Replaced fuel level sender, but no luck with dash gauge

[slanej]

So I had the flaky fuel gauge problem, and I replaced my Fuel Level Sender as per Hans' excellent video tutorial. Everything went smoothly (except for removing and replacing the lock ring - have plenty of PB Blaster on hand!) But now when I start the car, my fuel gauge pegs at full, then after a minute or so drops to dead empty. A minute after that, the low-fuel idiot chime goes off, even though I know I have 3/4 tank.

Just for something to take my mind off of the gasoline migraine I got trying to perform the fix, I ran a code check and I'm getting a never-before-seen P0452 - Evap Pressure Sensor Low, but I'm not getting a CEL illuminating.

I've been reading through the forums at length, but can't find a definitive answer. Does anybody know a step-by-step diagnostic procedure that can work me through the process (e.g., check O-ring 1st, check fuel pump wiring 2nd, check pressure sensor 3rd, etc)? Perhaps something from the service manual?

Anyway, I'm kind of banging my head against the wall, here. Any help you can provide would be greatly appreciated.

[Archon]

The code is for the fuel tank pressure sensor. So, it's likely that you don't have the o-ring seated properly.

Most tests would require pulling the pump again. If you wanted to try resistance measurements, you could see what the measurement is now, and then again after either filling up, or after using some of the fuel. See if the value changes.

[slanej]

Thanks for the reply, Archon.

(1) I punched off the lock ring again and inspected the O-ring. No tears, twists or damage that I can see. How likely is it to misalign the O-ring when re-seating the locking ring? I've got to wail on the lock ring pretty hard and evenly over it's circumference to get it back into place. Is there any way to prevent misalignment of the O-ring when locking it down?

(2) I pulled and reseated the wiring connectors. When restarting, I had some slight fluctuations on the dash gauge between the startup Full reading and the dead empty a minute later. But the same result, ultimately.

You speak of resistance measurements. I think I know where you're going with this, but just how do I perform these measurements? (Keep in mind that I'm hardly a mechanic or electrical expert, and that anything you tell me could potentially help others in the forum.)

Thanks in advance.

E.

[Archon]

You could try putting some Vaseline around the O-ring, allowing things to slide around it easier. Then, take it in to have the code cleared and see if it comes back.

Disconnect the connector to the fuel pump, place the meter leads across the connections on the pump where the wires for the fuel gauge sensor connects. Probably a purple (or purple and white) and black and white (or black). See what resistance is read. If you're lower on gas, fill it up and check to see if the resistance value has changed.

[slanej]

Archon:

I have one of those el cheapo Radio Shack pen-type auto-ranging multimeters. I suspect its sufficient for the job at hand. But just one question for my own edification. Do you know offhand what the lo and high resistance across that circuit SHOULD be? I'd like to tuck the answer away in my little "book of secrets."

Thanks in advance.

E.

[Archon]

That meter should do fine. I do not know what the actual resistance values are. The higher the fuel level, the higher the resistance should be.

[slanej]

Archon:

In your last post where you said fuel GAUGE pressure sensor, did you mean fuel TANK pressure sensor (i.e., the topmost sensor when looking down on the fuel pump assembly)? If so, two things:

(a) I found out that with a full tank the resistance should read 250 Ohms, 50 Ohms at empty.
(b) I test across the leftmost and rightmost leads, correct? I can't see them inside the connector collar, so I'm just poking around until I get a reading.

E.

[Archon]

Sorry, I looked back and did not see anywhere that I said fuel gauge pressure sensor. In my first response, I mentioned the code was for the fuel tank pressure sensor.

On the top of the fuel pump, there should be two connectors. One is for the fuel tank pressure sensor, the other is a combination for the fuel pump and the fuel gauge sender. Look for a purple and a black wire. When you disconnect the connector, measure the resistance between the pins on the pump where those wire connect.

[MrPOPE]

I have the same crazy fuel guage problem.... i replaced the whole pump and it was not resolved. i read somethere the fuel guage itself in the dash has been known to just go.

[Archon]

They have been known to be a problem. That's part of what we're trying to determine here. If the sender in the tank is working properly, the problem likely lies in the dash.

[MrPOPE]

How hard is it to replace the whole dash Unit if the fuel guage is broken. I have seen them for about 75 bucks on Ebay.

[Archon]

This should give you a good idea on how to take things apart.
http://www.pontiacbonnevilleclub.com/fo ... dash#p1642

You also want to contact Andres (Harofreak00) as he rebuilds them.

