jahajazz

My 1996 z28 has a very odd fuel gage/pump problem. The car has 150k miles, and I just replaced the fuel pump 3 months ago. A week ago, it was taking a while to start, so I checked the fuel pressure. The pressure would rise to about 40psi, then drop to 0. But, once the car did start, it ran fine and held pressures at 36psi. I figured it was just a bad check valve so I ignored the problem.

A few days ago I went to start my car, and my fuel gage was all the way to the right, to the point where the needle couldnt move any further. It idled around 600rpms and then died. The engine wouldnt rev steadily, and with mid throttle it still died. I thought it was a bad fuel pump, but heres the weird part; the car ran perfect 45 minutes later. Today, a few times in the middle of driving the fuel gage needle would do the same thing, but the pump recovered fast enough to keep driving.

The last time I was driving the car tonight, the fuel gage needle fell all the way to the right, and the engine died. I pulled over, and the car started after like a minute. I was able to drive home after that, but the whole drive home the fuel gage needle would drift back and forth between the normal level and way right.

Currently, with the car idling, I can hear the pump dying and turning back on. So, is this just a pump failure? Or could there be a problem with the electricals somewhere? What can I do to find out?

Injuneer

Sounds like you may have a loose, corroded or damaged harness connector back at the pump.

Guest47904

The connection is opening intermittently. The first place to look is under the body behind the rear seat on the drivers side of the car. It's easiest to remove the left rear wheel to find it. There is a connection there that carries the fuel pump power and the fuel gauge signal. The fuel gauge uses a variable resister in the tank. The resistance goes down lower as the tank level goes down. If the signal is broken (that is open) the resistance is infinite and the gauge pegs. More than likely you are loosing both signals at that connector.

You at least want to check there first. If it's there but you still have problems, you may have to get to the top of the tank next. The connector is a double cavity that is fastened to the body. It has 2 connectors mated to it. The one your having problems with is the top cavity. More specifically the gray, purple and black wire on the top row.

Gray is fuel pump power.
Purple is fuel gauge signal.
Black is ground for both.

Visually check the connections and the entire length of the wires as much as you can get too. If you see nothing, try re-seating the connector. If that does nothing for you, it's time to back probe the three pins with a meter.

shoebox

More help to go along with what Speedy said.

jahajazz

Thanks speedy and shoebox for the info. Shoebox that site is excellent! Great photos of the area in question and I liked the straight forward electrical troubleshooting. This is what we found:
Fuel gauge read 7/8, ohms 60 (3/4)
Fuel gauge read E when wires disconnected with key on.
Fuel gause pegged when grounded with key on.
We made sure the connector was well seated and did not see any damage to the harness. The harness to the pump looked good as far in as could be seen and is new as came with the new pump installed 3 months ago.

We started car and it ran, but we could hear the pump fluctuating. Then after lowering the car and double checking, at first the pump did not come on when keyed, then was fine. We attached the fuel pressure gauge and while the pressure was 32 at 2500 rpm (held) it would twitch every 5 seconds or so about 1-2 psi.

We do have an issue where the radio and power windows cut out and beleive the BCM is invloved, could this be related? It seems that with the gauge and pump problem happening at the same time it must be electrical and the initial tests seem to have ruled out that portion of the wiring.

Any other ideas on what to check next would be greatly appreciated.

Guest47904

If you probed the gauge connector like Shoe's site shows, you proved the gauge works. But your problem is the signal is dropping out intermittently. So the connector has to be mated when you are checking it.

Probe the power connections to the pump and have someone monitor the meter while you hear the pump doing it's weird behavior.

Leave the connector together while you are back probing it. If either the gauge or the pump stop and start but the meter shows no change, you know the problem is some where between the connector you are probing and the tank.

At that point it would be time to lower the tank.

jahajazz

Speedy, what is the best way to back probe. This connector is not too accessible as it is supposed to be water proof. I thought I could pin the wire, but I don't want to pierce to insulation.

Guest47904

Snap-on dealers carry them too.

back probes

jahajazz

Speedy, we back probed the connection. We did not get a complete failure, but did collect these results:

When keyed, about 10.9 volts. Pump ran the 2-3 seconds then shut off until fully turning the key to engage starter. Engine started fine.
When running volts between 13.35 and 12.95 as time passed.
Could hear fuel pump change frequency while running without complete stopping, the voltage changed about .10 volt (ex. 13.20 to 13.10) when the frequency changed. The change was just a quick 1-2 second event.
The fuel gauge did not peg, but it did twitch in what seemed the same event as the pump sound change.

It seems the voltage does change when the pump sound does, but is .10 volt a true indication that the source voltage is at fault?

shoebox

.10v is not very significant, IMHO. I would also want to know if the switched ignition voltage to all circuits changed at the same time (not just in the f/p circuit). That seems more likely to me. But then, I still don't think that voltage change is necessarily indicative of a trouble.

jahajazz

How can I test to determine if the switched ignition voltage to all circuits changes at the same time, and not just the fuel pump circuit?

shoebox

Just put your meter on the fuse of any switch controlled circuit. Fuses have little access slots on the top.

jahajazz

Ok... Rechecked voltage last night. No fluctation at pump feed or ignition, even that crazy 1/10th. Decided to double check fuel pressure and saw the weirdest thing. Prime pressure 42, running 32. Shut the car off and the pressure bounced down to 28, then slowly started to RISE until it stopped at 40. Tried it several times and it repeated. I thought maybe having the rear on jackstands was involved, so put the car down, but it still happened. I bled the line at the schrader and it took two blue shop towels to soak it up until it started to spit. I also took the gas cap off thinking the tank was pressurizing, but no effect. Otherwise the darn thing ran fine. The car has smelled like gas after a long run in the summer, is the vapor canister involved?

Injuneer

Sounds like you may be heating the fuel in the tank, and building vapor pressure. A failing pump will cause the fuel to heat up. When you depressure at the Schrader, it should spray a little fuel, and drop the pressure rapidly. Sounds like pressure in your tank may have forced the fuel out of the line all the way from the tank to the Schrader, then started to "spit" when vapor started to flow out. By the time you went back and pulled the cap off, you had already released the pressure from the tank.

jahajazz

I actually ran the car with the cap off and it still did it. Funny I did not pull a check engine, it used to do that if I did not make sure the cap was on tight.