Fuel sending units read WRONG time after time!

Hey guys. Hoping someone may have some insight into this issue.

72 Barracuda. From what I can tell, stock tank. It had what appeared to be a stock sending unit in it when I got the car, and it was toast.

I bought a new one from Rock Auto....got it and found it didnt read right. Like it would read 1/2 tank when the tank was full. bent the float arm and after numerous trials and errors got it to read right at half and full. But found it had numerous dead spots....the gauge would go to empty for a few miles and then go back to the correct level.

Well, thinking it was a cheap sender I sent it back and bought a "better" one from a well knowm Mopar restoration outlet. The unit had a larger price, about $50 more as I recall....the unit LOOKS to be better quality for sure. Ok, so installed that and found that it read a half tank when the tank was full.

Many more trials and errors , which includes draining the tank to half so I can pull the sender , and I once again have it reading properly at half and full. But this time I find I still have two dead spots...one at half and one at 1/4 tank. I confirmed that it was the sender by running several tanks of fuel through it. Every time, same spots, same results...drops to empty. A few miles later and it goes back up. At least it was predictable.

I brought this to the attention of the vendor and he suggested I return it sinice it wasnt right.

Well, I have the new one in and once again, I have a half tank of fuel in it (8.25 gallons) and the gauge reads just barely off of empty.

Before I go thru yet another round of drains/bending/refills, does anyone have any ideas here?

I have checked the sender out of the tank, hooked up to the wiring/gauge.....when the float is all the way down the gauge (stock dash) read empty.....all the way up, it reads full. Just like it should.

The tank appears to be stock and it does hold just about 16.5 gallons or so.

The ground strap is new....and connected good. I even ran a second ground wire and ran it straight to the battery to see if that would change things....no change at all.

I for the life of me cant understand why I am having this weird issue.

It seems like the sender is the wrong one or something for the tank.

But I have looked and the same sender unit is specified for all years of e body cars.

Anyone? I am at wit's end!

I posted this at Moparts and have a thread going on Cuda-Challenger....I really need to get to the bottom of this....I know this is an A-body forum but you guys seem to know a whole lot about Mopars in general so hoping maybe someone has seen this issue before.

Thanks in advance....

If it reads correctly at empty and full travel outside of the tank, then it must not have the correct travel for the tank. Measure the sending unit from the mounting flange to the float in the full drop position, then measure the depth of the tank from mounting flange to the bottom. The measurements should be close to each other. If they aren't then the travel is not right for the tank. I would think the vendor or someone out there could give you the measurements and angles of a known good stock unit. Unfortunately both my cars have fuel cells and I don't have any idea what the measurements might be.

have you tried to run a new wire from the sending unit to the gauge, just tie wrap it for now, nothing perfect, just to see if the wire to the gause is not damaged somewhere, just a thought, I tried it on mine, I think my problem was bad grounding to the dash or something to do with the dash, going thru 2 or 3 sending units, grounding the tank, and everything you did, sounds to me that the tank end is good, also if you take the sending unit out again, try hokking it up the the wire and going up and down with it looking at the travel

I'm having the same issue. I've purchased 2 x senders now and with full tank of fuel it only goes up to half way on the gauge in my '76 Duster. Why the hell is that?

Mine has been doing this for years. I never bothered to figure it out. I asume it has something to do with the gauge or the lead or ground to it. Maybe a bad ground to the dash.

Start with a steady reference voltage. See http://www.allpar.com/history/mopar/electrical2.html

I guessing the problem is in the dash, not the senders. You cant recreate every condition in park that you would create on the road. In other words, what you thought was a dead spot in the sender was momentary circuit failure caused by bumps in the road, etc..

is your fuel gauge stock and never changed?

if you are using an aftermarket gauge, just know that chevy / ford use different ohms for the tank sender, so an aftermarket gauge will read wrong every time unless it's one made for a mopar sending unit in the tank.

if you are using the stock dash gauge, something could be wrong with it.
you tested it out of the car, but that is using external wiring. maybe the wiring harness behind the dash has some kind of short...

Logan, Thanks for that link! I think I see a project in my future!

I ended up soldering a small wire from the arm of the sender to the body.....it was suggested to me by someone else and it worked...

More info here in a thread I posted a few years back with the saga...

http://www.cuda-challenger.com/cc/index.php?topic=67680.0

great job!

I better probably clarify a little though -

When I say it worked, it did....mostly. LOL

When I fill up the tank, it reads full.

When it says there is a half tank it was pretty close.

When it reads 1/4 tank, though, not so sure it perfect.

When it hits 1/4 tank, I can get 10.5 or so gallons onto the tank. I was working with the idea that my tank holds 14 gallons. That would be close.

Well, I recently came across info that says my tank is an 18 gallon tank. Not 100% sure that is true or not.

That throws off the calculations obviously if it is.

For whatever reason, once I get below 1/2 tank it is off as far as accuracy, but rather than telling me it has more gas than I have , it says I have less......I figure that is like a built in "reserve" for when my gauge says "get gas" but my brain says "lets just go a little further before stopping for gas"...LOL

But this was as close as it got after many tries adjusting the arm....adding the wire made a huge difference in accuracy.

Now, to make things interesting, I will be swapping in a new tank soon.....same sender.....will have to confirm that things are still about the same....LOL

I guess that little bit of unknown is what makes these cars fun! :glasses7:

FWIW - I did check the wiring between the tank and my gauge and it was fine as far as ohms.

First step would be to find out exactly how much your tank holds.

Get a nice sealed fuel can full of 7 gallons of gas and run the car empty.

Fill it with tracked amounts of gas.

For example, run it really low to where you expect to run out some time soon, then dont drive too far away from a gas station. When it runs out you put 7 gallons in it and drive 10 miles to the nearest gas station and fill it up. What ever the numbers are just note them down.

