Thread: Fuel gauge nightmare

Hello. Just joined today and need HELP. I have a 02 SE with a 3.1. The fuel gauge at start up goes from full to empty .I already replaced the fuel sending unit and float but it still does the same thing. The weird thing is if I put it in gear while its dropping from full to empty it stops where its at. Can anybody help?

I am having the same problem with my 2001 GP 3.1, but it is coincident with the OBD II diagnostics throwing codes P0122, P0425, and P1635. My problem appears to be with a low 5V reference voltage that is supplied to various sensors being pulled down somewhere. See my post on this a few up the list from yours. If you replaced the fuel sending unit and still have the problem, and are not also getting those OBD codes, you most likely have a bad ground circuit (the black wire to the sending unit). From what I can tell from the circuit, if you lose the ground connection, the voltage on the sensor wire (purple) from the sending unit to the power train control module should go up to near the full 5 volts of the reference supply voltage, and drive the gauge indication to the full peg or to full empty (I haven't gotten that far in how the voltage should track with tank level). My Haynes manual diagram does not show the connection from the 5V reference to the sending unit, it must be internal to the power train control module (PTCM). It also appears that the fuel gauge itself in the dash is not driven directly by the voltage coming off the sending unit, but is fed serial data from a bus in the power train control that supplies a whole bunch of the dash indications from sensor inputs within the PTCM. If the wiring to the sending unit is correct, you should see about 0 volts relative to ground on the black lead, and something between 0 and 5 volts on the purple lead depending on the level in the tank 0 empty, 5 full (although it could be the opposite, like I said I haven't nuked this problem out that far yet. If you see about 5 volts on both the black and purple lead, it means that there is an open circuit somewhere the black lead that goes back to a ground connection in the PTCM. If you see 0 volts on both leads , then there is likely an open in the purple lead going back to the PTCM, or it is shorted to ground. If you are not following this, seek out someone with an electrical background to help you troubleshoot.

i have the same problem with my 02 se

I am getting the code 5-volt reference (a or 1 )circuit. I really cant afford to pay somebody to diagnose the problem so I have to try myself. Hopefully somebody on here has found or will find the solution.I used to do this stuff for a living but that was when it was a lot easier. Also had a code that said Baro censor so I replaced the Map sensor and it still idled fast. So I got a TPS and guess what still the same thing. Maybe its all associated with the same thing. Driven me nuts and trust me thats a short drive.

Found another site thats says something about a PCM. suppose to control the fuel gauge the TPS the MAP sensor and couple other things. Could replacing that solve the problem?

Originally Posted by ajhartle

Found another site thats says something about a PCM. suppose to control the fuel gauge the TPS the MAP sensor and couple other things. Could replacing that solve the problem?

It could, but if if its something else, like shorted wiring, you just threw your money way one something you didnt need. Easter egging the problem by changing sensors willy nilly is a waste of time and money. Try to eliminate the simple things like bad wiring, which is probably a more likely root cause first.

What i am going to do on my 2001 3.1 when it warms up outside a bit is unplug all the sensors supplied by the 5v reference (TPS, tank pressure, EGR, MAP, and gas tank, and measure the 5V reference signal on the reference voltage supply wiring, it looks like its the gray wire to each of the sensors. If its up to 5 V, with the ignition switch to the RUN position, then one of the sensors is the problem.

Start plugging them back in until one of them pulls the voltage to down, then you have found a likely suspect sensor.

If the 5V reference is still low with all the sensors disconnected, then there is likely a short or a partial short to ground on the 5V supply wiring to one of the senor's plugs, or in the common wiring from the PCU, OR the 5V reference supply itself in the PCU is shot.

You may have to disconnect and reconnect the 5V supply wiring to individual components to try to localize a short if you see a no or low voltage condition with all the sensors unplugged. For starters, if there is a single wire coming out of the PCU (I still have not had a chance to look), you might make the first break right at the wire coming out of the PCU. If there is no 5V at the PCU with all the feeds to the sensors isolated, then the problem is internal to the PCU.

If you do have voltage there, start looking looking for shorts or partial shorts of the wiring feeding the individual sensors. By a partial short I mean a low resistance path to ground that should not be there, could be due to insulation breakdown and salt spray on the exposed wiring this time of year. Look for visual indications of damaged insulation, some discussions of the problem I have found indicate that the wiring under the PCU sometimes can break down and cause shorts on the 5V reference wiring. Hopefully the 5V supply circuit inside the PCU doesn't get taken down for good by an external short.

BTW, it looks like the fuel gauge circuit gets its 5V excitation internally to the PCU, the wiring external to the PCU for that circuit including the level sensor appears to be just a variable resistance to the PCU ground circuit.

