Open Menu
Open Menu

Getting an old hot-rod hero back into service: repairing a P-51 Mustang fuel gauge

Published in

The fuel gauge on my Ford A/B speedster is from a World War II P-51-B Mustang fighter plane. Unfortunately, it stopped working just in time for the car’s maiden run.

When you’re building cars—especially hot rods—you never really know what you’re going to get into.

Take my Ford A/B speedster, for example. With a modern aftermarket body like those produced by Mercury (no relation to Ford), Laurel, and Morton & Brett in the Twenties for the Model T, and a full complement of found components, ranging from a Moto-Vox “pancake” horn to a cooling shroud from a B-17 Flying Fortress’s 50-caliber machine gun that I’m using as a spreader bar, there’s ample opportunity to dive headfirst down a rabbit hole when you get to tinkering around.

Currently on the workbench is the car’s fuel gauge. I had wanted something that didn’t require an electrical sending unit, something simple, durable, interesting, and fairly inexpensive. What I found, believe it or not, was a gauge from a P-51 Mustang’s fuselage auxiliary fuel tank.

With its upright horn shape (like a ship’s ventilator in miniature), it was perfect for its location over the pilot’s left shoulder, and it’s likewise well suited for its station atop my car’s fuel tank behind the seats. I had the tank modified so it could be positioned just to the right of center and be seen in the rearview mirror by the driver.

Fuel gauge installed to be viewable in the rearview mirror by the driver. Note, though there is still work to be done on the speedster, it has been out for a few spins in the mud and snow. But that’s another story.

When I purchased the gauge a few years ago, all worked as it should: Manipulating the float arm swung the needle. But the build has taken longer than expected, and, alas, when I finally filled the tank with fuel, the gauge continued to read “E.”

My guess was that, as we had done nothing to adjust the arm and float for a tank that is shaped and sized differently from the one it was intended for, it was not lifting. I figured, worse-case scenario, it was brushing the side of the tank and not floating, and I’d simply shorten the arm.

Not including the float arm, the body of the unit measures 16-inches long. The float is comprised of compressed cork washers and is varnished to prevent absorption of fuel.

So, as part of my incremental winter-night workshop strategy, the plan was to remove the assembly one evening and assess the situation, then I’d make the necessary modifications another night, and put it all back in and calibrate it on the third.

But when I unscrewed the six hex button-head cap screws (not original to the gauge) and lifted the unit out, I immediately realized that the problem was not in the action of the float. The needle did not move when I lifted and lowered the arm. And down that rabbit hole I went…

In keeping with my work strategy, I should have quit at that point for the night, but I wanted to get an idea of what I was up against so I could work on it in my head until I could get back at it tomorrow.

I did hesitate a moment before beginning to take the gauge apart. After all, it is a 75-year-old historical artifact, the product of the efforts of the mighty American war machine that defeated the Axis. Breaking it would suck.

But, for the same reasons I purchased it to begin with–its simplicity and durability–I decided that I would give it a go. The alternative would be to relegate it to ornamental status, a state not worthy of its connection with the legendary P-51 Mustang.

As I studied the gauge, I quickly realized that I had been correct when I bought it. Having been built for war-time service, and short of a direct hit–which, given its location inside a fuel tank, would suggest the pilot had bigger problems–it was designed to continue to do its job no matter what. And should it in fact be damaged, it could easily be repaired, or cannibalized to fix another fuel gauge.

Flipping the gauge over reveals a ring with two holes. As there appears to be no other way to access the mechanism inside the gauge head, I assume that the ring is threaded and that inserting a punch or an awl into one of the holes and leveraging against it will allow the ring to be turned. I’m right; once started, it moves rather easily. This is feeling like a Japanese puzzle box: Just pay close attention, think methodically, experiment carefully, and take pictures.

After the threaded ring has been spun off, I cautiously separate the gauge housing from the tubular base, expecting to see some kind of mechanical connection between the bevel gears and the stuck needle, but there is none. 

Once the mechanism is removed from the gauge housing, you can see that it uses a magnet to move the needle. You can also get a clear idea of just how easy servicing was intended to be: The magnet is painted red on the tip that aligns with the point of the needle, and every single component has a simple set screw, allowing it to be quickly removed, replaced, or reused–absolutely essential qualities of equipment operating in a combat environment. I move the float arm, and all parts of the mechanism mesh well and work smoothly. By process of elimination, the problem should simply be that the needle is stuck on its spindle, likely due to a little corrosion after so many years.

While my five minutes in the garage had turned into 25, it was time well spent. My goals for the night had been to remove the fuel gauge and identify why it wasn’t working. I had done both.

Once back in the house for the night, I do what I always do: research the hell out of it. And while I find no technical information on the gauge, I do learn that there was a real story behind why my gauge exists at all.

The P-51 had been tasked with protecting Allied heavy bombers that cruised at high altitude during their precision daylight raids on the Axis, but its performance was lackluster at altitudes above 15,000 feet.

A veteran pilot with the Royal Air Force speculated that swapping the P-51-A’s 1,200-hp, 1, V-12 Allison engine–with its single-stage supercharger–for the 1,620-hp, 1, V-12 Rolls-Royce Merlin 60 Series power plant–with its two-speed, two-stage supercharger–would help.

It did, and in 1942, the result was gains of 250 horsepower, 50 mph, and a service ceiling of nearly 42,000 feet.

