Weird fuel problem

It's a black 1990 car, older style body with the more modern wrap around dash with a digital speedo/fuel gauge and analogue gauges for rpm, oil, coolant, voltage etc. Really good shape, only done 80,000 and it lived in Japan till 2006.
I'm in Scotland so its a pretty rare car over here, just imported a new weatherstripping set for it (given all the rain we get I figured it was a good idea to replace it) and I've replaced the faulty TPS sensor....car is running great aside from one slight issue!
I've looked and looked and looked and although there are loads of threads about fuel sender and gauge problems I cant find any that match my problem.....seem to be unique, or I'm just not looking in the right places!
So.....my problem......basically the fuel gauge does not read right. With the engine off and ignition on, the gauge is accurate.......BUT.......as soon as I start the engine the gauge slowly moves up to read "FULL". If I turn the engine off and just put the ignition on, the gauge goes back to reading accurately.
I've removed the sender and cleaned the float arm connector like is described on various forums but it's not made any difference.
I read that the fuel gauge will show full when there is an open circuit (i.e no signal being received from the sender)........so it's as though having the engine is running is causing an open circuit (no signal), but it's closed and sending a signal with just the ign on......really odd as I don't see why the signal would change just because the engine is running.....
This weekend, I'm planning to try running an earth cable directly from the -ve terminal on the battery to the sender unit body and also cleaning the multiplug on the sender unit body, incase it is a bad earth......but thought I'd check on here to see if there was anyone who has had the same issue and knew what the cause was?
I don't think it is the sender unit itself as it is reading right with just the ign on. If you take it out the tank and move the float arm up and down with the ign on, the dash gauge moves up and down correctly......
Thanks,
Alastair
Does it have RH or LH drive??
I think you are on the right track with a possible grounding (or maybe power) issue.
How is fuel pressure with ign On vs running (remove & plug the vac line to the FPR for this test), as the pump uses the same pwr/gnd as the sender?
Not tested the fuel pressure with the ign on vs running. I've not got a pressure gauge to hook up to it (is on my list of "tools to buy" though lol). Car has LOADS of power, idles perfectly and always starts instantly, even if its been sitting a few days, so I assume the pressure is fine?
once I loosen the wires...all good...a real dumb *** attack...just a shot
Right, I took the entire unit back out the tank yesterday and cleaned all the connections on the body of the unit and on the multiplug.
The male end of the multiplug (on the cars main harness) has 4 wires (black, purple, tanish grey, green), the female end (coming out the sender) is has only 3 wires (black, purple, grey).......is that normal or is it the wrong sender that's in it?
I assume that:
black = -ve earth
purple = fuel level signal output
grey= fuel pump +ve
With the sender unit OUT the car but still plugged into the harness, with the ignition ON but engine OFF, I can move the float arm up and down and the dash gauge moves up and down from FULL to the RESERVE warning coming on......so works as it should.
With the sender unit OUT the car but still plugged into the harness, with the engine RUNNING, the dash gauge starts correctly but moves up to FULL within a few seconds, regardless of the float position.
With the sender unplugged and a multimeter connected across the Purple and Black wires, I can move the float arm up and down and the reading on the tester moves between 1.6 to 10.6
With the sender unit in the tank and unplugged the reading is 6.4 so just over half a tank.
As soon as I plug the sender plug into the main harness, the reading jumps instantly from 6.4 to off the scale. So this is leading me to believe that there is a short on the Purple wire, somewhere between the multiplug and the gauge?
Any ideas on:
a) How the hell I'll find the short without pulling the whole car apart?
b) Could I cut the purple wire at the pump and run a new wire directly to the gauge cluster....if so, what wire do I connect to on the gauge itself?
c) Could I cut the purple wire and wire it to a stand alone, aftermarket fuel gauge.....if so, anyone able to recommend one?
I've ordered a FSM for the car, but it'll take a few weeks to get here by time UK customs are done with screwing me on import duty, so I'm working blind just now.
Right, I took the entire unit back out the tank yesterday and cleaned all the connections on the body of the unit and on the multiplug.
The male end of the multiplug (on the cars main harness) has 4 wires (black, purple, tanish grey, green), the female end (coming out the sender) is has only 3 wires (black, purple, grey).......is that normal or is it the wrong sender that's in it?
I assume that:
black = -ve earth
purple = fuel level signal output
grey= fuel pump +ve
With the sender unit OUT the car but still plugged into the harness, with the ignition ON but engine OFF, I can move the float arm up and down and the dash gauge moves up and down from FULL to the RESERVE warning coming on......so works as it should.
With the sender unit OUT the car but still plugged into the harness, with the engine RUNNING, the dash gauge starts correctly but moves up to FULL within a few seconds, regardless of the float position.
With the sender unplugged and a multimeter connected across the Purple and Black wires, I can move the float arm up and down and the reading on the tester moves between 1.6 to 10.6
With the sender unit in the tank and unplugged the reading is 6.4 so just over half a tank.
As soon as I plug the sender plug into the main harness, the reading jumps instantly from 6.4 to off the scale. So this is leading me to believe that there is a short on the Purple wire, somewhere between the multiplug and the gauge?
Any ideas on:
a) How the hell I'll find the short without pulling the whole car apart?
b) Could I cut the purple wire at the pump and run a new wire directly to the gauge cluster....if so, what wire do I connect to on the gauge itself?
c) Could I cut the purple wire and wire it to a stand alone, aftermarket fuel gauge.....if so, anyone able to recommend one?
I've ordered a FSM for the car, but it'll take a few weeks to get here by time UK customs are done with screwing me on import duty, so I'm working blind just now.