If all this seems to complex for you to do yourself, find a friend who is electrically inclined to help you. Paying a shop to chase electrical problems can get real expensive real fast, and take a lot of time, unless you get someone who has a brain and a knack for doing it, and isn't just a mindless parts swapper. My gut, and over 40 years of experience in working on electrical and electronics systems, including my vehicles, says I'm going to find this to be a bad wiring problem and not a sensor problem.

If my 3.1 GP troubleshooting indicates that I have lost the 5V supply within the PCU, rather than shell out big bucks for a replacement PCU, I'll try to build up a small external 12 to 5V DC supply (should only need a few parts) and feed the sensors with it. The first attempt will likely be three D cell batteries in series, should bring me close to 4.5 volts, probably enough to make things work. I'll keep you posted.

One thing I wish I had is a schematic that shows the internals of the PCU, in my Haynes manual , its just shown as a blank box, and I don't think the lead pin out numbers on the PCU are shown correctly. Anyone know where I can find this info?

I unplugged the egr valve and the fuel gauge went back up. Is that the answer my EGR valve? nothing change when I unplugged any of the other sensors.

I have found on a handful of cars that the camshaft position sensor under neath the power steering pump wires get chaffed and the 5 volt reference wire shorts on the intake or power steering pump causing many weird problems and setting various 5 volt reference codes.

Well wasted more money on a egr valve.Wasnt it however I did find wires spliced together hidden under those black flex pipe wire holders. I unhooked the wires where they spliced them on at a time. Got to the red wire unhooked it and the idle went to normal the fuel gauge read correct and so far no check engine light.I did reconnect the other wires all except the red and so far so good. Now I have to figure where the red wire leads to.

Yes I was

Originally Posted by ajhartle

Well wasted more money on a egr valve.Wasnt it however I did find wires spliced together hidden under those black flex pipe wire holders. I unhooked the wires where they spliced them on at a time. Got to the red wire unhooked it and the idle went to normal the fuel gauge read correct and so far no check engine light.I did reconnect the other wires all except the red and so far so good. Now I have to figure where the red wire leads to.

Looks like you are getting close, hope the EGR valve didn't set you back too much, but at least you have ruled it out as the issue and what you determined with the red wire is suggesting that the problem indeed lies in the EGR circuit.

Is it RED or PINK, my Haynes wiring diagram shows the following wires on the EGR valve for your engine/year (same as my 2001 3.1), no RED wire noted:

Pink - EGR Valve 12VDC solenoid excitation (comes from main 12V PCM IGN 15A fuse in the underhood fuse panel near the battery - with this harness wire disconnected at the EGR valve it should show +12VDC to ground (battery negative terminal) with the ignition switch in BULB TEST or START. I think by BULB TEST they mean the position you turn the key to just before going to START, basically where the switch sits once the car is running.

Gray (1 of 2) - EGR solenoid control signal - comes out of the PCM module. Not sure exactly what you should see there, by the circuit it is the other end of a solenoid coil that is powered by the pink wire, and returns the 12 Volts DC from the pink wire to ground through the PCM. The amount of current though the solenoid changes the position of the EGR valve stem to whats called for by the PCM based on various other signal inputs to the PCM, including the stem postion feedback signal from the potentiometer.

Gray (2 of 2) - 5V Reference Voltage for EGR valve stem position sensor potentiometer. This is the one that should have the +5V to ground on it that also feeds the other stuff like the fuel gauge. The pot is mechanically connected to the EGR valve and tells the PCM the actual position of the valve at any point in time. The solenoid and the pot are part of a feedback control system to put the EGR valve where it has to be. With the EGR valve unplugged you should see 5V there with the igniton switch in BULB TEST (i.e run).

Brown - EGR position signal - comes of the wiper of the position sensor potentiometer that has 5V reference on one end (the 2nd gray wire), and ground (the black wire discussed below). The voltage on this wire varies porpotrionally with the EGR positon between somewhere between 5V an 0 volts, proportional to valve position with the igntion switch in BULB TEST (run).

Black - EGR positon sensor pot ground return- tied to car ground at the PCM.

I"m wondering if whomever did the splices you found crossed the two gray wires up! That would likely really screw up the 5V reference signal, because it would be connected to the 12V from the battery through the valve positioning solenoid coil. You can tell which gray wire should be the one for the EGR solenoid and which is for the 5V excitation of the EGR position sensor potetiometer as follows: (BTW, shame on the GM engineer/designer making two wires on completely different circuits the same color on a 5 wire connector!) :

With the ignition shut off, and the connector installed on the EGR valve, after noting gray wire going into the harness is connected to which lead off the connector, undo the splices on the gray wires or otherwise cut them off from the harness so you can "look into" the EGR valve with an ohmmeter on the pieces of gray wire coming off the connector. Mark the wires so you know what your gray wore connections were going int this test. The connecter side gray wire, when not conneted to the harness wiring that goes to the solenoid will have a very low resistance with the plug connected to the EGR valve, and an open circuit with the connector disconnected from the EGR valve.

