Tale of the faulty fuel gauge!!!

-
cruiser

cruiser

Well-Known Member
Joined
Feb 26, 2016
Messages
811
Reaction score
900
Location
Bloomington, MN
Okay guys, I'm really stumped here. I cannot get the fuel gauge in the dash of my very stock 74 Duster to work right. When I fill the tank completely, it only shows about 7/8 full on the gauge. After driving for 40 miles, it shows half a tank remaining. The sending unit is new, as is the tank. The gauge cluster was rebuilt by a very good gauge guy, and it has a solid state voltage limiter. I've tried a different sending unit - no help. The new one does the same thing. I've had the original tank sending unit rebuilt, and it does the same thing too. I've replaced the ground strap on the tank with a new one. Grounding the sending unit wire to a body ground with the ignition on sends the dash fuel gauge to somewhat past full. Any ideas?
 
Mine does about the same thing. I haven’t gotten around to troubleshooting it yet, so I would like to follow along.
 
This is a common issue with aftermarket sending units for mopars. The factory unit and aftermarket are designed completely different. Lots of threads on here about that issue.
 
So you think the problem is with the sending unit in the tank, not the dash gauge or wiring?
 
Could be anything, in my case it was the old float (brass) began to take on gasoline and would not float correctly any more.
 
Sounds like from reading a bunch of other threads, that if you have an aftermarket tank sender this is just how it works. It looks like the design of the aftermarket sender is different enough that the gauge no longer works properly. It is unclear if a Meter Match gadget can fix this. I will do some more research.
 
The problem is most likely the sending unit not being designed for the gauge it's driving. The gauge works based on the current supplied by the variable resistor on the sending unit. Variable resistors typically come in two formats, linear taper, or logarithmic taper. With a linear taper unit, the amount of resistance changes the same amount with float height anywhere on the scale. With a log taper, the resistance changes on a logarithmic scale with float travel, so the first three inches might change it by 50 ohms, and the next 3 inches could change it by 500 ohms. Many times, aftermarket sending units will use a different electrical taper than the OEM unit. What this means, is that at a given float position, the aftermarket unit is not supplying the same resistance to the circuit that the OEM sender did, and the gauge reads improperly. This happens most frequently with a sending unit that's "designed" to fit multiple applications. If you can find an OEM sending unit, or one with the correct taper, the gauge will likely work properly.
 
Based on the chart below that I copied from another thread, we can see what Garry is describing. At 9 gallons (about 1/2 full) the mopar sender is about 28 ohms and the aftermarket is about 48 ohms.

That means that at a half tank of gas, the aftermarket sender will read about 3.5 gallons, or less than 1/4 tank.

The graph below shows it visually. As the resistance goes up, the gauge goes down. The blue line goes up much faster than the red line.

If you look at where the blue line crosses 63 ohms (empty on the mopar sender) it would imply that the aftermarket sender would read empty on the gauge at 5-6 gallons left.


upload_2018-12-27_18-30-14.png


upload_2018-12-27_18-35-42.png
 
Has anyone used a Meter Match gadget to correct this issue successfully? Never tried one myself.

Meter Match End point calibrations would likely not solve this problem. However thier literature says you can also pick two calibration points elsewhere on the curve. I suspect you could substantially improve gauge accuracy with this gadget at a cost of about $53 or so.
 
Picture of original and replacement fuel senders. Mechanically a lot different, I believe the electrical design is also a lot different.

upload_2018-12-27_19-17-28.jpeg
 
Tristarrradiator.com rebuilds them.
I found this in an old thread here. There are probably others. Don’t know if you have to have a core.
 
Either sender should report 10 ohms at full and that should take the gauge needle to 'F'. The physical/mechanical difference in the aftermarket sender means the needle will stay at 'F' longer then drop like a hammer to somewhere between 1/4 and 1/2 tank. There is another thread here that shows this difference in float arm length/reach/travel. Anyway... There just might be a problem within your gauge too since you stated that the needle doesn't go all the way to full. Only way to know is check for the 10 ohms at full.
Could meter match module adjust that proven 10 ohms to 8 ohms thus pushing the needle higher? I have no clue. Good luck with it
 
Okay, here's the latest. Just dropped off the factory original sending unit at APT Instruments of Bloomington MN. Very nice guy named Chuck owns the place and has previously rebuilt the gauge cluster in my 74 Duster. We checked the factory sender on an Ohms meter, and the resistance readings were all over the place. He's going to completely refurbish it with new points across the rheostat, and new windings inside as well. I'll report back in a few weeks on how it went. He says that the Chinese and Argentinian ones that are repopped are crap, and he has a whole bunch of experience rebuilding MOPAR instruments, so I tend to believe him. More to follow.
 
cruiser said:
Okay, here's the latest. Just dropped off the factory original sending unit at APT Instruments of Bloomington MN. Very nice guy named Chuck owns the place and has previously rebuilt the gauge cluster in my 74 Duster. We checked the factory sender on an Ohms meter, and the resistance readings were all over the place. He's going to completely refurbish it with new points across the rheostat, and new windings inside as well. I'll report back in a few weeks on how it went. He says that the Chinese and Argentinian ones that are repopped are crap, and he has a whole bunch of experience rebuilding MOPAR instruments, so I tend to believe him. More to follow.
Click to expand...
Old thread revisited. How did the rebuild work?
 
You must log in or register to reply here.
-

Back
Top