Total in from gas can minus total estimated consumed during the 10 miles to the gas station (maybe 1 gallon) + amount to fill tank = your total tank volume.

Ive done this with every car i've owned just to find out how large the tank is and how much fuel I have left after the gauge reads E.

Rice Nuker said:

First step would be to find out exactly how much your tank holds.

Get a nice sealed fuel can full of 7 gallons of gas and run the car empty.

Fill it with tracked amounts of gas.

For example, run it really low to where you expect to run out some time soon, then dont drive too far away from a gas station. When it runs out you put 7 gallons in it and drive 10 miles to the nearest gas station and fill it up. What ever the numbers are just note them down.

Total in from gas can minus total estimated consumed during the 10 miles to the gas station (maybe 1 gallon) + amount to fill tank = your total tank volume.

Ive done this with every car i've owned just to find out how large the tank is and how much fuel I have left after the gauge reads E.

Click to expand...

Actually, I had measured my current (assume original) tank when I had it down and empty. Added 5 gallons to the empty tank to get me to the gas station, about 3 miles away, and added another 9 gallons. That is why I have been figuring 14 gallons. But like I said, according to the books my tank holds 18....LOL

When I install my new tank current one has a slight seepage) I plan on doing the same thing, and crossing my fingers that I do not have to play with the damn sender again....that was such a pain to calibrate it.

Well.. I didnt read thoroughly, It sounds like its working reliably and pretty accurately too. Nice fix. Im checking it on the other site now. Great thing to have in an old car. That's a great looking car btw.

Did you end up getting a VR for the dash or did you make one or just decided it wasnt the issue? I noticed the link to the BWD R307 and wondered if you bought that..

If you're using the old gauge and a replacement sender, that's your issue. The stock gauge is linear and the new replacement senders are non linear and they will work incorrectly. It sucks because it still happens with an autometer gauge

Fst73_360 said:

If you're using the old gauge and a replacement sender, that's your issue. The stock gauge is linear and the new replacement senders are non linear and they will work incorrectly. It sucks because it still happens with an autometer gauge

Click to expand...

Well, I am not sure what you mean by that.....

I used an OEM style sender.....when I pulled the old original one it looked the same as what I was putting in.....??? So not sure....but if that is the issue I guess that explains it.....

But wouldnt an "oem style" part be the correct type?

So I had some similar problems awhile back and someone showed me the flat stock strap that is supposed to run from the outlet tube on the sender to the metal fuel line. If it isn't in place, the little hose piece that connects the two tubes keeps that ground from occurring. If the sender was changed that little strap may have been discarded and the system doesn't work dependably. The gasket on the sender should keep the ground from occurring there. Sometimes you catch a ground and sometimes you don't. You have one wire to the sender - but no ground that allows the flow through the unit without the strap, or a ground wire that attaches the two pieces, or runs from the sender tube to a frame ground. Just a thought...

shadango said:

Well, I am not sure what you mean by that.....

I used an OEM style sender.....when I pulled the old original one it looked the same as what I was putting in.....??? So not sure....but if that is the issue I guess that explains it.....

But wouldnt an "oem style" part be the correct type?

Click to expand...

if you want laymans terms, the electronic signal in the gauge is not the same as the one in the sender. Do a search for yourself " mopar linear and non linear fuel sender

str12-340 said:

So I had some similar problems awhile back and someone showed me the flat stock strap that is supposed to run from the outlet tube on the sender to the metal fuel line. If it isn't in place, the little hose piece that connects the two tubes keeps that ground from occurring. If the sender was changed that little strap may have been discarded and the system doesn't work dependably. The gasket on the sender should keep the ground from occurring there. Sometimes you catch a ground and sometimes you don't. You have one wire to the sender - but no ground that allows the flow through the unit without the strap, or a ground wire that attaches the two pieces, or runs from the sender tube to a frame ground. Just a thought...

Click to expand...

Yep, I have that...one of the first thinbgs I played with (folks here and elsewhere suggested it).

FST73_360 -- would like more info on this . I had one of the replacement senders apart and it was setup exactly like the old one........basically a rheostat-type of thing...a coil stretched between two poles and an arm that rubbed against it.

Some folks were saying that the coil in some of the aftermarkets was distorted and lo and behold, that is what I found too....


I guess the "linear" versus "non linear" has to do with how the coils are spaced I guess?

I still do not understand why aftermarket manufacturers would change this - it doesnt make sense to alter the way a replacement part like this works when the same gauges will be used with them......:banghead:...it doesnt make any sense. LOL

I have had this problem for a long time. I put in a new sending unit as the old one was rusted off. .

On the test bench I may have found a fix for this problem. But until I get the speedo working I can not install the permanent fix. The parts cost me less than $5. I will let you know in a few days if I get it fixed permanently

Please let us know if your fix works.

90 percent of the time it is a ground issue.i have found it is best to ground the sender to the body ground the gauge cluster to the body and ground the body to the engine.you can leave the factory ones on but you-must-complete the circuit.

Mr.FT said:

Please let us know if your fix works.

Click to expand...

I put it in and SNAFU. I must have hooked up something wrong, because it did work on the test bench. I will look at it again with fresh eyes tomorrow. I am pretty confident that the problem is not with the sending unit. will advise

Edited Tuesday the 20th,
Ok someone figure this out for me. I have my modification installed and the cluster is hanging out of the dash. I am showing 36 Ohms on my ohm meter coming from sending unit. I carefully put the cluster into the dash and the resistance drops to zero. I have tried it many times and it happens every time.
I am off to the garage to figure out something else. If any one thinks of an answer to why it changes, leave a post and I come in at least every hour to check the board