Now, with the two gray leads from the connecter still disconnected from the harness gray wires, turn the ignition on (but do not start the car). The harness lead that carries the 5V reference signal should show +5 V to ground. I am not sure what the harness side gray wire for the solenoid valve control should be, but it most likely will be low, since it appears to be a return to ground for the 12V from the pink wire. The piece of the gray wire on the connector side, with the key on, and the EGR valve connected, that is the gray wire for the solenoid return, should show +12V to ground (its getting it through the pink wire, with no current flow, there is no voltage drop through the coil.

At this point you should now know which gray wire on the harness side is the 5V reference, and which gray wire on the EGR connector side is the solenoid return wire. These two gray wires do not/should not be connected to each other. The gray harness side wire with the 5 V reference signal goes to the other gray connecter side wire.

You following all this? This is what I would do if it were my car with your issues, but if you follow my advice, don't come after me if something gets fried! And if it turns out that the gray wires were in fact hooked up right, and the only way you can clear the problem at this point is disconnect the RED/pink wire, I'm not any further alnog than you at this point.....not sure how driveablility might be affected. Now to go figure out my cars issue...going to do some troubleshooting on it before the snow flies! The wiring on the TPS connector looks kind of shaky for starters....

Thank you for the info. Very usable

Originally Posted by AJ1G

Is it RED or PINK, my Haynes wiring diagram shows the following wires on the EGR valve for your engine/year (same as my 2001 3.1), no RED wire noted:

Well, it looks like the wire is really RED, have been working on my problem. The EGR connector's wires on my 2001 3.1 also have a RED wire, no PINK, also one of the gray wires has a black trace, which is NOT noted on the Haynes diagram. Don't know who is responsible for the "typical" wiring diagram in the Haynes manual, Haynes or GM, but someone has at least tow errors on that one connector. I'm suspecting that the diagram was reverse engineered by Haynes, the red wire the reverse engineer was looking at was a faded red that he IDed as pink, and he missed seeing the black trace on one of the two gray wires.

That said, it should be easier for you to check your respliced wiring. Perhaps who ever did the splicing of those leads also missed the gray and gray/black trace and crossed them up them up. Should be able to check just visually, provided you can see the black tracer on both sides of the splice.

I'm about 99% certain I have found and fixed my problem..wiring to the throttle position sensor which is one of the sensors fed by the 5V reference voltage that is associated with the P1635 code, was pinched into the joint where the rubber air intake duct, the other end of which connects to the air mass meter, fits onto the inlet to the throttle body. Insulation was all chafed,wiring was all exposed under the chafe, and probably shorting to ground on the air mass meter side of the joint. Pulled the shorted wires apart and now my gas gauge reads exactly where is should. Have not yet fully confirmed everything else is fixed, because the HARD PLASTIC HOSE THAT FITS INTO THE NIPPLE ON TOP OF THE RUBBER AIR DUCT JUST UPSTREAM OF THE THROTTLE BODY POPPED OFF WHATEVER THE OTHER END CONNECTS TO WHEN I WAS TRYING TO LOOSEN IT UP IN PREPARATION FOR PULLING THE DUCT OFF THE THROTTLE BODY! Pardon the ALL CAPS, I'm venting because I haven't a clue as to where it goes, looked all over that end of the engine bay, and in the Haynes manual. If that hadn't happened, it would have been a ten minute fix for the original problem! Where does it go, what is it? Seems like it might be the evap emissions canister evacuation line. I just don't see anything it should connect to. If I can't find out in the next few minutes what it sucks on, I'm going to just plug the nipple on the rubber air duct for the time being and leave the damn thing off. Whatever it connected to must not have been a very tight fit in the first place, judging by the way the opposite end connection is designed. ARRRGH!!!!

My original problem is all fixed....normal gas gauge readings, all OBDII codes cleared, no Service Engine Soon light! I suspect I might have gotten the TPS sensor wire jammed up into the throttle body/inlet duct joint when I was putting the duct back on the throttle body several months ago after having to pull the duct off to get access to the air filter to change it. The manual says you can just flip the latches that hold the air box cover and squeeze the air duct to get at the air filter but that was pretty much impossible, I had to take the duct off. The TPS wire was right next to where the duct clamps onto the throttle body and evidently got caught in there while putting the duct back on, not easy to see it, but something to watch out for! BTW, the plastic hose is the oil breather vent pipe, plugs into the rear side of the engine on the valve cover.