Street News

Leveling Out: Deconstructing the Mystery of the Incorrect Fuel Gauge

Lead

advertisement for Steeroids

Imagine this scenario- after a long, hard winter, the weather has finally turned pleasant again. The sun is shining, the birds are singing, and even the grass has begun to green (which really means you have to mow soon, but we are just trying to set up a theme here).  The weather is so nice you decide that you just have to run into town with the top down on the Impala (or Ford, Mopar, whatever you got). You fire it up and cruise into town. You glance down at the fuel gauge, reading a nice 1\4-tank, so you pass the first gas station. Then you get a whiff of that peculiar smell that can only mean one thing- you just ran out of gas.

If you own any older car, you know the situation. The gauge reads 1\4-tank, but the tank is bone dry. You put 5 gallons in the 18-gallon tank and it reads beyond full. Sometimes you bought the car that way. On your way out of the guy’s driveway he stops you and says “Hey, by the way, fill ‘er up before 1\4 tank, otherwise you’re empty.” Thanks for telling me that AFTER I paid you, buddy.  The bottom line is you have a problem. Something between the gauge and the fuel tank is not working. Nobody likes messing with the gas tank, you start throwing money at the problem with a new gauge (you might even try hanging an aftermarket gauge below the dash), but is that going to fix the problem? You need to take a little extra time and diagnose the situation before you spend a dime. It is very easy to pinpoint the problem if you take the time to do it. Otherwise you’re just out of gas.

There are three potential causes- gauge, sending unit, and wiring. Each area should be tested in order to find the problem. In addition, every car is different. Most GM cars are very simple, just the gauge and the sending unit, but many early Fords and some Mopars have voltage regulators at the gauge itself. Voltage varies by the manufacturer, model, and year, but for vehicles with a fuel gauge voltage regulator, it should be around 5-volts. IF the car does have a voltage regulator (this will require a little research for your particular car), then the temperature gauge (not dummy light) will be tied to the same voltage regulator. If the temp gauge is working but the fuel gauge is not, the regulator is not the problem.

Moving along the chain, we start in the dash. Unless you want to get real familiar with the underside of the dash, it is a good idea to pull the gauge panel. This gives you better access to the back of the panel. You don’t have to take the panel out of the car, but if you do, the tests will be easier. If your panel has a printed circuit board and bundled plugs, then you need to determine which wires are for the fuel gauge. More research for you. Testing the gauge is easy. You need 12 volts (run through the regulator if so equipped), a ground, and a second ground lead. We used a battery charger on the bench to make things easy. A fuel gauge is in reality just an ohm-meter, it measures resistance. GM vehicles use 0-90 ohms almost exclusively (every muscle car era GM is 0-90 ohm), where the gauge reads empty at 0 ohms and full at 90 ohms. Ford uses two ranges, either 73-10 ohm or 240-33 ohm, it could be either one. The 73-10 ohm senders are non-linear, meaning full at 10, half at 25, empty at 50 and below E at 73. The 240-33 ohm senders are linear and more accurate, reading 240 empty and 33 ohms when full. Knowing what your car has is critical to this test.

To test the gauge, you need to assemble a series of resistors (available from Radio Shack) that total the full level and the empty level. For GM that is easy, all you need is 6 15-ohm resistors wired together in series (end to end). For the other ranges, you need two, one for full and one for empty. When pairing resistors, a simple series configuration adds the resistance numbers together. You may even be able to find a single resistor that fits the mark you want. You don’t have to be exact. A 250 or 220 ohm resistor will work for the 240 range. A potentiometer or rheostat may also work, as long as it covers the full range.

Testing the sending unit is easy, but because most sending units are only accessible from the top of the tank, the tank must be removed or at least dropped from the body. Sometimes you can get to it without fully removing the tank, but for any GM A-body, you may as well settle in for the full job. The sending unit is easy to test. You need a multi-meter set to ohms and that’s it. You will need to pull the sending unit from the tank during the testing, as you need to be able to move the sending unit lever.  The sending unit should display the required ohm reading at empty and full.

With a little patience and some diligent diagnostic work, you can get your fuel gauge reading accurate again. If your gauge is shot, you need to either replace it or have it rebuilt. Auto Instruments offers rebuilding services that range from basic “get it working” to full-on restoration with fresh screen printing, chroming, and restoring plastic bezels. Repairing the gauge requires special calibration and should be left to a pro. Year One sells new reproduction gauges (as well as new sending units) for many models. Replacement sending units are available for most vehicles, your local parts store might even stock them, and are drop-in easy. Nobody likes running out of gas, so get your tools out and do something about it.

