Fuel Guage

[jcatnite]

My fuel guage gets a little flaky near empty. It started when I changed fuel tanks and sending units a few years back. I was thinking about calibrating it and found this site: https://www.mgaguru.com/mgtech/electric/fg_04.htm

The only problem is the resistance he is talking about isn't the resistance I'm seeing on the old sending unit on the work bench. I'm seeing a spread of around 240ohms full and around 18ohms empty. I think I can still calibrate it using these numbers but hoped someone could verify them for me. I used a digital multimeter to read the resistance but it should be the same as an old analog one right?

Thanks,
JC

 

[jlaird]

Regularman is our resident expert on fuel guages.

 

[jcatnite]

Thanks Jack,
I was kind of waiting for Kim or Doug to pipe in and give the expert diagnosis...lol. I would like to calibrate the meter without playing the "one gallon at a time" game. If I could build a small test unit similar to the one in the guru's website, it could be passed from member to member that needed it. Although I do have a spare tank and sending unit...
JC

 

[dklawson]

"me Doug" or another Doug?

The info on Barney Gaylord's site (MGA Guru) is for the gauge systems prior to the mid-1960s... before the voltage stabilizer was added to Smiths gauges.

You can use the same basic idea that you found on the MGA Guru site, however, as you noted, the resistance range is different and going in the opposite direction. Also, since the later bimetallic gauges operate a bit differently, the adjustments are a little different. I've only calibrated two temperature gauges to match senders I had. That's not really enough experience for me to explain the adjustments. Hopefully Kim will post soon. I'm sure he's calibrated more bimetallic gauges than I have.

 

[jcatnite]

Thanks Doug,
Yes you are the Doug I was refering too. I once read an article you authored on resistance matching to utilize an aftermarket sending unit. In much I have read regarding the later units, it seems that all the magical adjustments are made at the sending unit. I was hoping to perform some compensation at the meter itself. It seems as if they coils can still be shifted slightly but they are now on rivets instead of studs with nuts. I might just try to pick up a used on on eBay and disassemble it to see what it is made of. You guys might want to hold a memorial for it now. I doubt I will get it back to working order once torn down....lol. Anyway, I've already committed the sin of bending the float arm on my existing unit. Way back when, probably 3 years ago, when I bought a new tank and sending unit, the float bottomed out in the tank before the arm hit the low stop. I probably need to start fresh with a new sending unit as well. Thanks for the advice, I think Kim is out right now but I'm sure he will check in when he gets back.
JC

 

[dklawson]

The bimetallic gauges aren't wired internally like those you saw on the MGA-Guru site. If you look on the back of the bimetallic gauges you will find either two little cork disks or two holes (that used to be covered by cork disks). Remove the cork disks and look through those openings. You will see what appear to be screwdriver slots. Those two slots are used to tweak the bimetallic gauge calibration. As I said, I've only re-calibrated a couple of temperature gauges so I don't even remember which slot changes which aspect of the gauge (zero or span). Hopefully Kim will see your thread and post. If not, I'd send him a PM and ask for his advice. However, once I did stumble on a bimetallic gauge calibration article on the web. I'm sorry... I didn't bookmark it but you may be able to find one yourself.

I had an interesting experience with the resistance matching of fuel sending units. I was contacted by a guy in the U.K. restoring an early (pre-war) Talbot. He could not find a sending unit compatible with his car's original fuel gauge. After looking over his options and what he had I calculated the parallel resistor he'd need to make the correction. HOWEVER, I neglected to tell him that he needed high-watt resistors and needed to put them on a heat sink. The resistor worked... but he had it behind the dash where it promptly got VERY hot and eventually started to smoke. I suggested he try again with a aluminum finned power resistor mounted remotely on a frame rail. Hopefully that fix has given him a more permanent solution.

 

[regularman]

The only way I have found to really calibrate a gauge accurately is to calibrate the sender with the gauge. That way if either is off a bit then that can be corrected out. The new replacement senders are not of the good quality like the originals. I have checked out 2 of them and found them to be way out of range especially on the full side and had to remove some windings to get the resistance correct. Also beware any play in the sender. What usually happen when there is side to side play is that on one end or the other the gauge will read way out when the float goes to the top or bottom of its range because when it hits the stop the play allows the wiper to move off wire. Mine does it sometimes when I fill the tank to the brim. It will read empty until I burn about a quart out of the tank and then it goes to full. I can live with this but will fix it should I ever pull the tank off again. Let me know if I can help at all. What year spridget? That 18 ohms is too high, you got some extra resistance in there somewhere. It should read very close to 0 ohms at that end.

 

[jcatnite]

Thanks Kim,
The car is a 76. The sender I was measuring is an old one that I pulled from the car when I bought it. The PO "woke it up" by putting gas in the tank and starter fluid down the carb. The rusty tank ruined the stromberg and several mechanical fuel pumps. I bought the gas tank "kit" (tank and sending unit) from Vicky Brit as one of the first purchases. Couldn't ever get the carb right so now I"m running an electric pump and a weber. At any rate, I will pull the tank next maintenance period I have and see what each gallon actually equates to on the meter. I might try to find a 0 to 500 ohm pot and put it in line with the meter to see where its zero and span actually lie as well. Mostly just fiddling to try and learn how things work.
JC

 

[regularman]

Thanks Kim,
The car is a 76. The sender I was measuring is an old one that I pulled from the car when I bought it. The PO "woke it up" by putting gas in the tank and starter fluid down the carb. The rusty tank ruined the stromberg and several mechanical fuel pumps. I bought the gas tank "kit" (tank and sending unit) from Vicky Brit as one of the first purchases. Couldn't ever get the carb right so now I"m running an electric pump and a weber. At any rate, I will pull the tank next maintenance period I have and see what each gallon actually equates to on the meter. I might try to find a 0 to 500 ohm pot and put it in line with the meter to see where its zero and span actually lie as well. Mostly just fiddling to try and learn how things work.
JC

If you take it apart over the winter, send me the sender and gauge and I will go through them. I don't think a pot would help if the resistance readings you give are correct because its already too high and adding more wont help.

 

[jcatnite]

Thanks Kim,
I may take you up on that. I wasn't going to add resistance to the sender circuit. I was thinking of using a pot in the place of a sending unit. That way I can chart out the high, low, and midpoint of my existing meter. Then I will know what I need the sending unit to read when I pull it.

Mostly just for experimenting.
JC