftermarket once I decide which brand will look the best.

When I do the swap, I'll replace the wiring all the way back to the tank since it won't be routed through the factory circuit board harness any longer.

Thanks for the tips & suggestions!

[Captkaos]

I am looking at the '74 service manual. This is what it lists to check:

Never reads empty of read full all the time with ignition on -> check for disconnected/loose tank unit feed wire at tank unit -> check cluster fuse -> check for proper connections at Dash unit to printer circuit, IP harness to printed circuit, Flat wire to IP harness, trunk (frame ground) harness to flat wire

Check for grounded tank unit lead (worn signal wire)

If you disconnect the signal wire to the tank the guage should go to full. If it doesn't and you think the dash guage is working properly, check for broken printed circuit or a short in the wiring.

The way you are describing this makes me think your signal wire is shorted/grounded...

BTW, I can possibly scan this for you, and email it to you if you want.

[NeCrOmAnCeR]

Sorry, mis understood what you were trying to say about testing the ground wire. The easiest way to check the gauge is to ground the sensor wire. With the wire grounded the gauge should read empty, with the wire disconnected the gauge should read full.

As another member mentioned, this wire provides a ground ,(through the sending unit) for the gauge. As the resistance of the sender changes, more current flows through one of the windings in the gauge. Within the gauge there are two sets of windings, one with a fixed current (set by the white ceramic resistor connected to the back of the gauge) and the other coil experiences varying current flow through the sender. As the current in the second coil changes so does the magnetic field surrounding a permanent magnet (connected to the needle) causing the gauge needle to rotate.

Sounds like you have checked nearly everything, so ground the wire at the sending unit, and check the gauge. If it still doesn't register a change, check the wires themselves (all of them).