Looking for some advice here. I have a 2009 Z06 I put twin turbos on. I developed a controller for a fuel pump not available in the US yet, so I have had the tanks out a few times. I eliminated the crossover mess. My tanks are connected via a dash 12 line at the bottom. Cool thing is I can drop tanks in 20 minutes, bad part is now the gas gauge rarely works.

I don't understand why GM made this so damn complicated, all they needed to do was average the signals and call it a day. My question is, how can I make the computer happy and why is it unhappy? I've read the tanks equalize when not running, so my signals should basically be always sending equal amounts like on startup.

Any advice on this? Thanks!

 

The passenger side tank has a fuel level sender as does the driver side. The system is designed to syphon all the fuel out of the passenger side tank until it is empty. At that time the fuel gauge will read approximately 1/2 tank. The drivers side level sender then takes the fuel gauge the rest of the way to empty.

BJK

 

Thanks man, but I knew that already. I was looking for a bit more in depth view of why the computer is not happy right now seeing identical signals coming from the senders. I know they both send some % of the 5v source back depending on level.

I really thought GM was just averaging the signals, because that logic would work for every scenario and give an accurate over all fuel reading, but that is obviously not the case.

 

Didn't he say one empties first, duplicate that and problem solved

 

What is your level sender? What codes are stored?

 

Here are photos of the driver's side fuel pump with sender and float, and a closeup view of just the level sender. Note: The close up view is for a sender used in '07 and later. The '05 and '06 senders have the green wires attached at the bottom of the circuit board as shown in the fuel pump assy photo. They are not interchangeable.

BJK

 

It looks like GM's algorithm is that if the level sensor in the passenger tank reads a fuel level, then it is added to the capacity (9 gals) of the drivers side tank. If there is no reading from the passenger tank, it then shows the level of the drivers tank.

Why not average the 2? Maybe dealing with the sloshing of one tank was easier than dealing with 2, while trying to provide an accurate fuel gauge.

 

Did you determine that by experiment? What happens if you substitute 150Ω resistors for both level senders? If it's summing the decoded signals, the gauge should show a little below 1/2 tank. If it's doing what you describe, it should show 3/4 tank. If it's programmed to recognize that combination as evidence of a bad sender and/or failed venturi pump, it should throw a code.

 

Good idea, I will check the code and see if that provides some insight. When I eliminated the crossover I really assumed they were just averaging the signals, but I suppose it's looking for a more complicated scenario like you said to detect failed components. I may have to program a chip to emulate the OEM behavior, but I'd like to avoid that of course

 

How do you have your level sender(s) wired up now, and what's going wrong? Gauge at zero, gauge showing the wrong level, CEL, etc.

 

You would of been better off if you had installed a low volume pump inside the passenger tank. Have that one pump into the drivers tank. Then all would be happy. If the passenger tank stays full and the drivers is low, You will not be able to refill the drivers tank. It has a shut off to keep from getting fuel in the evap. lines.
I don't understand why most won't install a higher volume pump and splice into the factory fuel setup. The factory pump will still work correctly and drain both tanks. Seems pretty simple to me and will support large hp applications.

 

They are hooked up as stock, sometimes it would read and most of the time no. The gauge goes to zero and the read out says check fuel system / low fuel.

Yeah I see what you are saying. In a perfect world I would have rather done that. When we took the tanks out, the main reason I joined them at the bottom and eliminated the crossover was because I literally didn't think we could get the crossover hooked back up without dropping the transmission, that thing is a biatch to work with if you have ever touched it! This was on the end of a year build, so I opted for the easy way out. It is quite nice I will say, but I'm paying for it with the gauge!

On a side note, the pump is working great. It sits in the stock holder and is supporting 1000+ RWHP on e85.

Thanks for everyones input, not sure what I will do at this point.

 

There are a few different zones that GM uses. If it falls out of the correct zone it will throw on the check engine light and default to empty on the gauge. Same thing if the right tank is full and the left tank starts to empty.

 

Contact SinisterC6. He has a long thread about his build by Vengeance Racing. They removed his fuel tanks and provided another. You may get some help on how they got the fuel gauge to work.

http://forums.corvetteforum.com/c6-f...ce-racing.html

http://forums.corvetteforum.com/1584497047-post453.html

http://forums.corvetteforum.com/1584497774-post457.html

BJK

 

If you look at his video, The fuel gauge is on empty. They might not have worked on the fuel gauge issue yet.

 

Yeah, I saw that, but that was the first time the car was started. I doubt they wanted much fuel when everything is new. None the less, contacting them to see how they resolve the issue can't be all that bad.

BJK

 

Wow interesting build, pretty neat. Thanks for the link. I may get in touch with them. If nothing else I can just utilize one side signal only and use half the gauge, better than nothin!