Strange Fuel Gauge Issue

[emceebrooks]

Guys,
I've searched through the forum to try to find this specific problem. If someone has already posted an answer to this and I've missed it, I'd appreciate a link to it. If not, here goes....

We replaced both the fuel gauge and fuel sending unit in April. The gauge has been operating perfectly since then. In July, we were out of town for about a month and the car sat unused during that month. Last week, as I pulled out of the garage, I noticed it was on empty (It had been a little above halfway before our trip). When I stopped at the gas station, I turned off the car, put only 4.5 gallons in the tank and the gas gauge when back to "full". Then when I started the car, it immediately shot down to "empty".

After some more fiddling with it this week-- when the car is off, the gas gauge seems to works perfectly. When the ignition is turned on accessory or the engine started, the gauge goes immediately to empty. It seems electrical to me, but where to start?

Thanks for your help.

[bbodie52]

There are two electromagnetic coils inside the fuel gauge. The one that is on the FULL side serves as a constant magnetic pull on that side. The strength of the magnetic field on the FULL side never changes when power is on. The other coil on the EMPTY side of the gauge has more windings, so it is capable of exerting a magnetic field that is greater then the other coil. It is also a variable coil that is controlled by the position of the float in the fuel tank. When the fuel tank is full the magnetic field exerted on the empty side of the gauge is very weak, because the sending unit at the fuel tank is set to maximum resistance when the tank is full. This causes the other coil on the FULL side to have a stronger magnetic field and pull the needle all the way to the right, which indicates that the tank is FULL.

As the tank is drained the float in the tank moves and the variable resistor decreases the resistance setting, which increases the amount of current flowing through the coil on the E side of the gauge. As the tank approaches EMPTY the magnetic field on the E side becomes stronger than the constant field on the F side, so the needle is gradually pulled towards E.

The E coil and the F coil both receive voltage from the same source, and only when the key is in the ON position. When the key is OFF, neither coil is energized and the needle sits idle, with no pull to affect the needle position. So when the key is OFF the gauge is shut down and does not reflect the fuel level in the tank. It generally does not move when the key is turned off, so it appears to show the fuel level based on the last reading before the power to the gauge was cut. When the key is turned on the gauge is energized and updates based on the strength of the two magnetic fields in the two coils.

You indicated that whenever you turn the key ON, the needle swings immediately to E, regardless of the amount of fuel in the tank. This would seem to indicate that the coil on the F side (which is supposed to be energized with a constant magnetic field whenever the key is on) is not being energized at all. Both coils receive electricity from the same terminal on the back of the gauge at the same time. The coil on the F side is grounded via the fuel gauge housing to the instrument panel and to chassis ground. The coil on the E side is grounded through the gauge sending unit at the fuel tank. Even though the fuel tank is apparently full, and the fuel tank sending unit with therefore be at maximum resistance, there would still be some current flowing through the variable resistor at the fuel tank. Because of this, the coil on the E side would be energized with a weak magnetic field. If the F coil is not energized at all, the needle would move to the E side, since the variable coil would be the only one that is energized at all.

I would guess that for some reason the fuel gauge housing is no longer properly grounded. This would prevent the F coil from being energized, even though voltage is being applied to both coils within the fuel gauge at the same time. (The fact that the needle moves when the key is turned on proves that voltage is being applied to the gauge). The F coil, which should be grounded directly through the gauge housing to the instrument panel, is apparently not grounded, which would prevent current from flowing through the F coil. Although there is a possibility of an internal defect in the gauge that is preventing the F coil from being grounded, this would be unusual. It is more likely that the path to ground between the gauge housing and chassis ground is being disrupted somehow. You should be able to test this by attaching a wire to chassis ground and temporarily touching the wire to the gauge metal housing. With the key on, providing a temporary ground path to the housing should restore the proper display on the gauge to reflect a full gas tank. If grounding the gauge housing does not make any difference, the gauge itself may be suspect in that it may have an internal defect in the F coil itself.

I have included some illustrations, wiring schematics, and a complete description of the fuel gauge operation below.

Here is some material, including some theory on how the fuel gauge circuit functions. Hopefully this will help with your troubleshooting...

