One way to repair leaking (aluminum) fuel tanks

17 Jul

It’s a bit of a bummer but along with the past two weeks of daily rain I’ve been knocked low with some sort of summer flu and despite being so close to launch I’ve made little progress on the boat these past couple weeks. Looks like I may have to temper my expectations, yet again. I’m sure glad I don’t have a real deadline to work to! In the meantime, here is one of the projects I have been looking forward to writing about. It seems to have been not only an effective repair but one which I have not found a great deal of information on, either on the web or in print. For a those of you just curious about the boat it’s probably a bit long and dry but I hope the lack of brevity will be useful for anyone attempting a similar repair.

My Cape Dory 28, being an early production fiberglass boat, came out of the factory with many effective but unusual features. One of these is the fuel tanks. It is equipped with two cylindrical welded aluminum tanks of approximately 7-10 gallon capacity each (I haven’t bothered to measure them) which rest one to a side on custom fiberglass mounts in the cockpit lockers. In true Cape Dory style these mounts are quite beefy and have stood admirably to the test of time.

You can see the tank mount jutting into the left of the frame in this photo (from when we painted the lockers).

The tanks, unfortunately, fared a bit worse. The issue was the mounting brackets, or, more specifically, the material used to get a tight fit on the tanks. Each tank is held to the mount with two steel straps using some sort of synthetic fabric between the painted steel bracket and the tank. This fabric has a feel similar to that stuff that goes under carpets and, not surprisingly, seems to have held some moisture. Over the past thirty-seven years this fabric created a path for moisture to contact the tank and the end result was that my pressure testing and a close inspection revealed a band of corrosion where each strap contacted the tank. In places this pitting had eaten clean through the aluminum.

The worst of the corrosion. This hole goes straight through nearly 1/4″ of aluminum.

New tanks, of course, are not cheap, especially when they must be custom-made to fit an unusual mount. So I resolved to do everything I could to salvage these tanks. Despite the corrosion and pitting they were in surprisingly good shape for having spent all this time in a damp environment. Even the areas with severe pitting didn’t seem seriously  weakened, just pinholed. But before there was hope of repair, I needed to survey the extent of the damage. I started by cleaning the outsides of the tanks, a relatively easy task.

Pressure testing a nice shiny tank

 Then it came time to clean the insides. This stumped me for a while. Both tanks were gunked up with a couple pounds of solidified diesel which, fortunately, was not the rust I had initially taken it for. This was easy enough to clean up as all I had to do was shake them and pour out piles of brown flakes. The real problem was the emulsified diesel clinging to the sides of the tanks. This stuff did not want to come off! I tried soaking with bleach and Simple Green to no avail. Then I tried scrubbing, which sort of worked except that the only ‘access’ is through the 1.5″ hole for the fill fitting. Next I tried a pressure washer. Again this helped a bit but was largely ineffective. By this point I had the tanks in the backyard of the bicycle shop I work at and had started started rambling aimlessly around the shop looking for anything that might help. My eyes lit on an overflowing drawer of random nuts and bolts and I had a great idea. This drawer was dumped into a tank and chased with with a cup or two of citrus degreaser. Then I rolled and shook the tank all around the backyard until I thought I might pass out from the heat. When I rinsed it out, it was clean as a whistle, as were all those nuts and bolts! In the end it took me longer to get the bolts back out then to clean the tanks. Tip: if you ever try this, only us ferrous metals so you can pull it all back out with a magnet instead of shaking out the stainless ones one-by-one. With nice clean tanks I was very pleased to find almost no interior corrosion – it seemed the only serious damage was the exterior pitting.

With thee tanks cleaned out I called up a metal fab friend of mine and while she agreed to weld them we eventually decided this was a poor idea. Concerns were both the safety of welding around any residue left from the old fuel and worry that areas which were pitted but not actually leaking would continue to corrode.

The next thing to consider was epoxy or, as I sometimes call it, ‘the magic goop’. I did a lot of online research and found that this sort of repair is normally attacked from one of two angles. The most common is to coat the tank from the inside by pouring in a specialty epoxy which is designed to coat interior surfaces as you roll the tank around. This seemed likely to work but I had a few misgivings. The first was that this is a generally used to repair interior corrosion and I was uncomfortable with my seriously pitted tanks being repaired only by a thin layer off epoxy on the inside. Even with a sealed epoxy liner I didn’t like the chance of existing pitting trapping water and leading to further corrosion of the tank structure. I was also worried that the interior surfaces were still not sufficiently clean for a proper epoxy bond. So this idea was scrapped.

 Next I looked at repairs from the outside. The most common technique here is to coat the damaged area with a liberal application of a specialty thickened epoxy. Once again this seemed effective but doubts remained. Despite the tenacity of modern epoxies I read multiple accounts where this mechanical bond had failed over time. The fuel tanks, more so even than water tanks, are an area where I am not willing to take any risks. While BP can get away relatively unscathed after releasing 4.9 million barrels of crude into the Gulf there have been incidences of sailors losing their boats (over cleanup costs) after entirely accidental releases of mere cups of fuel through leaking tanks and automatic bilge pumps.

