I have a `92 SSEi... the fuel guage zeros when I fill up, then comes back after about 30-35 miles of driving... also, it doesn't go below 1/8 tank... I'm not a mechanical guy... is it difficult/expensive to fix...? any help appreciated... thanks...

Sounds like you need a new float/sending unit in the gas tank. If the gauge works normally through at least part of its range, I would say the gauge itself is probably fine but it'* getting bogus signals from the sending unit in the tank.

thanks... any guestimate as to cost or difficulty of replacement...?

There is a TSD on your problem. It involves installing a resistor in the wiring. I've seen it on this site somewhere.

Yes, it is a known circuit problem. When I had my instrument cluster rebuilt, it included a fix for that. I think the "real" fix is a different program in the cluster, but it makes sense that a resistor would also fix it, by never sending that "extremely full" voltage to the cluster.

However, the accuracy of my gauge is still the pits, it'* all over the place and I have not figured that out yet.

Fuel Gauge Problems 92/93

Fuel Gauge Goes to Empty When the Fuel Tank Is Full
Bulletin Number: 93-8-10
Reference Number: 308102
Publish Date: 2/93

Subject:
FUEL GAUGE READS EMPTY WITH FULL FUEL TANK (INSTALL 2000 OHM RESISTOR IN PARALLEL W/FUEL SENDER)

Models Affected:
1992-93 PONTIAC SSE AND SSEI

CONDITION:

Some 1992-93 Pontiac SSE and SSEI vehicles may exhibit a condition where the fuel gauge needle goes to empty when the fuel tank is completely full. If this occurs, the chime sounds and the "CHECK GAUGES" message comes on the display. After a short period of time (30 seconds to 10 minutes of driving), the gauge will recover and begin working again.

CAUSE:

The cause of this condition is a high resistance build-up on the fuel tank sender. The cluster interprets this high resistance (typically 105 ohms) as an electrical fault and sends the fuel gauge needle to empty.

CORRECTION:

The condition may be corrected by installing a 2000 ohm (2 kohm) resistor in parallel with the fuel sender.

DIAGNOSIS AND REPAIR:

Refer to section 8A-81 in the Service Manual for circuit details.

1. If the owner explains that the fuel gauge goes to empty when the fuel tank is filled then begins working after some time, follow steps 2 through 9 below. If the fuel gauge is at empty all of the time, follow the diagnostic procedures in the Service Manual for the cause of the condition.

2. Disconnect the negative battery cable.

3. Remove the lower rear seat cushion.

4. Locate the conduit on the left side of the car and expose the Purple (circuit # 30) and Black/White (circuit # 803) wires. Be careful to get the correct wires since there is also a purple/white wire in the opposite bundle.


The correct purple and black/white wires are in the wiring bundle that has the clear level ride line (ELC) in it.



5. Strip back about 1/2 inch of insulation from each wire as shown in Figure 1. Do not cut the wires.

6. Solder one end of the resistor to the purple wire and the other end to the black/white wire as shown in the figure. A 1/4 or 1/2 watt 2000 ohm resistor is recommended. However, the resistor can be from 1800 ohms (1.8 kohms) to 2200 ohms (2.2 kohms). You can buy this resistor at your local electronics store. Radio Shack normally carries a 2.2 kohm resistor.

7. Tape the connections to prevent any electrical shorts.

8. Put the wires back into the conduit, install the rear seat pad, and connect the battery.

9. Turn the ignition and verify the fuel gauge works correctly.


WARRANTY INFORMATION:

Operation No.:
T5338

Labor Time:
.4 Hr.

thanks guys... it'* a great help that I don't have to guess what the problem is...

But the problem with his '92 isn't that the gauge reads empty when full, it reads it has some gas in it when it doesn't. My '96 Olds 88 does the same thing. Our problem is the opposite of the service bulletin that fuddyduddy posted.
Would it still fix our problem?

my `92 showed that I had 1/8 tank when I ran out of gas last saturday (3rd day I owned the car)... maybe the higher voltage causes both problems...

A bit of both, actually; his original post reads, in part:

Since there is a known problem with a known repair, the best thing to do at this point would be to apply the known repair, then see if any further symptoms remain.