Just recently, my gas gauge has been acting strange. I had what I suspected to be bad gas in the tank, so what I did was drain the tank with a siphon pump w/metal fitting on the end. Worked great, I clicked the ignition on a few times to see how much I had drained, read to empty just fine. I took out what I think is practically ALL the gas.

Next time I turned the ignition switch to ON, the gauge pegged out at Full. It has done this every time that I have switched to ON. I figured maybe I jimmied the float or something, but to be sure, I took a look at the troubleshooting guide here at TC.

SO, I took the connector w/ the tan wire off the connector to the sending unit in the tank. I grounded it to the TRUNK pan w/ one of those screwdrivers with a fuse in it, where the trunk has a bare metal spot, and sure enough the gauge moved.

HOWEVER, unlike the instructions for gas gauge troubleshooting, it didn't go down to Empty (I gave it maybe 5 seconds or so- maybe not long enough?). It quickly went down to 1/2 and stopped there, or just a tad under. If I reconnect my gauge via the connector in the trunk, it just pegs out to Full- it doesn't drop any at all. This is what is throwing me off. If the gauge had a bad ground, wouldn't it NOT function at all, whether it had the tank hooked up or the tester hooked up?

Any thing else I should look at? Any suggestions are appreciated! :D

Might try playing around with the wires on the gage. Early gages didn't have a separate ground wire to the gage but relied on tight screws to make the ground. Perhaps a poor ground to the gage or maybe the resistor on the gage backside cracked.

SO, I took the connector w/ the tan wire off the connector to the sending unit in the tank. I grounded it to the TRUNK pan w/ one of those screwdrivers with a fuse in it, where the trunk has a bare metal spot, and sure enough the gauge moved.

HOWEVER, unlike the instructions for gas gauge troubleshooting, it didn't go down to Empty (I gave it maybe 5 seconds or so- maybe not long enough?). It quickly went down to 1/2 and stopped there, or just a tad under. If I reconnect my gauge via the connector in the trunk, it just pegs out to Full- it doesn't drop any at all. This is what is throwing me off. If the gauge had a bad ground, wouldn't it NOT function at all, whether it had the tank hooked up or the tester hooked up?


Yr of car??? There is high resistance in the circuit somewhere, most likely the ground. Hook a long test wire to the tan wire and ground it to the engine block, if it peds to "E" then its a bad ground. Did you check the ground wire off the sender?

Yr of car??? There is high resistance in the circuit somewhere, most likely the ground. Hook a long test wire to the tan wire and ground it to the engine block, if it peds to "E" then its a bad ground. Did you check the ground wire off the sender?


Thanks for the replies! The car is a '70 Nova. It is interesting you mention the high resistance in the circuit somewhere. Could that be caused during my testing by the "probe" I was using? The screwdriver type probe with the fuse in it? I measured Zero resistance in that by putting one end of the OHM meter on the end of the wire from probe, and the other end on the end of the probe that I stuck in the connector.

I did a search here, and found a post about the same sort of fuel guage pegging problem; one of the replies suggested using a fused screwdriver probe. The reason I used that is I didn't want to damage the guage somehow if there was nothing wrong with it in the first place.

If there is no way of damaging the gauge, I will try it with just an unfused wire- both to the spot I grounded it at first, and to the block with a long enough wire.


The odd thing about this all is the fact that the gauge responded, but just didn't respond as it should have according to the tech article.

As for testing the gas tank ground, what I did was take the OHM meter, put one end on one side of the bolt, one end on the other, and it measured 0. This was just as it sat, no power or anything to the sending unit.

As for looking at the ceramic resistor on the gauge, I'll try to get a little mirror or something similar up there to see if its cracked. I really don't want to take apart the instrument panel if at all possible.

Thanks for the replies! The car is a '70 Nova. It is interesting you mention the high resistance in the circuit somewhere. Could that be caused during my testing by the "probe" I was using? The screwdriver type probe with the fuse in it? I measured Zero resistance in that by putting one end of the OHM meter on the end of the wire from probe, and the other end on the end of the probe that I stuck in the connector.
If it has any sort of light on it yes it could. Just use regular wire with a inline fuse.

I did a search here, and found a post about the same sort of fuel guage pegging problem; one of the replies suggested using a fused screwdriver probe. The reason I used that is I didn't want to damage the guage somehow if there was nothing wrong with it in the first place.

