dzywicki

I hooked up my VCDS cable this morning and did another recording of the fuel level sensor resistance and fuel volume left. To do this, open VCDS, go to component 17 (Instrument cluster) and then open measuring block 26. This will show fuel level sender 1 resistance and tank volume.
I was hoping someone could help me with a diagnosis and do a recording of your car and share it. Here is a link to my results. you can view it through VC-Sope after it is recorded (https://www.dropbox.com/s/ne6rpgblzn...x-xxx.CSV?dl=0)

For comparison purposes, I have about 5/8 of a tank left. The resistance swings up when going around turns (that makes sense), but it doesn't seem to come back tot he same level. It almost seems like one of the floats is getting hung up on something. I have been in the tank several times now and haven't seen anything obvious.
I think my next experiment will be to disconnect the float on the drivers side and see what the gauge does.

Thanks in advance if someone can share a log recording of their fuel level sender readings form VAGCOM

dzywicki

This problem is officially killing me now. Had some extra time today and decided to unplug the float on the drivers side. Did that and it had no effect on the gauge level or the remaining miles. I didn't have the time to unplug the one on the passengers side today but will try that next to see the effect. At this point, it has to either be the float on the drivers side or something in the wiring from the floats to the dash.
The part that is really confusing me is that both floats seem to read in the same range. The start reporting a resistance about halfway up they motion and then read until they are all the way up. When the tank is low, both of the floats seem to be below the level where they have a resistance reading.
Checked a friends 2014 A4 today and his fuel level sender readings had values for fuel level sender 1 and 2. On mine is shows 0's for sensor 2. Makes me suspicious of that second float now.

dzywicki

I think I'm getting closer. I decided to just remove the float on the driver side float to see what the effect was. Got down to under half a tank it seemed to start reading normally with just the passenger side float. Put in enough gas to bring it over half full it went right up to full. It seems like the resistance piece in both floats is the same and now I have an extra so I'll try swapping in the old one that seems like it was good or disconnect the passenger side one and reconnect the drivers side one and see what happens there. I can't tell if the two floats work together at the same time or if one only reads 1/2 tank up and the other reads 1/2 tank down. Pain in my ***, but I hope I'm getting closer here. I have looked all over the interwebs and have not seen any other thread describing this problem. I see lots of people with a 0 reading when the tank is full but not this one.

kelisko

I just found this: C6 - 2 fuel tanks.... - CorvetteForum - Chevrolet Corvette Forum Discussion

Read it well. Maybe the fuel cleaner I suggested earlier was the actual solution to your problem. Quoting Joe Lynch:

"A rather common problem on the C5s occurs when the computer checks the level difference between the two tanks. If the difference in the readings exceeds the tolerance set in the computer, you get a check fuel system message and the gas gauge goes to zero. This can occur when the tank level gauge potentiometers corrode some and change the reading slightly, usually at about 1/2 tank. The cure is to put a little Techron additive in the tank, refill, clear the message, and continue driving. Happened to our C5 once, happened to 3 cars out of 10 on a club trip one weekend.

First time it happened to me, I thought I had broken a fuel line since the gauge indication suddenly went to zero.

Near as I can tell, the computer tolerance set in the computer is just a little too tight for the "normal" variations that can occur in the level sensor readings. Replacing the sensors does not help--the problem occurs again as soon as they change some. The Techron additive seems to prevent it from happening again. I just put a bottle in every 6-12 months, no problems after that. And I carry an extra bottle to present to new C5 club members on their first trip with us."

dzywicki

Definitely interesting reading. there are two very different kinds of floats floating around out there. One has the variable resistor part exposed to fuel and I think those contacts will corrode over time. the kind of variable resistor in these floats is sealed. There is a little magnet that moves in an arc over the resistor to give the reading. That resistor piece is sealed and I don't really see what techron would clean in this system. I will try it by getting some techron and soaking the float directly in it since I have the float completely out. I did some testing yesterday and the float seems to read fairly smoothly in the range where it reads. And the two floats read in the same range, but in opposite directions. So it definitely seems one reads 1/2 down and the other reads 1/2 up. Seems like a pretty simple system and has got to be a simple problem here somewhere. Only other thing I can think of is there is a split in the sealed resistor unit that does something weird if fuel gets on it and gets inside. I thought I had a picture of the other side of that resistor unit, but I don't seem to.

