DIY tank pressure test – how?

[BabaYaga]

Hi all,

this summer I noticed a slight and intermittent leak from my s/s diesel tank.
I have now drained it down and taken it out of the boat. What is the best method for finding the hole? Any views welcome here.
I understand that when tanks are manufactured they are often pressure tested to ensure that they are not leaking – could this be a good way also to locate a leak? Perhaps with the help of some soapy water?
If someone has actually done this and could describe the procedures on a DIY basis I would be most grateful.

 

[noelex]

I have never done it, but blocking the exits and using a dingy foot pump should work. I would not use much pressure however its a lot of force on the large surface area of the tank walls.

 

[gruntle]

Commercial tanks are only tested to about 5 Psi, which you should be able to manage with a dinghy pump, although it will take a fair bit of pumping to raise the pressure, because you are filling a relatively large volume. A car foot pump might be better, because they usually have a gauge fitted. Of course, if you can fill the tank with water, you wont have to fill such a large volume, and the leak should be easier to find. You might want to consider this anyway, if you are going to have the repair welded, as diesel vapour can be quite lively when you apply heat to it, and washing the tank out makes it much less likely to go bang

 

[Bobobolinsky]

If you don't want to pressurise the tank, go to the welding supplies and purchase a tin of crack detection developer, it's a third part of "dye penetrant crack detection" Fill the tank with diesel, preferably red. Clean the ouside of the tank with thinners and dry off with lint free rag. Now spray the tank with a nice thin but complete coating of the aerosol crack developer. Leave to soak overnight. Next morning there will be an oily stain where the pin hole is

 

[electrosys]

I've had occasion to pressure-test two kinds of 'tank' recently - the first was the buoyancy chambers on a Mirror dinghy I renovated. The solid rubber bung was replaced by a 'holed' bung into which I glued the brass core of a car tyre valve, and pumped up to 5 psi with an electric tyre pump, the kind with an integral pressure gauge. If the gauge reading didn't move after 15 mins, I moved on to the next chamber.

The second was the steel frame of a heavy launch trolley designed to be immersed - and I wanted to check my welding was 100%. So a small hole was drilled into the box section and another car tyre valve - this time with the rubber seat cut cleanly with a stanley knife - was epoxied over the hole. Same process, except I used 15 psi for 30 mins.

I was ready with the soapy water, but luckily didn't need it.

 

[DownWest]

As above, except the gauges on most pumps are not much cop at low readings. I did a radiator yesterday. Cut the valve out of an innertube with a circle of rubber left, drilled a hole in a bit of ply and clamped the valve over the filler. Blocked the two other hose connnections with more rubber and ply clamped on. With a tyre inflater I gently bled in some pressure while brushing soapy water over the suspect area.
In your case, a leak is most likely in the bottom, or a split seam. If in the bottom, you will need to cut out the whole area, as a pine hole will be the first of many.

 

[pete]

The way we test our drysuits is to connect a vacuum cleaner (Henry) set to blow and whilst it is running test with soapy water your tank will not be under that much pressure in normal use.
Pete

 

[rivonia]

Hi my fuel tank-diesel is at the welders now. I took mine out as it had a leak. The way I tested it is as follows.
1...Clean out the inside of the tank completly and make sure it is nice and properly clean.

2...Pour in a pint of full cream milk

3...Put down some newspaper and on top of this some toilet paper

4...Place the tank edge on the paper and leave for a few mins, result= damp patches where the pinholes are.

5...Mark the areas with a permanant marker pen.

Do the same proceedure for all edges. These instructions were given to me by a friend who is a yacht surveyor.

Good luck

Merry Christmas

Peter

 

[dt4134]

Commercial tanks are only tested to about 5 Psi,

That makes sense. My s/s diesel tank has a little plaque on it that says tested to 0.3bar, which is roughly 5psi.

 

[fisherman]

When the gas engineer tested my pipework he used a water gauge. If you had a water column 10ft high in polythene tube, say, attached to the tank that would do it.

 

[fisherman]

If you do have a leak in a diesel tank I'm told the only thing will stop it is soap, (as from a bar of soap).

 

[BabaYaga]

Quite a number of interesting suggestions there, much appreciated.
Obviously there is not just one, single established method for finding a leak in a tank.
I think I will start with the car tyre valve/rubber bung/soapy water approach and if it fails I will have several other options to try.
Thank you all for posting.

 

[sails]

I just pressure tested 4 2000ltr tanks on my boat. The long established method for marine tanks was to pressure test them at 4 to 5 psi for 24 hours. Use domestic water system test caps to seal vents and fill pipes. Make a "tree" with a 10psi guage , an air compressor attment and ball valve. Fill with air from a compressor watching the 10 lb gauge closely. Usesing a misting bottle filled with dish soap and water check all seams, spigots and hoses.
My tanks passed the 24 hr test. But not the fist 2 trys.
I then pumped 4000ltrs of diesel into the tanks....... and have 2 very very slow drips........ 6 drips per 24 hours on one ball valve and one sight glass fitting.
I have since learnt that the new rues for marine tanks are a 24hr pressure test and then filling the tank with water with a 10ft 1/2 inch pvc tube standing verticaly. Mark the tube at the level of water with a scratch on the inside of the tube and wait the 24hrs. Check for moisture at all fittings.
Hope this helps.
Hindsight on my tanks, Checking the 10psi guage often during the day and evening I noticed a 1/3 psi difference intank pressure. The next day, as the air temp warmed so the tank pressures increased but not back to the exact spot at which I started the day before. A hair width lower. I brushed this off as ambiant air temp differences. My mistake.
After repairs are made on the tank the air test allows for quick repairs.

 

[Graham_Wright]

Just a word of caution. 5 psi over, say, a 24" X 12" X 12" tank is a force of around 3 tons! If that goes pop, perhaps so do you!

Fill the tank with water to almost full then top up with air. The water is incompressible and will not store energy and you wont have so much pumping to establish the required pressure.