If there is no way of damaging the gauge, I will try it with just an unfused wire- both to the spot I grounded it at first, and to the block with a long enough wire.
You can't damage the gauge by grounding the sending wire. A fuse is just there for safety, incase you have the wrong wire, its hot, and accidently ground it. If you don't use a fuse just make sure you have the right wire. The connector in the trunk, the other wires are not normally hot unless the parking lights, brake lights backup lights or a turn signal is on.

The odd thing about this all is the fact that the gauge responded, but just didn't respond as it should have according to the tech article.

As for testing the gas tank ground, what I did was take the OHM meter, put one end on one side of the bolt, one end on the other, and it measured 0. This was just as it sat, no power or anything to the sending unit.
From the sending wire to the sender and the ground wire you should get 0 Ohms if the tank is empty. 90 if its full.
As for looking at the ceramic resistor on the gauge, I'll try to get a little mirror or something similar up there to see if its cracked. I really don't want to take apart the instrument panel if at all possible. These don't go bad very often, all the ones I seen that were bad were from damage after the dash has been taken apart.

If it has any sort of light on it yes it could. Just use regular wire with a inline fuse.


*Edit*

I could have sworn that thing in that probe was a fuse, but it was a light bulb :rolleyes:.. I grounded the gauge like I did yesterday, this time without the bulb, and sure enough it swooped its way all the way down to Empty. So, my gauge is OK.

I guess it HAS to be a ground!? I took a quick look at the tank ground on the frame, and it seemed OK, but I didn't take it off. I suppose I should take that off and make sure its alright.

Can other bad grounds cause this too? Like maybe a bad instrument cluster ground? I have three things hooked up to an underdash ground : tach, inst. cluster, and washer switch. Is this too much? Thanks for the input! I really appreciate it.

*Edit


Problem
|
|
V

Heh- stupid me. You know what?! I think that the thing in that little probe IS a light and not a fuse, now that I think about it! It just looks like a fuse. :rolleyes:.. I'm so used to fuses that look like little cartridges, I kind of forgot what I was looking at- LOL!

I guess it HAS to be a ground!? I took a quick look at the tank ground on the frame, and it seemed OK, but I didn't take it off. I suppose I should take that off and make sure its alright.
Yes, take it off, clean the frame and the ring. You need a metal to metal connection. It doesn't take much corrosion to cause problems


Can other bad grounds cause this too? Like maybe a bad instrument cluster ground? I have three things hooked up to an underdash ground : tach, inst. cluster, and washer switch. Is this too much? Thanks for the input! I really appreciate it.

3 grounds in one place is fine as long as there clean and tight.

Thanks a lot for the help!

Now that I know the guage is OK, I can figure out what the deal is with the errant readings on it. I am thinking that its either a screwy ground under the dash or at the tank. I'll also have to check the sending unit, but it was reading just fine when I was draining the tank. I hope I didn't smack something on the sending unit. Like I said, I used a siphon pump w/a brass end on it so I could get as much gas out as possible.

i had a problem with my gauge going only to 1/4 when full....then when the fuel level hit 1/4 it would work properly to empty...lol, i lived with this for a couple years, coz really, it was adaquate, but recently i had the dash apart and checked the connections and resistance of the tan wire, which was working fine...well, imagine my non surprise when i found the power wire was only getting like 3. sumpin volts at any given time (funny, thats just about 1/4 the proper voltage it should be getting) so...i got the power up to full voltage and it works fine, so if everything else checks out...dont forget to check that...ive never seen this prob covered in the forums, so here it is, good luck

Thanks for the tip!

I did some more checking yesterday. I figured a ground is a ground, right? If the tank grounds to the frame, and it is OK, and I found a good ground in the trunk to use, its the same thing, right- ? So I cleaned up BOTH the ground wire on the frame for the tank, the spot under it, AND the inst. cluster ground. I then put the positive end of the Ohm meter in the tank wire connector coming off the sending unit in the trunk while grounding the neg wire to the ground I was using in the trunk. There was no reading.

So what happens now when you ground the tan wire back at the tank?

yes!...do tell, we're on the edge of our seats here, what happened?

Edge of your seats...LOL! :D

Well, When I did ground the tan wire (single one coming FROM the sending unit) at tank the gauge does OK. It swoops on down to Empty with no issue. Like I posted in an earlier post, the probe with what I thought was a fuse was actually a light bulb! LOL. That was what was causing the 1/2 reading at first. Its an old school kind of probe, and the bulb looks just like a fuse from the car.

