What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Tank testing question

amaris

Well Known Member
Hi Everyone,

Testing my first tank. I'm using the balloon method. I was having a hard time getting the balloon sealed properly so I used a piece of 3/8 tubing over the vent tube fitting with hose clamp and then zip-tied the balloon on. After some pieces of electrical tape, it seemed to not be leaking anymore.

I did the kids bubbles test twice with no signs of leakage. The balloon deflates over a period of 3-4 hours but never fully deflates, it just has a light pressure on it. If I squeeze it, it deflates and then goes back to the same size.

Anyone else experience this? I just want to make sure I'm sealed so I can build on.

Thanks for the input!
 
It takes very little temperature change for the balloon to change in size over time, both directions. I pulled my tanks out of a shaded hangar onto the sunny ramp and my balloons filled up fast! The balloon method is kind of rough but if you did the soap bubble test along all seams you should be good.
Most leaks I see are slow seeps over time. Very difficult to test for those unless you fill them with fuel and let them sit over time. I wouldn't recommend it due to the obvious hassle and hazards. If it was me,I would consider them ready for use.
 
I didn't like the balloon test as all of them leak on their own through their skin, some faster than others and make it tough to get a good sense of leak vs no leak. I recently went to the store and bought a length of clear tubing and made a manometer and the tanks both held pressure for a week each then I disconnected the tubing. 100% sure now of no leaks.
 
Thanks. I did the test in the late morning knowing the garage would warm more as it got later. So while expecting to see it inflate further as it warmed it was the opposite.
 
Much better and easier to just take a long section of clear vinyl tubing and make a manometer. Make a "U" shape hanging down and put some water in the tube. Inflate the tank just enough to push the water down about a foot as compared to the other side of the U. You will see the water level go up and down with the temperature and barometric changes, but unless the levels on both ends of the tubing go to the same level (ie, a leak) then you can be sure your tanks are OK.
 
I didn't like the balloon test as all of them leak on their own through their skin, some faster than others and make it tough to get a good sense of leak vs no leak. I recently went to the store and bought a length of clear tubing and made a manometer and the tanks both held pressure for a week each then I disconnected the tubing. 100% sure now of no leaks.

All the balloon is supposed to do is hold pressure while you do the soap test. The soap test, properly done, will identify any leaks.
 
Hi Everyone,

Testing my first tank. I'm using the balloon method. I was having a hard time getting the balloon sealed properly so I used a piece of 3/8 tubing over the vent tube fitting with hose clamp and then zip-tied the balloon on. After some pieces of electrical tape, it seemed to not be leaking anymore.

I did the kids bubbles test twice with no signs of leakage. The balloon deflates over a period of 3-4 hours but never fully deflates, it just has a light pressure on it. If I squeeze it, it deflates and then goes back to the same size.

Anyone else experience this? I just want to make sure I'm sealed so I can build on.

Thanks for the input!



Allen, The balloon is just a good way to hold some pressure and not risk damaging the tanks. There are several tank pressure test documents from Vans,
See: http://www.vansairforce.com/community/showthread.php?t=150838&highlight=tank

The manometer test is good, but tricky. Perfect if you want to soap all the joints for leaks. Tricky if you want to use it as a leak detector like I did.

Using a manometer, the difference between the column heights is the air pressure. 12 inches of water pressure is 6" deflection from zero. Use no more that about 30 " of water to do the test. You might see bulging even at that.

Then, you can bubble test all the joints. A tiny leak will bubble a lot. Good luck.
 
As has been mentioned, the balloon is not meant to "test" for leaks. For that matter, a manomometer is not all that good for finding leaks either. It is much too sensitive to temp and barometric pressure. Soapy water in a spray bottle while the tank is sealed and a small amount of air pressure is applied to the tank (1 psi is plenty) will tell you if you have leaks. Think of the balloon as a "relief valve" for when you pressurize the tank. If too much pressure the balloon will burst before any seams on the tank.
 