kelisko

You never know. Give it a try. But I don't think soaking the float directly in the concentrated solvent is a good idea. I'd mix it with some fuel instead. Be cautious though, and wear gloves.

dzywicki

Been busy rebuilding a Jeep Cherokee, coaching basketball and flag football and haven't had much time to deal with this fuel gauge issue lately. I removed the driver side float and definitely see that it primarily affects 1/2 way full and up. I had an extra drivers side float from my replacement. The resistor is the same on both sides, but the orientation of the float is different. I swapped the resistor, soldered the plug on and put it back. It seemed ok for a day and then started the same thing with going up. I got frustrated and unplugged the connector from the fuel pump/sender and saw that the gauge went to full. After I saw that, I went through the connectors and pressed then in with a small screwdriver, spread the wires out and then pressed the connector on and pushed the the inner connectors really hard to ensure there was a good connection.
That seemed to do the trick and the gauge seems to be reading normally now. I think what is happening is the small connectors were losing contact or had a break in the wire causing an intermittent break. That causes the gauge the go up like when there is no conneciton.
Just wanted to add this to my post in case someone else ever comes across this. I have searched all over the interwebs and main problem I see is the gauge reading 0. I guess when the connections are bad in these cars, the gauge goes up.
I still don't see how fuel cleaner would do anything to these floats. It's a small magnetic piece with no physical conneciton to the variable resistor. The only way I could see a chemical cleaner doing something is if there was a leak int he variable resistor and then I would guess you'd have different problems.

dzywicki

Final update to this saga. The fuel gauge has been behaving as it should. No more gauge going up. I was not able to determine if the problem was in the connections to the fuel pump/float piece or in the wires on the backside of the connector. Earlier in the thread I posted a picture of the wiring harness. The whole harness is bent and shoved under the metal support so the cover can go on. It must be either the connections were periodically losing contact int he plug or one of the wires has a break that I was able fix by straightening and moving the harness. With as many times as I removed that plug, I am leaning towards the actual wires themselves. I'm not touching anything until it happens again. Keeping my fingers crossed that I am done with this project for a long time.

dzywicki

Hopefully this is the real final update to this saga. The fuel gauge seemed to work normally for a few weeks and then went back to being wonky. But it was a different wonky. when the tank was more than 1/2 full, it seemed to read double what it should and when less than 1/2 full would read 1/2. Not 100% since I couldn't really tell how much fuel was in the tank.
Finally got fed up with it and went back in this weekend. The smaller connector was being a real PITA to come off. I finally ended up prying out one of the push points and it let it release. I had a spare and I pried off the same piece slid it back in when I replaced it and it held fine. Fought with that for 2 hours just trying to get the connector off.
I had checked the resistance at the two float connectors at the connecter and it seemed like the one for the main tank wasn't always giving me a resistance reading. I ended up taking the fuel pump basket out and hitting the connectors with a soldering iron to reflow the soldier. I put it all back in and my gauge that was reading close to a 3/4 tank was now reading just over 1/2. this was a good sign. Added ~5 gallons and my gauge went to 3/4 tank. Seemed to take awhile to get there but I suppose it'a always like that to account for tank sloshing while driving.
Going to fill up and then watch as the level drops. I had not been driving much the last 2 months (been great motorcycle weather in Northern VA) so it didn't bother me as much. I sure hope this is the last iteration because I am out of ideas if this doesn't work out. I had read about "micro fractures" in some connections in some VW/Audi forums. Seems far fetched, but may be a possible condition.
For the record, after taking this all apart several times, it still makes no sense to me how fuel injector cleaner could change any behavior with the floats. They are sealed and pretty low tech.

kelisko

Could also be a missing software update apparently: https://www.audiworld.com/forums/a6-.../#post23886674