01.We set up a battery charger for the power supply on the bench. To keep the clamps from touching, they were clamped to the edge of the table. The yellow test lead runs to the positive cable, while the green and white leads attach to the ground cable.

01. We set up a battery charger for the power supply on the bench. To keep the clamps from touching, they were clamped to the edge of the table. The yellow test lead runs to the positive cable, while the green and white leads attach to the ground cable.

02.We made this tester from 6 15-ohm resistors wired in series. With the multi-meter leads on both ends, we can test the gauge at full 90-ohms (we are working with a GM product here).

02. We made this tester from 6 15-ohm resistors wired in series. With the multi-meter leads on both ends, we can test the gauge at full 90-ohms (we are working with a GM product here).

03.By moving the test clamp to the middle of the series, we get 45 ohms, which should read at the half-full mark.

03. By moving the test clamp to the middle of the series, we get 45 ohms, which should read at the half-full mark.

04.We tested a gauge panel we pulled from a 1967 Buick Skylark. Later model cars use more printed circuit boards, which can cause problems too. Look for burnt traces. Notice the white board between the two terminals to the right. This board is wrapped with a thin wire. This wire can burn, break, or corrode. If you have any of these issues, the gauge is probably not working.

04. We tested a gauge panel we pulled from a 1967 Buick Skylark. Later model cars use more printed circuit boards, which can cause problems too. Look for burnt traces. Notice the white board between the two terminals to the right. This board is wrapped with a thin wire. This wire can burn, break, or corrode. If you have any of these issues, the gauge is probably not working.

05.We attached the ground to lead to the ground terminal and the positive to the positive terminal. The positive terminal reads resistance to the ground through the sending unit lead. At this point, the gauge should read well past full. If it does not, you might have the polarity reversed, double check.

05. We attached the ground to lead to the ground terminal and the positive to the positive terminal. The positive terminal reads resistance to the ground through the sending unit lead. At this point, the gauge should read well past full. If it does not, you might have the polarity reversed, double check.

06.The second ground lead, which is the stand-in for the signal wire, was attached to the sending unit terminal. Using the resistor chain we created earlier, we can test the range for the gauge. To get an empty reading, the resistor chain is removed and direct ground attached.

06. The second ground lead, which is the stand-in for the signal wire, was attached to the sending unit terminal. Using the resistor chain we created earlier, we can test the range for the gauge. To get an empty reading, the resistor chain is removed and direct ground attached.

07.With 90-ohms, this is the reading you should see.

07. With 90-ohms, this is the reading you should see.

08.We switched to the middle of the resistor chain, which is 15-ohm three resistors at 45-ohms. Got a perfect half-full reading.

08. We switched to the middle of the resistor chain, which is 15-ohm three resistors at 45-ohms. Got a perfect half-full reading.

09.With the ground directly on the sending unit post, the needle moved to the empty position. This is a fully-functioning gauge.

09. With the ground directly on the sending unit post, the needle moved to the empty position. This is a fully-functioning gauge.

10.Moving to the gas tank, we tested the sender. The sending unit needs a chassis ground to complete the circuit. We first tested the continuity between the sender terminal and the ground. No good. That is the first sign of trouble.

10. Moving to the gas tank, we tested the sender. The sending unit needs a chassis ground to complete the circuit. We first tested the continuity between the sender terminal and the ground. No good. That is the first sign of trouble.

11.We also checked continuity from the ground tab to the end of the ground wire, which was good.

11. We also checked continuity from the ground tab to the end of the ground wire, which was good.

12.The sending unit must be removed for the next step, so using a hammer and a prybar, we drove the ring retainer out of the position. There are three tabs, you need to hit each one, just smacking on one makes it more difficult to remove.

12. The sending unit must be removed for the next step, so using a hammer and a prybar, we drove the ring retainer out of the position. There are three tabs, you need to hit each one, just smacking on one makes it more difficult to remove.

13.Then the sending unit was removed. You have to snake the sending unit, filter sock, and float out of the tank.

13. Then the sending unit was removed. You have to snake the sending unit, filter sock, and float out of the tank.

14.We set the meter to ohms (200 range) and touched one lead to the ground wire and the other to the bottom of the sender terminal, nothing.

14. We set the meter to ohms (200 range) and touched one lead to the ground wire and the other to the bottom of the sender terminal, nothing.

