Van's Air Force
Don't miss anything! Register now for full access to the definitive RV support community.
fuel tank pressure check
- Thread starter dspender
- Start date
Trust but verify
I am an old school kind of guy. Was taught to trust but verify.
I trust they sealed the tanks but would want to take them off and verify that they are sealed with a pressure test. I would want to spray the tank (including the back) with a leak checker while they are pressurized.
It is a lot easier to take the tanks off now, pressure test, and fix any leaks now than it will be to after the tanks have fuel in them.
I am an old school kind of guy. Was taught to trust but verify.
I trust they sealed the tanks but would want to take them off and verify that they are sealed with a pressure test. I would want to spray the tank (including the back) with a leak checker while they are pressurized.
It is a lot easier to take the tanks off now, pressure test, and fix any leaks now than it will be to after the tanks have fuel in them.
Paperwork from Van's that came with the QB wings said to check them. I was hoping that check could be done with them on the wing rather than removing them to check. I suspect I could keep them on, however I would need to verify that those rivets/surfaces on the aft tank surface and that are covered by the inbd upper and lower skins show no sign of air leak.
QB tanks
One of my QB tanks leaked. I filled them with water, and rigged a gallon paint can suspended 10 feet above to provide hydro- static pressure and let them set for a few days. First leak I was able to fix from the fuel sender hole. Rinse and repeat. Second leak took several days to show and required removing the tank from the wing and cutting a hole in the back to apply sealant along the entire inboard rib to lower skin joint. There was no sealant at that joint from the factory! How that got thru quality control I'll never know. Other tank was fine. Must have been sealed by a different worker.
One of my QB tanks leaked. I filled them with water, and rigged a gallon paint can suspended 10 feet above to provide hydro- static pressure and let them set for a few days. First leak I was able to fix from the fuel sender hole. Rinse and repeat. Second leak took several days to show and required removing the tank from the wing and cutting a hole in the back to apply sealant along the entire inboard rib to lower skin joint. There was no sealant at that joint from the factory! How that got thru quality control I'll never know. Other tank was fine. Must have been sealed by a different worker.
On my QB wings, I've gone through each step in the plans to verify completion. To be as comprehensive as possible I personally felt that removing the tanks was important.
As I did it, there have been multiple benefits. Keep in mind that this is a 20-30 minute task (not a huge learning curve).
1) I learned how everything was put together so in the future any leaks would be better understood.
2) My first attempt at removing a fuel tank was NOT after it was painted!
3) Installation of the fuel floats was a little easier on the bench.
Most importantly though, after removing my first tank I found that the factory left me an EZ-out floating around between the tank baffle and the spar. I can't imagine what that would have done after a few years of flying.
Obviously I'd recommend taking a look at yours...
[URL=http://s296.photobucket.com/user/senecaprop/media/0fc580a4-0edf-4bcc-a8fb-58550506f889_zpswwsezgym.jpg.html]
[/URL]
As I did it, there have been multiple benefits. Keep in mind that this is a 20-30 minute task (not a huge learning curve).
1) I learned how everything was put together so in the future any leaks would be better understood.
2) My first attempt at removing a fuel tank was NOT after it was painted!
3) Installation of the fuel floats was a little easier on the bench.
Most importantly though, after removing my first tank I found that the factory left me an EZ-out floating around between the tank baffle and the spar. I can't imagine what that would have done after a few years of flying.
Obviously I'd recommend taking a look at yours...
[URL=http://s296.photobucket.com/user/senecaprop/media/0fc580a4-0edf-4bcc-a8fb-58550506f889_zpswwsezgym.jpg.html]
[/URL]
Auburntsts
Well Known Member
Why not do the balloon test first with the tanks on the spar? If that fails, remove and leak check. I built SB wings, so perhaps I'm missing something obvious...
The balloon acts as a pressure relief valve, and is not the leak test in and of itself as there's a number of reasons outside of a leak that could cause the balloon to deflate over time.