Done some more reading online......am I right in thinking that cars with the early dash 84-89 have a 3 wire pigtail on their fuel sender units and the later 90-96 cars with the newer dash have a 4 wire pigtail on their sender units??
Just wondering if someone has installed the wrong fuel sender at some point in the past as my factory harness has a 4 wire pigtail going to the sender but the sender only has 3 wire pigtail going to the harness????
The Best of Corvette for Corvette Enthusiasts

I am leaning towards it being a wiring issue - something disconnected either inside the tank or to the CCM.
Here is some troubleshooting that will help you determine if the sender is working correctly:
http://forums.corvetteforum.com/1571003242-post3.html
Good luck,
Matthew
You don't need to do all the resistor tests. Just moving the float arm will tell you if the sender is working correctly.
The fuel gauge on the dash is heavily damped to avoid it bouncing around when the fuel sloshes in the tank. It takes about 5 seconds for a change in the sender to appear as a change on the dash.
One thing to watch for is to make sure the dash shows no bars when the float arm is at the bottom extreme. Otherwise you'll think you have fuel in the tank when you don't.
Cliff.......if I take the sender out the tank and leave it connected, to the harness and then move the float arm up and down, the dash gauge moves up and down in relation to me moving the float arm. the gauge will go all the way from full bars, down to 0 bars with the RESERVE light on and then back to full bars.
So does that mean the sender is definitely working fine then and the fault is in the wiring harness going from the sender to the gauge?
The gauge only reads wrongly with the sender in the tank and engine running. Out the tank with just the IGN on, it reads fine???
Only other thing I thought of yesterday, would be voltage......with just the IGN on there is only 12V running in the system, but with the engine running there will be 14.4V as the alternator is putting out voltage....am I away on wheels here, or could the higher voltage output from the alternator be affecting it?

That being said, I remember another member doing a fuel pump on his 90 after I did my 89 and the wiring inside the tank was different. There should be a purple wire inside the tank (from float windings to the top of the assembly) that provides the fuel level. I would check that.
http://www.corvetteforum.guru/module...orumpost157934
Here is the install link: http://www.blowerworks.net/download/GSS340IG.pdf
You don't need to do all the resistor tests. Just moving the float arm will tell you if the sender is working correctly.
The fuel gauge on the dash is heavily damped to avoid it bouncing around when the fuel sloshes in the tank. It takes about 5 seconds for a change in the sender to appear as a change on the dash.
One thing to watch for is to make sure the dash shows no bars when the float arm is at the bottom extreme. Otherwise you'll think you have fuel in the tank when you don't.

In that case I would do what Cliff says. Pull the sender and move the float arm, test the resistance on the purple wire.
Last edited by Matatk; Jun 17, 2014 at 10:29 AM.