15.Next we tried the top of the sender itself, where the solid wire strap connects the sender to the terminal. Here we got a 104.6 reading (which would be above full). This means the connection between the sender and the terminal is bad.

15. Next we tried the top of the sender itself, where the solid wire strap connects the sender to the terminal. Here we got a 104.6 reading (which would be above full). This means the connection between the sender and the terminal is bad.

16.We could replace the strap, but before going through the trouble, we tested the empty position. This read 24.6 ohms. Even if we replaced the strap, the sender is shot. This would yield about 1\4 of a tank when empty. This is the most common problem with fuel sending units.

16. We could replace the strap, but before going through the trouble, we tested the empty position. This read 24.6 ohms. Even if we replaced the strap, the sender is shot. This would yield about 1\4 of a tank when empty. This is the most common problem with fuel sending units.

17.Another potential problem for the sending unit is the float. Most floats are solid plastic, design to, well, float. Sometimes they are brass or hollow plastic, which can develop holes, allowing them to soak up gas and read empty when you actually have plenty of gas.

17. Another potential problem for the sending unit is the float. Most floats are solid plastic, design to, well, float. Sometimes they are brass or hollow plastic, which can develop holes, allowing them to soak up gas and read empty when you actually have plenty of gas.

18.Replacing the sender is simple. Here are two different senders. The bottom unit is a stock replacement sending unit we picked up from Year One. This will install just like the stock one. The top unit is from AutoMeter and is a 240-33 ohm Ford unit. This is a universal sender (for a Ford gauge) and requires drilling the tank and bending the arm to read correctly. This is also good for custom tanks.

18. Replacing the sender is simple. Here are two different senders. The bottom unit is a stock replacement sending unit we picked up from Year One. This will install just like the stock one. The top unit is from AutoMeter and is a 240-33 ohm Ford unit. This is a universal sender (for a Ford gauge) and requires drilling the tank and bending the arm to read correctly. This is also good for custom tanks.

19.We tested the new sending unit on the bench. We set the arm in the full position (which is slightly below its maximum) and measured it. 89 ohms, it actually went up to 109 in the maxed-out position.

19. We tested the new sending unit on the bench. We set the arm in the full position (which is slightly below its maximum) and measured it. 89 ohms, it actually went up to 109 in the maxed-out position.

20.In the full down position, it read 2.8. Like we said earlier, the gauge is analog, so the reading won’t be perfect. 2 or 3 ohms off is fine. 20 is not.

20. In the full down position, it read 2.8. Like we said earlier, the gauge is analog, so the reading won’t be perfect. 2 or 3 ohms off is fine. 20 is not.

21.Another sore spot is the sender terminal. These items are incredibly hard to find, so don’t lose it. If you have to rewire the car, save a generous amount from this to create a lead.

21. Another sore spot is the sender terminal. These items are incredibly hard to find, so don’t lose it. If you have to rewire the car, save a generous amount from this to create a lead.

22.If you have to, a female spade terminal will work, but not in stock form.

22. If you have to, a female spade terminal will work, but not in stock form.

23.Using a small flat-blade screwdriver, we opened up the spades and then crimped the terminal onto the sender stud. This will work, but be prepared to have to reinstall it occasionally. In other words, try really hard to not lose or destroy the stock terminal.

23. Using a small flat-blade screwdriver, we opened up the spades and then crimped the terminal onto the sender stud. This will work, but be prepared to have to reinstall it occasionally. In other words, try really hard to not lose or destroy the stock terminal.

Sources:

Auto Instruments

http://www.autoinstruments.com/

AutoMeter

http://www.autometer.com/

Year One

https://www.yearone.com/

 

About Jefferson Bryant (196 Articles)
A life-long gearhead, Street Tech Magazine founder and editor Jefferson Bryant spends more time in the shop than anywhere else. His career began in the car audio industry as a shop manager, eventually working his way into a position at Rockford Fosgate as a product designer. In 2003, he began writing tech articles for magazines, and has been working as an automotive journalist ever since. His work has been featured in Car Craft, Hot Rod, Rod & Custom, Truckin’, Mopar Muscle, and many more. Jefferson has also written 5 books and produced countless videos. Jefferson operates Red Dirt Rodz, his personal garage studio, where all of his magazine articles and tech videos are produced. You can follow Jefferson on Facebook (Jefferson Bryant), Twitter (71Buickfreak), and YouTube (RedDirtRodz).

Leave a comment

Your email address will not be published.


*