Bad fuel gauge readings\"fix\" The post below got me to actually attempt fixing the erroneous gas gauge readings I was getting: ful sender fix writeup My problem was twofold...gauge would read just under half a tank when actually only about a 1/4 tank remained. Full was actually full though. The other problem was that the gauge would bounce like mad when I hit a bump, or any time the fuel was obviously sloshing around in the tank. So today I dropped the tank to replace the sender. Then I remembered the above post, found it, and decided I had nothing to lose attempting to fix it, with the price of these senders. (TBI in my case) Pulled the sender out, and started playing with the float while it was wired up, but out of the tank. In a couple of spots during the floats travel, the gauge would bounce around, then settle down once I passed those spots. Decided to swap the gauge with another one, since I don't have the interior put in yet. (still lol) The "bounciness" of the gauge was gone, and it now went lower than 1/4 tank, although not to E. The gauge that was "bouncy" ALSO responded to float changes immediately, whereas the other gauge I have was much slower to respond to float level changes. Obviously the gauges are "dampened" somehow, but no clue how thats done...probably not worth fixing for the normal big fuel gauges anyways since they are so common. Anyways, once I determined part of the problem was the gauge, I decided to play around with the sender I now had out. For anyone that hasn't pulled one out, don't be afraid of the sending unit. It is not complex at all. I was thinking of taking pictures, but there really is no point. They are too simple. Once I got the torx screws out, I realized the very fine wrapped wire was broken in a few spots about where the springs hold the piece it is wrapped around. (these are at the extreme nds of the wrapped wire "section) If I'm not mistaken, this wrapped wire is called a rheostat. Anyways, since the gauge still worked to some extent, I figured the wire being broken there was not a problem. I did notice that most of the electrical stuff on the sender had some form of corrosion on it. Not having a scotch brite pad, I used 0000 steel wool. Wouldn't be my first choice, but it is very "soft" and worked ok. Had to clean some of the stuck fibers out afterwards, but not tough. I've never used contact cleaner, but anything that will clean electrical contacts and copper is highly recommended before this job. (Laugh if you will, but I realized back when I was in the food service industry, that tomato paste will make copper shine like you wouldn't believe. Didn't have any around, or I would have tried on these pieces, as they are very fine, and hard to reach in some cases. Just rinse clean after a couple of minutes!) Cleaning the wire and the contact on the float arm did make some difference, but still wouldn't go all the way to E (but closer) and still above F a bit. Tried to check Ohms at full/empty, but I wasn't getting the 0-90 range you should see. Pulled the thing back apart, then cleaned the springs and the small metal piece that is attached to the sender itself. Upon closer inspection, both the springs and that metal piece are part of the circuit, and both were dirty/gunked up. Put it back together, checked Ohms, now it is between 1-100, which is close enough I assume. Hooked it back up to the truck, and now the "new" guage reads E when empty, and only slightly over full when the float is maxed out. Total success. As a side note, to double check my conclusion that the original gauge was bad, I hooked it back up with the now properly working sender, and the bounciness was gone, no doubt because the dead spots on the rheostat were now gone. However, the gauge still followed the float instantaneously, which would indicate an overactive(immediately changes fuel level as you start up or down a hill, right when you accelerate or decelerate) gauge that still reads correctly when not moving, or under steady flat travel, and is not "bouncy", is more than likely bad. It appears that the sender can be somewhat "tuned" as you can change the angle on the rheostat with the screws/springs (depending on how tight they are), which means that you can SLIGHTLY alter how far the float arm contact "travels" on the rheostat. I may play around with that a bit tomorrow to see if there is enough to be perceptible. I can't see any reason for those springs to be there except for "tuning" the gauge reading, and/or PERHAPS some sort of shock absorbtion, although that seems unlikely, as there is no need for that...by design, there is nothing that bumps will affect that the springs would cancel. Anyone have a pre-tbi sender out of their vehicle? I'm very curious to see if the rheostat is the same for all tanks. As far as I can deduce, they most likely are, which is a good thing of course for being able to swap bad ones, although a tank one should never really go bad. I checked what appeared to be the rheostat "housing" part number on gmpartsdirect.com but it didn't come back as a good number, so I have a feeling these aren't a dealer item either. Anything I missed? Any questions? I hope this helps people that are motivated enough to pull their tank, to try and fix fuel gauge readings, and perhaps helps narrow down where to look first!