Yes, but if the balloon remains the same size for 7 days..? Ultimately, I agree with the above poster about the benefits of removal vs. the difficulty of removal, but I believe the OP was referring to leaks. Obviously, the OP cannot bring the wing inside for temp stabilization as I was able to do with only the tanks, so it may be more difficult to track deflation/inflation due to ambient conditions. Sure, if the balloon increases/decreases with the same mean, then, yes, probably not a leak. I had fits sealing the balloon on my second tank (first was 7 days with no change), but if done successfully, and your balloon deflates, I'm pretty comfortable calling that a leak test...
rocketman1988
Well Known Member
More like impossible...
I have the QB wings a removed the tanks for a leak check. My shop is insulated but not climate controlled. Just the difference in temperature from morning through noon was enough to change the size of the balloon, not to mention all the other factors. I can't even imagine trying to do it outdoors. I ended up using the manometer method with soap water to check for leaks...and didn't find any. Hopefully the QB worker used plenty of pro seal the day my tanks were built.
I have the QB wings a removed the tanks for a leak check. My shop is insulated but not climate controlled. Just the difference in temperature from morning through noon was enough to change the size of the balloon, not to mention all the other factors. I can't even imagine trying to do it outdoors. I ended up using the manometer method with soap water to check for leaks...and didn't find any. Hopefully the QB worker used plenty of pro seal the day my tanks were built.
AlexPeterson
Well Known Member
Whoa - 10' of water head is a lot, around 5 psi! I suspect joints that wouldn't ever leak in service might eventually leak at that pressure. I'd limit is to perhaps 1' - 2' of head. Best is to use air, limited by a very fine pressure regulator (not the normal compressor types...), and a blow off column of water about 1-2' high, and use soapy water to look for bubbles.
Sorry for the crummy picture, but here you can see I've got a very low pressure regulator, and the outlet goes to a loop of tubing. I've put water in the tube, and pressurize until I get about 12-18" differential. Just to the right of the picture the tube is open, so if the tank were inadvertently over pressurized, it would just blow the water and vent air. There is about 2" differential when this pic was taken.
Jesse
Well Known Member
I've built, fixed and tested many tanks. The only way ivw found to pressure test the tanks and be sure they don't leak is to put them somewhere that they will sit in the correct dihedral position and fill them with Avgas and let them for several days. You will know you have a leak by the blue die left when he fuel evaporates or by the wet fuel itself. It should be done somewhere well-ventilated, like a hangar (not your garage). Cap the outlet, install a vent that is higher than the filler cap, fill them all the way up, cap them and wait. Leaks will make themselves obvious before long.
head pressure
Actually, the head pressure was much less than 5 PSI, as it was restricted to 1/4 ID tubing at the source. PSI means square inch. 1/4 inch ID tubing is much less area than a square inch, and in any case, the pressure wasn't high enough to bust the balloon that was also included in the test rig.
Also the tank most certainly would have, and in fact did leak as the entire rib to bottom skin fillet of sealant was not there, due to failure from the applier and inspector.(as stated in my post)
I would also disagree that it's "much better to use air", as that test was first performed with the air method and failed to display any leaks. This water test was done because one of my RV8 QB tanks also leaked due to a similar QB defect. I learned back then to test with a liquid or your wasting your time. Using Avgas could be better as it has a lower viscosity and could possibly find a leak that is too small for water, but using highly flammable liquid is far too dangerous in my opinion, unless you have a test rig that can control vapors, static electricity, potential spills, etc. I didn't and I feel that the water works fine. One note is that the water will not damage the tank, but you must purge for a couple of days with a very low pressure/volume of air after the test. Hook an air source to one of the inboard fittings and leave the gas cap off, then let it run. Another note: The fuel sender will read a direct short when submerged in water (avgas isn't a conductor),so don't plan on calibrating your gauges during this process, at least past the first few gallons until the water is high enough the reach the rheostat portion of the sender. (the bottom of the range is what the senders are really about anyway) Until the sender is submerged, you can accurately measure the reserve quantity if you make sure the tank is in the right attitude. Measure the resistance thru thru the sender to ground with an Ohmmeter and then simulate that resistance with a large (0-50 Ohm) rheostat when calibrating the EFIS. Be sure and remove the sender after you get the readings, and replace it with a cover plate as the water will cause the sender to rust and fail during the couple of days of testing. I just used the sender that I had previously ruined as a cover, before I discovered this fact!
