Flex fuel and fuel tank inconsistent readings

Car is a 2004 GTO with a P59 PCM, 12587604 OS upgraded to 2 BAR OS with flex fuel enabled. You can see in the log and below that the Alcohol % is all over the place:

- 00:00 time 25.0%
- 00:41 time 35.9%
- 02:17 time 06.6%
- 03:14 time 24.6%
- 04:15 time 00.0%
- 10:25 time 25.0%

Obviously this is messing up trying to scan AFR error as the EQ ratio is all over the place even though the fuel in the tank is the same blend.


Also my Fuel tank level is reporting values all over the place:

- 00:00 time 0.0 gal
- 03:56 time 1.0 gal
- 06:52 time 4.0 gal
- 10:43 time 4.7 gal
- 11:52 time 1.6 gal

The fuel level gauge in the dash shows empty with a very low fuel level when I first start the car. After about a minute or so I get a "fuel gauge error contact dealer" message. Sometimes the gauge will start working normally for a brief period of time before going into error mode again. Unfortunately the gauge does not seem to follow the fuel tank level PID in the scanner.

I'm attaching my tune, log and layout. Any ideas where to start looking?

Thanks!

First think I would check us the BCM harness behind the glove box for frayed/shorted wiring. Very common in the GTO's. As to the flex sensor, where are power and ground for it wired to? Bad ground or noisy power will cause them to act funny

Quote:

---

Originally Posted by 5ft24

First think I would check us the BCM harness behind the glove box for frayed/shorted wiring. Very common in the GTO's. As to the flex sensor, where are power and ground for it wired to? Bad ground or noisy power will cause them to act funny

---

I have already checked the BCM harness and it looked fine, I'll go ahead and give it one more look just in case I missed something.

The power and ground for the flex sensor are tapped into the power and ground for the low coolant sensor, the one at the bottom of the coolant expansion tank on the driver's inner fender.

Did you check the sensor frequency too? It's been my experience that the sensor freq is just displaying raw output as expected. But the alcohol content param is a filtered value based on other values, such as the fuel composition transition/delay values etc.

If you confirm that FF sensor outfreq freq is constant, then you should be able to eliminate it as part of the issue and maybe this is purely a fuel level/gauge issue etc.

Quote:

---

Originally Posted by Pulse_GTO

I have already checked the BCM harness and it looked fine, I'll go ahead and give it one more look just in case I missed something.

The power and ground for the flex sensor are tapped into the power and ground for the low coolant sensor, the one at the bottom of the coolant expansion tank on the driver's inner fender.

---

I don't have wiring diagrams in front of me, so forgive me it this is a dumb question LOL.
Is the low coolant sensor powered by 5V or 12V?
I used the power to the ignition coils and my Flex sensor power and a seperate ground..

Output of the sensor should be 50hz for 0% and 150hz for 100%, so if you have a multimeter that reads frequency, you can check for that as well

Quote:

---

Originally Posted by 5ft24

I don't have wiring diagrams in front of me, so forgive me it this is a dumb question LOL.
Is the low coolant sensor powered by 5V or 12V?
I used the power to the ignition coils and my Flex sensor power and a seperate ground..

Output of the sensor should be 50hz for 0% and 150hz for 100%, so if you have a multimeter that reads frequency, you can check for that as well

---

I recheck my wiring and I have I believe it is wired up correctly:

-Out to PCM pin 56
-GRD to ground wire from low coolant level switch. I confirmed this by measuring ohms with one lead on the cylinder head and the other in the plug.
-Vcc to 12v switched. I have this running to a switched 12v source in the under hood fuse box.

The low coolant sensor is powered by 12v, I tested this key on with the multimeter probes in the plug.

I do have a multimeter that can read frequency. Do I test between Pin 56 and ground? I could not find a PID in the scanner to log sensor Hz. I assume I would at least have to test the frequency with key on, or maybe just supply 12v to it from the battery, disconnect the wire from the PCM and then test the Hz? I'm just afraid of possibly shorting something out in the PCM.

Yeah that's strange it's not in the scanner. I actually have EFILive but it's in there. I was able to set up torque pro to see it as well with a BT OBD2 dongle on my phone too.

Yep, from pin 56 to ground for freq measurement. If it looks stable with it powered on without the engine running, then measure it while the engine is running to see if it's picking up interference

So I measured the frequency and with the the key on and engine off the readings were all over the place, most of them were below 50 hz with some spiking as high as a little over 100 hz. Then after about a minute it settled to 0 hz. I moved the ground to the inner fender well and the readings were similar. I'm wondering if the sensor is bad. Seeing that a new one is not that expensive, I'll order one and try again.

I finally got a chance to replace the sensor and so far the alcohol percentage is showing a stable reading. Granted I was only able to check it at idle and slightly revving as the weather was too bad to take the car around the block. I'll report back either way to let people know if replacing the sensor was the fix.

This time the fuel gauge did not move and the fuel tank level sensor voltage was always 5v in the scanner. Guess I'll try disconnecting the wiring at the fuel pump module and see if I can make the gauge read full with 0 volts. I would assume that if it does, then my issue is with my float or some wiring in the fuel pump module, if not I would need to keep looking in the wiring harness. I still have my old original fuel pump module and I can try plugging that one in as well and move the float arm and see if the gauge moves with it.

I've driven the car a little and so far the sensor seems to be doing it's job. I added some 93 to the tank and the alcohol percentage dropped from 24% to 20%. Readings stayed consistent throughout the drive. Hopefully a new sensor fixed the issue. I know they are not directional, but I did reverse the flow direction on the new sensor after looking at other setups, just in case. On another note, I had to get a special disconnect tool to remove it.

Yeah if you get the compact sensor and depending on the style of connect fitting....huge PITA to remove. I ended up cutting up a normal plastic disconnect tool, wedging it in there and was able to get it apart after like 30 mins of fiddling with it.