Shop car context: VIN 108689 sat in dry Texas storage for roughly fifty years with fuel in the tank. That combination — long sit, dry climate — produces a very specific kind of contamination, and it is not the same problem as a tank that rotted in a damp barn. Everything below happened on this car.

§1 · The question that decides everything

Every tank thread on every forum collapses into the same argument: seal it or don’t. The argument is unresolvable in the abstract because it depends entirely on a diagnosis nobody makes explicitly.

Is the tank contaminated, or is it failing?

A contaminated tank has varnish, residue, and light surface oxidation. The steel underneath is intact. A failing tank has scale, flaking, delamination, pinholes — steel that is actively leaving. These need opposite treatments, and treating the first like the second is how good original tanks get filled with liner that later peels into the fuel system.

So the first job isn’t cleaning. It’s finding out which one you have.

Fuel tank in place in the front luggage compartment before removal Top of the fuel tank showing filler neck and sender mounting
Before anything comes apart. The tank sits in the nose above the compartment floor — photograph the strap routing, line runs, and sender wiring before you disturb them.

§2 · Getting it out

Drain first, at the tap, into something you can inspect. What comes out is your first piece of evidence — colour, sediment, smell. Ours came out as expected for fifty-year-old gasoline: dark, varnishy, unmistakable.

Brass fuel tap and line at the tank outlet, seen from underneath
The tap and outlet line. Brass fittings of this age will round off if you're careless — correct-size flare wrenches, not an adjustable.

The sender comes out before the tank does. On the A it’s a swing-arm float on a top-mounted flange, and it lifts straight out once the screws are off and the wire is disconnected.

Do this while it's out: put a multimeter across the sender and sweep the arm slowly from empty to full, recording the resistance at each end. Published figures for 356 senders vary meaningfully by type — one range circulating for the later sender is roughly 6 Ω empty to 200 Ω full, while the earlier bottom-mount type is quoted around 107 Ω to 2 Ω. Those are not interchangeable, and a perfectly restored gauge reads wrong against a mismatched sender. Measure yours rather than inferring it, and hold the number until you can check it against whatever your instrument is calibrated to. We are doing exactly this, and the figures will be added here once confirmed.
Fuel sender removed from the tank, swing arm and float visible
The sender out of the tank: swing arm, float, top-mount flange. Check the float for fuel ingress — a float that has taken on fuel sinks, and the gauge reads empty on a full tank.

§3 · Diagnosis, on the ground

With the tank out and drained, look inside with a bright light and a phone camera on a stick. You are looking for one thing: does material come loose?

Ours didn’t. No loose rust, no scale, no flaking. What was in there was varnished gasoline residue and light surface oxidation — a dirty tank, not a dying one.

Fuel tank removed, sitting on the driveway, top view Fuel tank removed, showing the underside and flaking exterior paint
Out and on the ground. The exterior paint is flaking, which looks alarming and means nothing about the interior — the two surfaces lived completely different lives.

§4 · The process we used

  1. One week vinegar soak. Fill it, cap it, leave it. Household white vinegar. Mild acid, slow, and it does not attack sound steel the way aggressive acids do.
  2. Water rinse. Thoroughly, repeatedly.
  3. Sun dry at around 80 °F.
  4. Vacuum the interior.
  5. Air hose until the interior is genuinely dry, not nearly dry.
  6. Mineral spirits swished throughout to displace any water that survived steps 3–5.
  7. Drain.
  8. Fogging oil coating the interior.
Read the vinegar. Ours darkened considerably over the week. That colour is the contamination coming off, and it's the most useful feedback in the whole process — a soak that comes out nearly clear means either a clean tank or a soak that isn't working, and you want to know which before you move on.

The water-removal steps are not optional padding. You have just filled a bare steel tank with an aqueous solution. Every one of steps 2–7 exists to get that water back out before it does in a week what fifty years of gasoline didn’t.

§5 · Why we didn’t seal it

We left the interior as bare steel. No liner.

The reasoning, in order:

  • The diagnosis didn’t call for it. Sealer solves a failing tank. Ours wasn’t failing.
  • Sealer is effectively irreversible. Removing a failed liner from a 356 tank is a genuinely miserable job, and a liner that fails does so by peeling into the fuel system.
  • It obscures the original interior. On a car where the argument is preservation, coating an intact original surface is the same category of decision as replacing a good dial face.
  • Modern liners and ethanol have a mixed record. Formulations have improved, but the failure mode when they don’t hold is severe.
Where the honest disagreement lives: plenty of experienced restorers seal every tank on principle, and their argument isn't stupid — bare steel plus modern ethanol-blended fuel will oxidize, and a lined tank is a solved problem you never revisit. The counter-argument is that you have replaced a maintainable condition with an unmaintainable one. We chose maintainable. If your car is going to sit for years between drives, or you can't get ethanol-free fuel, the calculus genuinely shifts toward sealing.

§6 · What we signed up for

Choosing not to seal is not a decision that ends when the tank goes back in. Two obligations follow, and pretending otherwise would be dishonest:

Fogging oil is a preservative, not a coating. It has a shelf life. If the car sits before fuel returns — and on a restoration, it will — the interior needs re-inspection and probably re-treatment. Put a date on the calendar rather than trusting memory.

Ethanol restarts the clock. Bare steel and E10 will begin oxidizing. Ethanol-free fuel or a stabiliser stops being an enthusiast preference and becomes standard practice for this car.

§7 · What was underneath

Removing the tank exposed something the tank had been hiding: localized corrosion in the compartment floor beneath it, including several isolated perforations, with solid steel around them. Trapped moisture, over decades, in a spot nobody could see.

Front luggage compartment with the fuel tank removed Corrosion in the compartment floor beneath where the tank sat
The compartment with the tank out, and the corrosion underneath it. Right-hand photo is underexposed — a better one is coming before this area is stripped.
Honesty note: this repair is not done and we're not going to pretend it is. The plan as originally written — strip, epoxy prime, seam seal, paint — does not repair a perforation. Primer bridges nothing, and seam sealer over a hole traps moisture behind it and hides the problem from the next person to look. The method gets chosen after stripping shows the true extent, and it will be documented here either way. The more important question is why water sat there at all: repair the floor without addressing drainage and tank padding, and this comes back.
If you take one thing: pull the tank early in a restoration even if the tank itself is fine. What it's hiding is the reason to look.
Parts used Affiliate links · they fund the research
White vinegar
Enough to fill the tank — ours took several gallons. Household strength, nothing exotic.
Hardware store
Mineral spirits
Post-rinse displacement of any remaining water.
Hardware store
Fogging oil
Marine two-stroke storage fogging oil. A preservative, not a coating — see §6.
Pelican
Fuel sender gasket
Replace on reassembly if the original is hardened. PN to be confirmed against the invoice.
Stoddard

Where this guide comes from