JD White

2002 Z06. During the summer of 2017 my car developed a fuel gauge issue and produced DTC 1431 for the right tank and coincided with what I was seeing on the gauge. It would never read full and was inaccurate until it reached 5/16ths full and then was accurate until empty.

I tried the typical Techron fuel system cleaner and had improved readings but it ultimately never fixed it. I caved after a few months and replaced the entire right side siphon assembly in October. My first fill-up following installation correctly read full. During the following couple weeks to consume that tank the gauge level appeared to be accurate. Then, every fill-up after the initial fill-up read a maximum of about 7/8ths full.

After a month, I decided to install the ECS stage 1 fuel kit that I had bought and replace the driver side float & sender while I was there. I did so, and the same problem has remained ever since. I recently tore everything apart again to manually test the float arms and check inside the tanks for any warping or deformity. Manually raising both float arms to their max positions results in 2.5v output on each (I use HPTuners to view the voltages) and the gauge correctly reads full. Using a mirror and light, I did not see any abnormalities in either tank while they were empty. I also verified that the venturi in the siphon is not clogged.

After reinstalling everything, filling it up at the pump and then manually filling the vapor dome from a can to be absolutely certain that the system is full, the left sender will not read anything higher than 2.35v. The right sender always correctly reads 2.5v when full. The gauge reads 15/16ths, which is accurate based on the missing 0.15v from the left sender.

Besides having a tank deformity when fuel is present, I can't explain why the left float will not read 2.5v when the tank is full. I cannot say for certain that this is the same issue as the original issue since I never checked the voltages at the time but it seems that it may be different since replacing the right side siphon assembly did fix it for one fill-up & consumption cycle - or it could have been coincidental that it worked once and the factory siphon assembly was fine all along.

Anyone have any suggestions on where to go from here? It's getting to the point where I want to drop it at a dealer and tell them not to call me until it works!

drseth

Welcome to the club. I replaced the right side sensor twice, but the gauge only reads 7/8s of a tank when full. Both replacements worked at first, but 2 or 3 months later, that 4th or 5th fill-up, the gauge doesn't move all the way to Full.

It is something I have learned to ignore.

wcsinx

Well since you've already verified the reference voltage is correct on full-sweep, why not just tweak the float arm downwards a little so that you detect "full" slightly lower in the tank?

JD White

I had that thought but it's a last resort option since it's just a band aid and will generate a bit of the opposite problem when close to empty.

wcsinx

Disagree on it being a bandaid...

As I said, you verified the full position on the potentiometer is providing 0 resistance thus delivering the full reference voltage signal. So in doing so you have narrowed the problem down to the float not moving to the correct position in a full tank. I suspect if you compared the shape of the float arms on your old and new sending units, you'd find the old one has the float a smidge lower than the new one.

JD White

It's certainly possible that it's slightly different than the original arm since the one time it did work was when the original arm was still installed. However, assuming the tank isn't deformed, does the arm normally reach the top of the tank? If so, then I wouldn't feel that bending it is a band-aid. Strictly based on my observations, it doesn't appear to that it would reach the very top of the tank

wcsinx

In my experience, most floats will "top out" slightly before the tank is completely full. This is why a lot cars have an insensitivity to the first gallon or so of fuel being used before any movement shows on the gauge. I typically drive about 30 miles (1.5 gallons) before my gauge moves off full.

JD White

Good point - but that could be variable depending on design. Some designs don't have the float all the way to the top and any overfilling delays gauge movement and could cause false failures on poorly calibrated diagnostics. Other designs bring the float directly to the top of the tank so it may have both inherent sweep range variability and overfilling delay. For what it's worth, in all of my research on this I've also found that the 1431 DTC diagnostic is delayed 40 minutes from start to allow for the jet siphon to move the fuel from the right tank to the left.

As soon as I read your comment here it reminded me of the updated GM chart of float voltages for 97-early 03 C5s. Specifically for the passenger side they state full is 2.35-2.5v while the driver side is specifically 2.5v. It sounds like they expect the installation of the passenger float could vary a bit. Since that's only at full volume, it's not a big deal that it doesn't move for a few extra miles. Having it on the driver side obviously becomes an issue since it's calibrated to only recognize 2.5v as full. Combined with the manual sweep working correctly, this helps convince me. I'll bend it and see what happens.

JD White

Long-term follow-up on this in case someone else runs into the same issue. This turned out to be ECS blowing smoke. At least in my car, the float assembly was hitting the "hump" in the tank, physically prevent it from being able to rise to its maximum range. I had to cut the sender pedestal and weld in 1" worth of extension plate to get the sender back to it's original position. Thus, the extension distance was equivalent to the thickness of the fuel adapter ring.