All the balloon is supposed to do is hold pressure while you do the soap test. The soap test, properly done, will identify any leaks.

Yes, sorry about not being clear with the balloon - it by itself is not the test for the leaks merely acts like an over pressure valve so you can test. The issue with the spray bubbles it that a very slow leak won't be that obvious as to create lots of instant bubbles, it may take a while for a bubble to form and if the balloon is slowly leaking down through it's membrane, you may never see the bubble before the pressure is lost. The manometer does not leak pressure but does change levels in relation to temp and atmospheric pressure changes but unless the level on both sides of the tube end up level, the tank is holding pressure and not leaking. You can set it up and come back a week later and get valid results.
 
My way

I just finished my tanks. Here is how I did it. I bought some 1/4" plactic tubing from home dpeot. Heated the end upmand pushed it over the tank's vent to flare it. Added a B nut and AN sleeve and hand tightened it onto the vent fitting. Blew into the tank until i couldnt blow any more, took a swige of water and blew that into the tube for about 30 inches then kept on blowing until it filled up the manometer half way ( both sides level) . Stopped blowing and the manometer settled at about 24 " of water, which is just shy of 1 psi. Then i bubble tested all the joints and let it sit over night. My breath in the tank cooled over night and I lost 1.5" of water, but the bubble test showed no leakage.

After reworking the second tank, i add a second tube on the fuel pickup. One with water for the manometer, one to pressurize. This made it easier to fill with my breath without have to push a swig of water into the tube. I just used a 1/4 inch punch to seal up the pressurization line. Easy peasey.
 
Last edited:
After about 4 days, I've lost about 5" of deflection from my initial pressure. Which was about 12" total in a 3/8" ID tube. The tank is obviously holding at this pressure and I couldn't find any leaks at the full 12" of pressure.

Is it possible to have a leak only at higher pressure and be leak free at the lower pressure and still be okay?

I want to do one more full pressure bubble test to see if I can find anything. I'm also worried about over pressurizing the tank and doing damage.
 
I'm not sure if this has been mentioned in the past, but has anyone had success building a plastic lined water tank and submerging the fuel tank under water to look for leaks?

Cheers,

Phil
 
I poured fuel in them

About a gallon and a half. Left it on ends overnight, standing with aft side down leading edge up and leading edge down in the cradle. I tried to orient the tanks over about a week so the fuel would be concentrated on every seam or rivet hole.
I did try the balloon method on my first tank but was leery of over pressurizing the tank and was not confident in the result because it showed my first tank being leak free. And lo and behold it was!
The second tank was a different story and required access holes in the baffle where I squeezed the proseal out of the seam and set up a minor seep.
 
Bubble test?

After about 4 days, I've lost about 5" of deflection from my initial pressure. Which was about 12" total in a 3/8" ID tube. The tank is obviously holding at this pressure and I couldn't find any leaks at the full 12" of pressure.

Is it possible to have a leak only at higher pressure and be leak free at the lower pressure and still be okay?

I want to do one more full pressure bubble test to see if I can find anything. I'm also worried about over pressurizing the tank and doing damage.

I used 24" of water and only lost 11/2". But the soap bubbles are what showed the leak. Inwouldn tworry about over pressurizing if the pressure is less than 27" which is right about 1 psi.
 
After about 4 days, I've lost about 5" of deflection from my initial pressure. Which was about 12" total in a 3/8" ID tube. The tank is obviously holding at this pressure and I couldn't find any leaks at the full 12" of pressure.

Is it possible to have a leak only at higher pressure and be leak free at the lower pressure and still be okay?

I want to do one more full pressure bubble test to see if I can find anything. I'm also worried about over pressurizing the tank and doing damage.

IMO 12" is not enough to begin with, and with 5" loss you likely have a leak. Be sure to check around the cap too. 24-30" range is ideal (agrees with latest Vans instructions), I ran 44". Lost 0.2" in 24hrs. I also put some tape on the tank and used it to shoot the temperature. The pressure changes a good bit with temperature. Also, be sure to re-zero the manometer at the end of the test. It will change. Dissolved gasses in the water come put and change the level.

Dunking in a tank may seem like a good idea but holding the tank underwater to displace 42 gallons will take 345 pounds of force. Just not practical. Way more work than the manometer.
 
Temperature Compensation

Using ideal gas laws and the ratio of absolute temperatures would seem to be a good correction factor, but it is not. It is close, but not perfect. The tank is a spring too, so the volume of the tank changes with pressure. It is, therefore, also affected by barometric pressure, but not predictable. I did not try to make a perfect correction factor, but by measuring the temperature, keeping in a stable environment, out of the light and getting lucky to catch the temperature at the same (as initial stabilized) point (and no barometer change-AWOS) yielded good results.
 
After about 4 days, I've lost about 5" of deflection from my initial pressure. Which was about 12" total in a 3/8" ID tube. The tank is obviously holding at this pressure and I couldn't find any leaks at the full 12" of pressure.

I tested my tanks using the manometer method. I waited till both were completed and set them up side by side on individual manometers. I "calibrated" the fluid level in each side so they were equal and then pressurized each tank to 27". You don't have to worry about over pressurizing the tanks as the water will just vent out of the manometer if you get carried away. I watched my tanks track within 1/2" of each other over three days. As they tracked together for that length of time, they are either leak free or both leak at the exact same rate. I'm working in an AZ hanger and saw some pretty wild swings with the daily temp changes, but they always tracked true with each other.
If the 5" change can't be explained by temperature, then I would suspect a leak in the tank or in your test system. I used a bit of fuel lube on all of the joints to ensure leak free connections.

https://www.dropbox.com/s/pr5olsz13od80lg/FT-33-170518.JPG?dl=0
 
control balloon

I did the balloon test. One for each tank. Then I got a third "control" balloon, blew it up to approximately the same size and tied it off as just a balloon. I left them in a quiet room in my house for several days. Theory: if the tank balloons change the same as the control balloon, no leaks.
 
I re-pressurized my tank and increased the pressure another 2" and moved the hose clamp, which revealed a small leak. I tightened that down which stopped the leak so I will check it next week when I return from OSH. I hope to find there's plenty of offset in the tube still!
 
I did the balloon test. One for each tank. Then I got a third "control" balloon, blew it up to approximately the same size and tied it off as just a balloon. I left them in a quiet room in my house for several days. Theory: if the tank balloons change the same as the control balloon, no leaks.
Sorry; doesn't work that way. One has the air volume of the balloon; the other has the air volume of the balloon plus all the expanded or contracted air of the tank. All the tank variation shows up in the balloon.
 
Outstanding! Very clever.

I tested my tanks using the manometer method. I waited till both were completed and set them up side by side on individual manometers. I "calibrated" the fluid level in each side so they were equal and then pressurized each tank to 27".<snip> . . . . they always tracked true with each other.

Perfect way to eliminate the guesswork. The odds are against two leaks exactly the same.
 
Manometer

I tested my tanks using the manometer method. I waited till both were completed and set them up side by side on individual manometers. I "calibrated" the fluid level in each side so they were equal and then pressurized each tank to 27". You don't have to worry about over pressurizing the tanks as the water will just vent out of the manometer if you get carried away. I watched my tanks track within 1/2" of each other over three days. As they tracked together for that length of time, they are either leak free or both leak at the exact same rate. I'm working in an AZ hanger and saw some pretty wild swings with the daily temp changes, but they always tracked true with each other.
If the 5" change can't be explained by temperature, then I would suspect a leak in the tank or in your test system. I used a bit of fuel lube on all of the joints to ensure leak free connections.

https://www.dropbox.com/s/pr5olsz13od80lg/FT-33-170518.JPG?dl=0

Same here. Mine were inside so fluctuations were minor. After a couple weeks, I called them good. They were both within 1/2" of where they started.
 
Back
Top