With this in mind I decided the only repair I was comfortable with was an epoxy and fiberglass coating which covered the tanks entirely, giving both the chemical bond of the epoxy resin to the fiberglass and the mechanical bond of epoxy to aluminum.

I started by ordering a quart of Gluv-It, a proprietary Marine-Tex epoxy meant for sealing and barrier coating, and by reading everything I could find on the subject, including this useful repair report from West System. I gathered fiberglass cloth, sandpaper, and all my other standard epoxy tools (cups, chip brushes, etc).

On the (first) day of the repair I sanded each tank by hand and with a random orbit sander using 80-grit paper while my friend Libby cut out light fiberglass cloth to fit each of the tanks. This was done in three pieces, with circles for the top and bottom and a rectangle to wrap around the body of the tank. From here on I did the tanks one at a time so as not to rush.

The first step was to mix a small batch of neat West System epoxy to wet out the tank. I had earlier called the West System Helpline and confirmed that there should be no bonding issues between competing brands of epoxies. For a superior mechanical bond I wet-sanded the neat epoxy into the aluminum using 120-grit paper. I then mixed up half my can of Gluv-It and added some West System 406 Colloidal Silica Filler as a thickener and strengthener. I painted this mixture, which was the consistency of chilled honey, onto the tank. There was enough for two coats with extra on the bands of pitting. Next I used more West System epoxy to wet out my pieces of fiberglass. The thickened Gluv-It would not have be as effective for wetting out fabric which is really a job for neat resin. This cloth was applied over the still-wet Gluv-It and carefully smoothed down, removing all air bubbles. I then made my first mistake, which was to set the tank on end and leave it for the epoxy to set up.

The repair:

The tank was wet sanded with epoxy and 120-grip paper


Then I catalyzed the two-part Gluv-It resin and mixed in some West System 406 filler
This mixture was liberally painted onto the tank with a chip brush

Before it set I wet out pre-cut fiberglass cloth with West System epoxy
It was applied to the tank

And smoothed down to eliminate air bubbles

Then I left the tank on end to dry – a mistake.

When I returned I found that while the repair seemed effective the curing fiberglass had sprung up in places, creating air bubbles. These were especially prominent at the edges between body and ends and on the end I had left resting on the floor. If these bubbles were to trap water rather than air they could be disastrous so I resolved myself to repairing them once the other tank was ready.

Pockets of air formed in the fiberglass as it cured

The second time around went a bit smoother. I skipped the 406 filler, which for some reason was very difficult to mix into the Gluv-It, and instead just waited to apply the epoxy until it was the consistency I felt would best paint on thickly without running. This seemed to work well. The only other difference from the first tank was that instead of leaving the tank on end to cure I rested it horizontally across a box and attended to it while it cured, pressing down bubbles as they appeared. As it cures epoxy gets progressively less malleable and stickier, making it easier to hold the fiberglass down where it wants to spring up. By tending to it during this time I had far fewer bubbles on the second tank.

When both tanks were fully cured I went over them carefully with a marker and utility knife, opening up any bubbles and marking spots where bare aluminum peeked through.

The raised areas where the fittings went in were an issue
As was this edge, particularly on the first tank.

I then mixed a batch of thickened West System epoxy and 406 filler and carefully gooped this into all of the spots I had marked. I took particular care around the edges of where the tank is tapped for fittings, trying to get a seal which will not allow any moisture to creep in between the fiberglass and the tank underneath.

It’s nothing a bit more epoxy won’t fix!

The end result is two aluminum tanks which are coated completely with epoxy and fiberglass, in effect creating a tank within a tank. With a bit of luck these should last the life of the boat, and the repair cost me well under $100, a fraction of the cost of new tanks. Of course I haven’t seen this repair spelled out by any of the experts and there’s certainly the possibility I’m missing something. As always, I’m open to criticism from anyone who wants to poke holes in my technique!

This article was syndicated from Safe At Harbour But Meant For The Sea: DIY Sailing with Paul Calder


  1. Thong Doan

    I would like to know if the gas tank hold up after couple years of service. I am thinking of doing the same to my leaky gas tank.

  2. Allen Sooter

    I am in the same situation but with a 90 gal tank, 900.00 to renew, so I am looking in to repair options. I can give you a tip on removing the fuel residue from the inside very easily.(mine was 1/8″ thick) An automotinve product called SuperClean from Castol, this stuff is great, I poured it in the tank full strenght from the jug and let it set for 4 hours. Rolled the tank to keep the sides wet about every 15 min. Then I took a garden hose and washed it out, the tank is spick and span on the inside. I also found that this will remove bottom paint as well, just put it in a garden sprayer and keep the paint wet for about 30 min. then wash the paint off with a hose and water and a deck brush.
    How long has the repair held up?

Leave a Reply

Your email address will not be published. Required fields are marked*. Comments are moderated and generally will be posted if they are on-topic and not abusive. For more information, please see our Comments Policy.

More from the AIM Marine Group