Study the schematic wiring diagrams in the shop manual and supplement for your vehicle. This is a very basic circuit, with power coming from the fuse block via a wire to a connector on the fuel gauge. You should be able to measure 12V DC at this connector with a multimeter (with the key ON) with the other multimeter lead grounded. The other fuel gauge connector is a wire that is routed to the fuel tank sending unit. The sending unit must be properly grounded to function. When the tank is empty, the sending unit basically is at 0 ohms resistance, and full current flows through the gas gauge, giving an indication of Empty. As the fuel level rises, sending unit resistance increases, which reduces current flow and moves the needle toward Full. You should be able to ground the wire at the fuel tank to obtain an Empty reading, and an open circuit (tank wire disconnected) should show a Full meter reading. Note that the gauge itself must also be properly grounded for the "F Coil" (as it is referred to in the description) to be properly energized to pull the needle to the F mark. If the ground connection on the instrument panel is faulty, the "F Coil" will not be energized properly, and the gauge will not function properly.

You should use a multimeter to verify the presence of 12V DC at the gauge input. The resistance (ohm) setting on the multimeter can be used to ensure that the gauge housing itself is properly grounded to the car chassis. You should also be able to measure 12V DC at the fuel tank wire, with the other test lead grounded. If the voltage is missing there, you may have a broken or loose wire between the gauge and the fuel tank sending unit. (All voltage checks are done with the key ON)

If the voltage is present where it should be (at the gauge input and at the fuel tank sending unit wire connector), and if you have a good chassis ground at the gauge housing and at the fuel tank sending unit, there is a possibility that the gauge itself has failed. If you determine that the gauge is faulty, you may be able to get a used fuel gauge on eBay.

Although the circuit itself is fairly simple, there are several connection points using multi-wire connectors that could become corroded or loose. Corrosion in a connector can add resistance to the circuit and lower voltage, which can affect the gauge reading. The gauge itself gets its ground from a ground strap that connects to all of the instruments attached to the plastic instrument panel housing. As shown in the illustration below, that single multi-point ground strap is grounded to the car chassis at one point by one screw. If that connection is corroded or loose, the instrument panel may be improperly grounded, which can affect the performance of the fuel gauge. At the other end of the circuit, a poor ground at the fuel tank sending unit can also impact the gauge. But an open condition (no connection of the tan wire) at the fuel tank sending unit should result in the fuel gauge needle being pulled fully to the right (F mark). 12V DC at the gauge and with the gauge properly grounded should apply full power to the "F Coil", resulting in a full deflection to the right. Only current flowing in the opposing coil (through the sending unit to Ground will apply a magnetic pull in the opposite direction, to pull the needle away from the "F Mark" and toward the "E Mark" as the current through the sending unit increases as the fuel in the tank is reduced.

The schematic wiring diagram and the written description below explains how the fuel gauge circuit functions.

Left-click each image once to enlarge, or a second time for maximum enlargement for better viewing...

1964 Corvair Fuel Gauge Circuit

1964 Corvair Passenger Car Combined Schematic

[emceebrooks]

Super helpful, thanks! I will give it a shot and let you know...thanks!

[emceebrooks]

Brodie, on a subsequent read, I think there's a typo on your response: "The coil on the F side is grounded via the fuel gauge housing to the instrument panel and to chassis ground. The coil on the F side is grounded through the gauge sending unit at the fuel tank." - To clarify you mean "The coil on the E (Empty) side is grounded through the gauge sending unit at the fuel tank", right? Because the sending unit increases the magnetic pull as the tank is emptied, right?

[bbodie52]

Yes. I guess at around 2:00 am my eyes did not pick up on the F that should have been an E! I corrected the error.

[emceebrooks]

Haha! I noticed you were pulling an all-nighter on that post. Thanks again!

[emceebrooks]

So I pulled everything apart this weekend and tested the gauge by connecting the housing to the chassis (while the key was "on") as you suggested, but no change; it operated the same way. Interestingly, in a lot of "offs and ons" while diagnosing this, I've noticed that the gauge doesn't return to full 100% of the time, when the key is turned off, but it does most of the time, which makes me think even more that the gauge might be defective. So, it sounds like the next step is to replace the gauge and see if it resolves? Thanks for your help! Marcus