This cutout depicts the Mustang-D variant, with the most obvious difference from the A, B, and C models being the teardrop glass cockpit that afforded the pilot 360-degrees of view and would become the type used on modern fighter aircraft. The P-51D, too, had an auxiliary fuselage tank, though it is not evident in this advertisement. Image courtesy of

And that brings us to my fuel gauge. The Merlin, built under license in America by Packard, was 250 pounds heavier than the Allison, and North American Aviation’s engineers decided to rebalance the center of gravity of the aircraft and increase its range by adding an 85-gallon auxiliary fuel tank in the fuselage aft of the pilot. That’s right, this gauge only exists to be installed on my hot rod because folks in ’42 decided to hot rod the P-51 Mustang.

Unlike the main fuel tanks located within the wings and utilizing gauges set into the deck of the aircraft, the auxiliary fuselage fuel tank was located behind the pilot and required an upright instrument (inside yellow box) to be seen. The simple, all-mechanical float system would have proved highly reliable in a combat environment where it was likely that an aircraft would sustain damage. Image from the U.S. Army Air Forces manual provided by World War II Performance

For take-off and climb-out, the flight manual instructed pilots to set the fuel-selector valve to draw from the fuselage tank because it “is the most direct system to the engine and is on a higher plane in relation to the engine. Use of the fuselage tank fuel will also move the C.G. [center of gravity] of the airplane forward to a more desirable position for flight.” Once at a safe altitude, pilots were to follow a procedure of fuel management that would require them to look over their left shoulder to reference this gauge.

Now, back to the repair of my gauge… I bring it in to work the next day to fiddle with on my break, and it looks like the only way to access the needle is by removing the glass lens.

Again, being cautious and taking time to observe and test, I’m hoping, will ensure I don’t damage anything needlessly.

I put my finger gently on the glass and attempt to move the lens around, which it does. So, it seems like it is held in place only by a metal compression ring.

I insert the tip of a flathead eye-glass screwdriver into the space and pry upward, being careful not to grind against the edge of the lens, and the ring begins coming out more easily than I expect. I had anticipated requiring needle-nose pliers, but I’m able to do so with my fingers.

With the lens out of the way, I am able to manipulate the needle, which sticks at first, but with a little force it’s freed. Nothing secures it to the spindle, so I easily lift it off, revealing that both, in fact, have hints of corrosion. Some gentle work with 1500-grade sandpaper cleans them up, and a drop of graphite oil ensures smooth, frictionless movement.

Thinking ahead to potential future repairs, I trace the lens on a piece of paper as a template should it ever break, and I label it so I don’t forget in the intervening years what it is. Then, I clean it and, because this gauge was never intended to be exposed to the elements, run a bead of clear silicone around the rim it seats on in the housing to prevent rain water from getting inside. Reinstalling the split ring is just as easy as removing it. 

Before reassembly, I put a few drops of the graphite oil on the bevel gears and the bushings.

Now, two tests loom on the horizon: 1.) Does operating the float arm move the needle? and 2.) Once on the tank, does the gauge provide a reasonable indication of the quality of fuel inside?

The first test is answered quickly. As soon as I work the mechanism back into the housing, the needle automatically spins and aligns itself correctly with the magnet.

With that one out of the way, I spin the ring that fixes the housing to the tube and float arm back into place, using the awl to get it tightened down.

Back in the garage, I sink it back into the tank, but don’t run the screws in yet (what if it doesn’t work?). The tank is empty, so pouring in 4 gallons of no-ethanol 91 octane will give me a known quantity to work with.

Test #2 is also a success! The repaired fuel gauge reads 20 gallons. I’m looking forward to filling the tank to its full 16-gallon capacity to test out my theory that the relationship between the amount of gas and the reading on the gauge will be a consistent 1:5 one: 4 = 20, 8 = 40, 12 = 60, and 16 = 80. It’s convenient, too, that though the P-51’s tank held 85 gallons, the gauge is only marked up to 80, making for quickly read 1/4 tank increments.

Finally, remembering that I only have Permatex Ultra Black RTV on hand, and that silicone-based sealants don’t play well with gasoline, I run to the auto parts store and pick up some Motoseal Ultimate Gasket maker.

Donning nitrile gloves, I squirt a thin bead of the gasket maker on the surface of both the flange on the tank and the one on the gauge, then use my gloved finger to spread the grey goo out evenly before mating the two surfaces.

You can just barely see the fuel gauge peeking up over the passenger’s seat.

But, rarely does working on a car come off without a hitch, or changes of opinion.

After I have smeared the liquid gasket all over both the flange on the tank and on the gauge, I realize that having done so would make any future gauge servicing just a little more difficult.

Getting mad at myself is a waste of energy. So… as much as I’d like to be able to call this project done, and as little as I want to go through the messy process of wiping off all the grey goo, I do so, carefully. To ensure that the ring will still unscrew easily in the future, I run it all the way out and clean the threads. All so tedious and avoidable.

I replace the liquid gasket with one I cut from a piece of Fel-Pro rubber-fiber gasket material, available in that company’s #3060 variety pack.

To ensure that the cap screws won’t vibrate lose, I apply a drop of Threadlocker Blue 242 (which, note, ironically comes in a red tube) to the threads of each.

While I’m predisposed toward over engineering and would like to use the high-strength Red Threadlocker on something as fairly permanent (and valuable) as this gauge, the need to heat the screws should I ever have to remove them from the fuel tank to above the flashpoint of gasoline is, um… “unappealing.” The Blue will make sure they stay locked down and help keep honest people honest.

And this is what I love about working on cars, especially hot rods… They present so many opportunities to learn history and new skills, but they do so in such a unique way.

As I repaired this fuel gauge, I couldn’t help thinking about the men and women in the factory who built them, the air-field mechanics who maintained them, and the pilots who relied on them.

Now, in a very small way, I have shared in that rare experience of those people in the past, just as we all do when we work on and drive old vehicles.

Keep ’em flying.