[slanej]

Archon:

For the life of me I can't get the test leads onto those pins for a resistance check. It's too tight a squeeze for my sausage-like fingers. I'm stumped, flummoxed, frustrated and just generally PO'd. I had my gearhead friend check the O-ring again and he says its OK. But I'm still getting the P0452 code. I replaced the fuel tank pressure sensor (cheap as it was) and, well, nada.

I read somewhere that you could pull the fuel pump and manually test the sender/float assembly against the dash gauge readout, but I'm reluctant to start this (or any) car with an open fuel tank just for a test. Ten years ago perhaps, but not now. Maybe my wife is right and I'm getting wiser with age.

But we're still poor and we don't have the money to replace the fuel pump. In fact, I'm not convinced that the pump is what's wrong anyway. Hell I can still drive the beast. Must be getting gasoline to the engine. I just have to keep ahead of the fuel situation.

What's next? I'd hate to think that the problem is the gauge in the dash - that's even more expensive than the fuel pump!

What about the wiring for the pump/sender unit? I noticed that the black wire with grey stripe looks kind of crimped just behind the connector. Is there a way to check the voltages that might lead me closer to a resolution?

I've gotta believe there's an enumerated fault tree somewhere online that could lead me in the right direction. Perhaps you can point me to it?

And, just so you know - and for anyone else reading this thread - I really appreciate the time you've taken with me so far. You've been very patient with this noob. I'm hoping we can trace this thing down and get it resolved without me going to the poorhouse.

Regards,

E.
Krystos Voskres!

[Archon]

I'm not aware of any specific fault tree for this problem. Even GM's servicing disk has little information regarding it.

If you can't get your meter's probes in to check the resistance of the fuel level sender, we'll work in the opposite direction. Get a relatively low resistance variable resistor. Using your meter, mark where the 50 ohm and 250 ohm points are on it when you turn the shaft. Start at 50 ohms, turn the ignition to on. Does the gauge read empty? Turn the shaft toward 250 ohms, and observe the gauge. Does it rise toward full? If this test works properly, your gauge and wiring are OK and the problem is with the fuel level sender. If not, the gauge or wiring is at fault.

Has the code for the fuel tank pressure sensor been reset, and then has come back? If so, and you're certain that the O-ring is sealing properly, it may be time to look at other places for leaks. Filler tube, for example.

Andrew (Harofreak00) rebuilds the dash for a reasonable cost if you need to go in that direction.

[slanej]

Archon:

When you say "variable resistor" do you mean a potentiometer connected on the center pin and one outer pin?

[Archon]

Archon:

When you say "variable resistor" do you mean a potentiometer connected on the center pin and one outer pin?

Yup. It would be best to use a linear one.

[slanej]

That's what I thought. Now all's I gotta do is find a nice 1 kOhm pot. I had an old Radio Shack (bring back my 1980's!) 5 kOhm pot but I tried it and it's not particularly accurate and very touchy to adjust. My HVAC guy thinks he may have a rheostat that'll work FOC.

Once I get the right pot/rheostat, I'm going to link it into the outbound connector (that hooks up to fuel pump using the purple and black leads, then test from there (keep in mind you're corresponding with a dummy, here)?

Meanwhile the P0452 code went away when I replaced the fuel tank pressure sensor. I cleared the code and I've driven about 60-70 miles without it coming back. So that's (hopefully) another thing out of the way. Let me get another 60 miles on it, and then I'll be ready to mark that item a success.

Thanks again for all of the help. I'm trying to map all of this out for future reference to all of the Bonneville die-hards out there.

E.

[MrPOPE]

I was over at the Pontiac Aztek Board... and they have the same exact problem. Those guage sender motors crap out at a high rate in pontiacs.

[smittyingr]

Any news on a fix for this? Im dealing with the same gauge problem. Put in a different pump that was told was good and still have the problem. Not sure if the sender on the new pump was good though. Guy was pretty shady. Lesson learned. I tested my old pumps sender and was getting wierd readings from 6 ohm at empty to 300 ohm at full and various numbers that didn't seem right throughout the floats path. ???

[95SLE]

OP:
I had the same problem two times and it was the gauge itself both times. The first time I replaced the cluster. The second time I had the entire cluster rebuilt. I like the cluster rebuilt option the best. The service replaced all the motors and lights in the cluster.

FWIW the SLE cluster is the hardest to find, probably next to the GXP, which was only available for 2 years.

I have had nothing but great service from Haro on any parts I have bought from him. Dependable quick shipping, packaged well and the parts were as advertised.