Aftermarket Fuel Tank Sending Unit Compatibility

[Fresh Air Inspector]

Hello All,

Had an interesting challenge when I replaced the original fuel sending unit with an aftermarket one; Spectra Premium FG62A (via Rock Auto). I don't know if anyone has experienced this issue but I thought it may save someone time and frustration correcting the problem.

Background
My '88 Caravelle (aka Gran Fury), was in a barn from 2009 until I purchased it in 2022. The previous owner replaced the fuel tank when he pulled the car out of the barn due to the fuel having corroded the tank. Unfortunately, he did not replace the fuel sending unit. :-(

One of the first things I did was pull the tank, installed a new sending unit (original was heavily corroded), adjusted the float on the new unit to match the old one and added about 17 litres (~4.5 US gal.), to the tank. I expected the gauge to register about 1/4 tank, unfortunately it did not. The needle struggled to move above empty. Curious, I added another 22 litres (~5.8 US gal.), to the tank for a total of ~39 litres (10.3 US gal.). The gauge didn't even manage 1/4 when it should have been just over 1/2 (68 litre / 18 US gal.).

Testing
First thing I did was test the gauge by shorting to ground the wire (dark blue in my car), from the tank to the gauge. The gauge immediately went to 'F' indicating the gauge was fine. Next, I measured the resistance of the sending unit - it showed ~33 Ohm. According to the specifications from Spectra the sending unit range is 75 Ohms empty and 10 Ohms full. According to the specs this is correct for ~1/2 tank of full. To say I was puzzled is an understatement.

I decided to experiment and pulled a 33 Ohm / 10 Watt resistor from my box of electronic bits, temporarily placed across the two wires from the sending unit (dark blue and grey in my case). Doing this creates a resistance of ~16.5 Ohm. Interestingly my gauge now shows just over 1/2 which is what I was expecting.

I added another 20 litres (~5.3 US gal.), for a total of ~59 litres (~15.6 US gal.), and the gauge registered close to F as expected.

Conclusion
Testing the sending unit verified that it operated within the stated specifications. Which means the Plymouth fuel gauge was expecting lower resistance values to be accurate. So, beware if you replace the sending unit and end up with unusual fuel gauge readings.

Below is a picture of the resistor I mounted in the trunk for easy access in case I want to substitute a variable resistor to 'fine tune' the fuel gauge readings. The other picture shows the wiring harness behind the rear seat back, drivers side. I tapped into the original dark blue with black tracer (DB / BK) and grey (GY), wires. Solder joints have since been coated and wrapped with electrical tape. Apologies for the blurry image, auto focus on the camera is a little too sensitive.

Hope it helps!

 

[Sub03]

Nice experiment!
Now it would be interesting to see if the gauge shows correct when the tank is empty, because with 75 ohm in parallel with your resistor only gives ~22.9 ohms with empty tank. If I understand correct you measured 33 ohms when the gauge was showing near 1/4.
With 22.9 ohms (empty tank) the gauge will probably be a tad over 1/4.
Some tuning may be necessary to get this right.

Do you know what resistance the gauge is calibrated for?

 

[69-]

Resistor in parallel with empty tank will give around 54 ohms.
(75+33)/2 -> 54 ohms.

That could work.

 

[Sub03]

Have to disagree on that one. The overall resistance in a parallel setup is always smaller than the smallest resistor.

1/Rtot = 1/R1 + 1/R2 = 0.436 --> Rtot = 1/0.436 = 22.9 Ohms.

Or

(75x33)/(75+33) = 22.9

 

[69-]

You are right.
1/ (1/R1 + 1/R2)

My mistake. Then it'll be challenging to get proper values.

Don't they have a formula editor?

 

[Fresh Air Inspector]

Hello All,

Great 'international' feedback. Sub03 - appreciate the compliment.

I have no idea what the fuel gauge input resistance range is. The factory service manual describes how to test the gauge BUT it is with the 'factory' testing tool. A box that has a three position switch; E 1/2 F. It doesn't list any resistor values. :-(

It will probably be the Spring of 2025 before I will be able to test the tank 'empty'. This car is a long term project. I've only just finished rebuilding the complete rear suspension / rear axle. The interior is only half finished and I haven't even started on rebuilding the front suspension not to mention the engine and transmission. I've collected all the bits to do the work except an overdrive automatic ('90 - '95 46RH [A518], is hard to find now).

If feel ambitious I may drain the fuel from the tank and put it in my Volvo - the only other vehicle I have that uses premium fuel (91 octane - the highest available around here :-( Which, unfortunately will limit my compression ratio increase to ~10:1.

If I have any additional information, I will update this thread.

 

[69-]

I built a tester for myself using the following resistors (found that years ago somewhere in the interweb)

Full - approx. 10-13 ohms
Mid - approx 30-35 ohms
Empty - approx 70-75 ohms

IIRC, I measured the gauge on my 69 vehicles some years ago with approx 10 ohms.
Result is, that you'll have 20 ohms (gauge and sending unit in series) on the "approx. 5V rail" (behind the voltage regulator).
According to Ohms law, you'll have 5V/20ohms => 0.25A (amps). And on the resistor you'll have to be able to dissipate 0.625W (watts -> 0.25A * 2.5V) of heat. Use a 1W resistor.

If you want to get rid of the fuel, drain it with a hose to a canister. Unfortunately, it makes no sense to send it to me (would be rejected by customs agents), but I would have been happy to burn it for you...

Fun fact: I replaced the mechanical voltage limiter with an electrical one on my older vehicles (69, to avoid the risk of burning the gauges with a "sticking" regulator, which would feed full 12V to the gauges and burn them in no time...).
Of course, I wanted to do the same with the LeBaron, but that renedered the warning light (LED) useless. It's two-transistor-setup works only with the slow 5V regulator in our cars (switching 12V on-off-on-off like once to twice a second). With real constant 5V it just doesn't work and is lit all the time...

 

[kkritsilas]

you can try a 150 Ohm 1W resistor. It will give you 50 Ohms empty.

Alternative is to use a Pot, say about 150-250 Ohms total resistance, and actually tune the pot to get the gauge to read exactly what you want. You'd need to have it enclsed in a sealed box of some sort. and secure the wiper so it doesn't move around under vibration.

 

[Halifax Hops]

Hope this helps. The NORS gauges use a board that is square for the wiper arm to ride over. The OEM they are triangular. Just the way it is with NORS Usually not right. Here is what it should use. Hope it helps.

 

[Fresh Air Inspector]

Hello KKritsilas,

Good point. The only 33 Ohm resistor I had in my parts box was a 10W. I figured it was overkill. A 1 or 2W resistor should be sufficient. I considered a pot to fine tune the accuracy. All of this will have to wait until the car is on the road - Spring of 2025 if all goes well.

Hello Halifax Hops,

Great piece of information. The factory service manual does not provide those resistor values, it just shows the 'special tool' with F, 1/2, E. Apparently the factory tool can be used to test the engine temperature gauge and oil pressure gauge (if equipped).

With our cars now being 35 plus years old and many parts 'unobtanium' being able to test and possibly repair a component is a huge benefit.

Great example of the benefits of a forum like this.