Anyway, I also tried to measure the resistance of the sending unit by putting the + probe in the tan wire connector (sending unit), and the - wire on the metal spot that I found in the trunk. There was no reading....Since the tank is EMPTY- it should read 0 ohms, right?

I figured since the gauge WORKED fine when I ground it to the trunk (and with that light bulb adding resistance, oddly enough), that this spot must be a good ground. The ground for the TANK is on the Frame (body, but same thing, I guess), so the spot in the trunk is as good as grounding the other probe to the actual sending unit ground, right?

From the sending wire to the sender and the ground wire you should get 0 Ohms if the tank is empty. 90 if its full.

You are doing this with the wires disconnected from the gauge (in the trunk)? The tan wire (sender) and the black wire (ground).
Anyway, I also tried to measure the resistance of the sending unit by putting the + probe in the tan wire connector (sending unit), and the - wire on the metal spot that I found in the trunk. There was no reading....Since the tank is EMPTY- it should read 0 ohms, right?
Yes, it should read O ohms if empty, but your no reading statement confuses me. What kind of meter are you using? Digital or analog. (analog meters have a needle) Zero ohms is the same thing as putting your 2 probes together. No reading is max ohms for your meter, such as when the probes are not touching anything. Post back

Greg U,

In testing the SENDING UNIT, I took the SINGLE (tank connector, right? It comes up from the trunk) TAN wire OUT of the DUAL Connector. I put the + probe from the ohm meter in the end of the single connector, and grounded the - ohm meter probe to the good ground I found in the trunk. As I stated before, I figured a ground is a ground, right? On a Nova, the ground from the tank grounds to a little area of the trunk pan (kind of) right above the pass side corner of the gas tank. I really cleaned up that area, as well as the rings.

I have a digitial multimeter, and it reads 1 (it looks like l ) when I do this test. In other words, it reads the same thing when I DON'T touch the probes together.

If I DO zero the meter by touching the probes, it then reads 0.00.

BTW- I saw on that occupations thread that you work on a stitcher? My dad worked on a Muller Martini stitcher for a LONG time. Printing is a neat industry, I have been around it and at times worked in it also :D (not on that scale, more like quickie print places).

Your car is 48G Forest Green, right? I like it :D . That is what color my Nova was (will be).

Thanks for the help!

I have a digitial multimeter, and it reads 1 (it looks like l ) when I do this test. In other words, it reads the same thing when I DON'T touch the probes together. Then current is unable to flow through the sender, the reason the gauge won't work. Something in the sender is open somewhere, either a broken wire or the windings in the sender. You may have to take out the sender for further testing, before replacement. If you do replace it, make sure to test it first, read this post (http://www.chevelles.com/forums/showthread.php?t=94345) The 1 is probley a | and means infinity, which is more ohms than the meter is capable of reading.


If I DO zero the meter by touching the probes, it then reads 0.00.
This is zero ohms and is what you should be getting with the tank empty.
BTW- I saw on that occupations thread that you work on a stitcher? My dad worked on a Muller Martini stitcher for a LONG time. Printing is a neat industry, I have been around it and at times worked in it also :D (not on that scale, more like quickie print places). I run a Sheridan/Harris FG stitcher

Your car is 48G Forest Green, right? I like it :D . That is what color my Nova was (will be).

Thanks for the help! I'm not sure of the exact color and the former owner couldn't remember, he had it painted in 1980.

I never dropped a tank before, but at least its not a gigantic gas tank.

I was really hoping that the ground to the gauge was bad and could be fixed easily. Dropping the tank is NOT happening until A) It cools down outside (if that EVER happens :mad: ) and B) I have the time to do it right. I haven't added any gas to the tank since I drained it- its been H O T, and I don't want it to sit. It will be worth a check again when I put some gas in the tank..Maybe its a "range" problem like hrd was having?

At least I know that if I think the gauge is bad, it can be tested across its range by grouding the wire to get E, AND using a probe with a bulb in it to get 1/2 tank :D !

For now, I'll just have to use the old pen and paper, and multiplication/division skills to keep track of when its time to get gas....LOL! Thank GOD its not my only car. Then I would be FORCED to brave the 115 degrees and get out there to do it :eek: