What's new
Van's Air Force

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

both tanks are done... now do I...

I

instructor_bill

Well Known Member
Sure, the standard practice is to pressure test the tanks with a balloon. I've done that and two leak-free tanks are complete. Now here's the golden question: Is a fuel tank test with full fuel a good idea?

Rather than hope they're both okay and proceed to paint (only to find a year later that I get blisters and leaks with a flying airplane) should I take it to the next level and fill'em up?
 
That's what I plan on doing and while a hassle it will make me feel a whole lot better about calling the tanks leak free. I even thought about filling them in before I put the rear baffle to make sure I have no leaks in the ribs.
 
When are you looking to paint? Each piece as you go or as an assembly closer to first flight? Here's my opinion: If your painting as you go then do you have something to use 42 gallons of avgas after testing? That's around 200 bucks to waste if you don't! If you do then why not. Now if your going to paint down the road closer to first flight then hold off and do it before you paint the assembled plane. Again just my 2 cents.
 
I plan to paint sub assemblies

I've got the tail and one wing complete, after I finish the second wing I plan to paint while I wait for the fuse kit to come (I'll have to order it first, but I'm waiting for this year's bonus.)

So, since I plan to paint the wings and tail before April-- should I throw some fuel in?
 
I just finished up my -10 tanks and am planning to fill one at a time to test for any leaks prior to painting. No need to fill both at the same time so I will only have 30 gallons of 100LL to buy and transport in 6 gallon cans. I will use a fine mesh filter funnel to drain it back into the cans. This will help flush any debris that may be in the tanks also. I will then use the fuel in my Cessna.
 
instructor_bill said:
Sure, the standard practice is to pressure test the tanks with a balloon. I've done that and two leak-free tanks are complete. Now here's the golden question: Is a fuel tank test with full fuel a good idea? Rather than hope they're both okay and proceed to paint (only to find a year later that I get blisters and leaks with a flying airplane) should I take it to the next level and fill'em up?
Click to expand...

IMHO a waste of time, not to mention an unnecessary fuel-handling danger. You've already balloon-checked them. No new leak is going to show itself until after they've seen dynamic service; G's and deltaT.
 
pazmanyflyer said:
When are you looking to paint? Each piece as you go or as an assembly closer to first flight? Here's my opinion: If your painting as you go then do you have something to use 42 gallons of avgas after testing? That's around 200 bucks to waste if you don't! If you do then why not. Now if your going to paint down the road closer to first flight then hold off and do it before you paint the assembled plane. Again just my 2 cents.
Click to expand...
you only need to fill one tank at a time and if you have no use for it, I'm sure a ride in an RV can be arranged.
 
A full tank, tare + gas, is on the order of 140 pounds to sling around. Try a few gallons, then rotate between both sides down and the baffle down each for a while to check all rivets. Why bother with 100LL; use mogas This whole exercise is a hassle not really worth the bother if it passed the balloon test (which presents its own problems with changes in temperature). Instead, pressurize the tank with air until the skin begins to pillow a bit, cap it off, then squirt the entire tank with soapy water. A leak will be more evident blowing bubbles than hoping to see a tiny amount of gas that might evaporate before you see it.

For real excitement, put in a little gas then pressurize it. This is called a bomb, but it will expose leaks.

John Siebold
 
Last edited:
For what it's worth, I balloon checked my RV-7 tanks for over 2 weeks, and had no leaks. But, like Dan said, after 2 years of service they are riddled with blisters. :(

Vic
 
so, I guess the answer is...

if a balloon test is good, then it's good enough to fly. If it's good enough to fly, then it's good enough to create blisters.

Isn't there a better way?

I was hoping for a "check for the smell of avgas" or "fill them and cover the tanks with____ and check for leaks by______. If no leaks then paint."

Oh well, I guess it's a necessary evil. At least I know that I won't have to account for both fuel flow and leak rate.:rolleyes:
 
Me Too

Balloon held for an eternity, painted, assembled plane, flew for 160 hrs, never (never) had any trace of a fluid leak, removed tanks to repair over 200 blisters :mad:. Just be sure to seal every joint with a nice fillet and cover all rivets, don't skimp on sealant.
 
Another consideration, Bill....

...is that an EFIS, like the Dynons, require fuel added one gallon at a time and a push of a button each time, to calibrate the fuel gauges. Maybe every two gallons, but then would be a good time to fillem' up,

Best,
 
Use a Manometer

I built a manometer from plans found elseware in these forums. When I pressurized the tank I found one leak around the fitting attaching the manometer to the tank. Once that leak was sealed, the tank held pressure. I let it sit like this for about 4 hours. When I checked the manometer the pressure had increased as the sun was now shining on the tank in the doorway. So I figure if the tank can hold pressure initially and handle expansion and contraction of the air inside, I betting that I will have a relatively leak free tank. Oh, I failed to mention that I used about 3/4 of a quart of sealant on the tank. No joint, rib, rivet, etc. went without some amount of sealant applied.
 
I've had both tanks complete and patiently waiting in the wing cradle for about a year. The fuel pickup and vent lines are capped. The other day someone asked what the quick drain valve was, and I demonstrated it's use. I was pleased to hear a nice hiss, showing that the tank had some kind of pressure differential to ambient.

The other tank is a different story. When first built, I used the balloon test. It almost passes. Spraying soapy water shows it has one tiny leak - the center terminal of the BNC connector for the capacitive fuel level sender. I opened up the tank later (don't use proseal AND the cork gaskets!) and slathered more proseal on the BNC from the inside. That didn't resolve the problem.

Has anyone else seen this? Not sure how to proceed short of replacing the BNC connector, or ditching it in favor of the float gauges.
 
Ok - I'll bite!

You all are my source of aircraft home building information so I'll ask a dumb question or two related to tank testing. Flame if you like but I'll consider your responses as a learning experience;

1. If gasoline must be used, why not use automobile gas to do the test? Then drain and use in your car, etc.?
2. Why not use water for the test? Sure, it weighs about 8.35#s/gallon instead of about 5.9#s but it sure is safer and cheaper. Would a take hold water and leak gasoline? Draining it a problem?
Thanks for the education.:)
 
E. D. Eliot said:
.......Why not use water for the test? Sure, it weighs about 8.35#s/gallon instead of about 5.9#s but it sure is safer and cheaper.......:)
Click to expand...
Water is likely to prove nothing. Among other things, its surface tension is far greater than any fuel which means it is more likely to remain contained under pressure compared to fuel.

Besides, if the experience of the Pazmany folks is relevant, and I believe it is, a routine pressure test is fine for uncovering and dealing with gross leaks but more subtle seepage that may well be the cause of paint blistering is a phenomena that usually occurs over a relatively long period of time. A fuel tank tested under pressure for days or even weeks may check out just fine and even be certified for sale if it is a quickbuild fuel tank...but as the experience of a few builders will show... still develop blisters sometime later due to fuel VAPOR seepage:

http://www.pazmany.com/newsletters/PL-1_and_2/64.pdf
 
chipf said:
The other tank is a different story. When first built, I used the balloon test. It almost passes. Spraying soapy water shows it has one tiny leak - the center terminal of the BNC connector for the capacitive fuel level sender. I opened up the tank later (don't use proseal AND the cork gaskets!) and slathered more proseal on the BNC from the inside. That didn't resolve the problem.

Has anyone else seen this? Not sure how to proceed short of replacing the BNC connector, or ditching it in favor of the float gauges.
Click to expand...

What about the other end of the wire? The tiny spaces between the conductors can be a leak path. BOTH ends need to be sealed to ensure a leak free installation.
 
I was thinking about that....

longranger said:
What about the other end of the wire? The tiny spaces between the conductors can be a leak path. BOTH ends need to be sealed to ensure a leak free installation.
Click to expand...

but the implications are too horrible to contemplate. It might also mean that the capacitive sender won't operate as expected.

Installing a float sender seems to be a simpler solution. Besides the float senders are less than having to buy the capacitive gauge conversion units.
 
Rick6a said:
Water is likely to prove nothing. Among other things, its surface tension is far greater than any fuel which means it is more likely to remain contained under pressure compared to fuel.
Click to expand...

Right on Rick.

I got curious about liquid vs vapor leaks and looked up the physics a few weeks ago. As I understand things, a very small hole means non-laminar liquid flow and a relationship of pressure = (4 x surface tension) / hole diameter. A generally accepted value for the surface tension of gasoline is around 26 dyne/cm or 0.000148462 lb/in, so liquid fuel shouldn't pass through a hole smaller than about 0.0006" diameter at the typical dynamic tank pressure of 1 psi.

If you assume a full tank with a head height of 10" and a fuel density of 45 lbs/ft^3, you would have a lowest point head pressure of about 0.26 psi. Add it to the 1 psi dynamic pressure for 1.26 psi and the "no liquid flow" hole size goes down to about 0.00047" for the bottom of the tank.

As for vapor....gases don't have a surface tension, so given the same pressure, a gas phase should happily flow through a hole which wouldn't pass a liquid phase.

As always, the real engineers should check me...I've been hit in the head a few times.
 
Last edited:
so then, DanH

Does that mean that a hole that will not pass gasseous air (nitrogen, oxygen, carbon dioxide and others) will in fact still be large enough to pass fuel vapor?

Fuel vapor under pressure?

I don't get it.
 
Using air in pressure test

In a former life I was a Sr. Engineer at Kelsey-Hayes, supplier of OEM brakes to Auto manufacturers. The 100% QA check for the brake caliper assembly line was an air pressure test. Air can find a smaller leak and is a lot less messy when it finds one.

The suggestion that vapor leaks that cause blisters develop during service is probably a good one. My Cherokee 180 had over 6300 hours and had some "smoking" rivets that would not have been vapor tight.

LarryT

DanH said:
Right on Rick.

I got curious about liquid vs vapor leaks and looked up the physics a few weeks ago. As I understand things, a very small hole means non-laminar liquid flow and a relationship of pressure = (4 x surface tension) / hole diameter. A generally accepted value for the surface tension of gasoline is around 26 dyne/cm or 0.000148462 lb/in, so liquid fuel shouldn't pass through a hole smaller than about 0.0006" diameter at the typical dynamic tank pressure of 1 psi.

If you assume a full tank with a head height of 10" and a fuel density of 45 lbs/ft^3, you would have a lowest point head pressure of about 0.26 psi. Add it to the 1 psi dynamic pressure for 1.26 psi and the "no liquid flow" hole size goes down to about 0.00047" for the bottom of the tank.

As for vapor....gases don't have a surface tension, so given the same pressure, a gas phase should happily flow through a hole which wouldn't pass a liquid phase.

As always, the real engineers should check me...I've been hit in the head a few times.
Click to expand...
 
You must log in or register to reply here.
Back
Top