Whoa - 10' of water head is a lot, around 5 psi! I suspect joints that wouldn't ever leak in service might eventually leak at that pressure. I'd limit is to perhaps 1' - 2' of head. Best is to use air, limited by a very fine pressure regulator (not the normal compressor types...), and a blow off column of water about 1-2' high, and use soapy water to look for bubbles.
Sorry for the crummy picture, but here you can see I've got a very low pressure regulator, and the outlet goes to a loop of tubing. I've put water in the tube, and pressurize until I get about 12-18" differential. Just to the right of the picture the tube is open, so if the tank were inadvertently over pressurized, it would just blow the water and vent air. There is about 2" differential when this pic was taken.
Actually, the head pressure was much less than 5 PSI, as it was restricted to 1/4 ID tubing at the source. PSI means square inch. 1/4 inch ID tubing is much less area than a square inch, and in any case, the pressure wasn't high enough to bust the balloon that was also included in the test rig.
Also the tank most certainly would have, and in fact did leak as the entire rib to bottom skin fillet of sealant was not there, due to failure from the applier and inspector.(as stated in my post)
I would also disagree that it's "much better to use air", as that test was first performed with the air method and failed to display any leaks. This water test was done because one of my RV8 QB tanks also leaked due to a similar QB defect. I learned back then to test with a liquid or your wasting your time. Using Avgas could be better as it has a lower viscosity and could possibly find a leak that is too small for water, but using highly flammable liquid is far too dangerous in my opinion, unless you have a test rig that can control vapors, static electricity, potential spills, etc. I didn't and I feel that the water works fine. One note is that the water will not damage the tank, but you must purge for a couple of days with a very low pressure/volume of air after the test. Hook an air source to one of the inboard fittings and leave the gas cap off, then let it run. Another note: The fuel sender will read a direct short when submerged in water (avgas isn't a conductor),so don't plan on calibrating your gauges during this process, at least past the first few gallons until the water is high enough the reach the rheostat portion of the sender. (the bottom of the range is what the senders are really about anyway) Until the sender is submerged, you can accurately measure the reserve quantity if you make sure the tank is in the right attitude. Measure the resistance thru thru the sender to ground with an Ohmmeter and then simulate that resistance with a large (0-50 Ohm) rheostat when calibrating the EFIS. Be sure and remove the sender after you get the readings, and replace it with a cover plate as the water will cause the sender to rust and fail during the couple of days of testing. I just used the sender that I had previously ruined as a cover, before I discovered this fact!
rocketman1988
Well Known Member
Be careful...
If you choose to use avgas to test, check your insurance first! My builders policy through falcon and global, is only valid until you put gas in the tanks...or so they tell me. Not to mention the fumes in the shop every time you open the tanks...
If you choose to use avgas to test, check your insurance first! My builders policy through falcon and global, is only valid until you put gas in the tanks...or so they tell me. Not to mention the fumes in the shop every time you open the tanks...
RVbySDI
Well Known Member
Cannot agree more with this statement. Simply attaching a balloon to an opening on the fuel tank and applying some arbitrary amount of air pressure in the tank is not a leak test. That balloon is just a pressure relief valve! The leak test involves thorough visual examination of all the seams, rivets, filets where air, H2O, fuel, or whatever, may be escaping. How you do that visual examination is the actual LEAK TEST! As you have seen here, and perhaps read elsewhere, there are many different methods of performning that visual examination. Chose one you are comfortable with and use it. But use it on the tank in a manner that will allow you to examine every minute inch of the tank. Doing a test with the tank attached to the wing is not going to cut it. You would never be able to inspect the rear baffle area (which incidentally, is where the vast majority of leaks occur).
Doing the inspection now will be a great deal easier than having to do it after you are flying. Oh, ok, go ahead and ask me how I know.
Jesse
Well Known Member
If the tanks are not on the plane, I doubt the fuel-in-the-tanks would not cancel your builder's insurance. If in a hangar, the fumes will be there the same testing this way as when the plane is finished and you do a preflight, so I don't see any more danger testing the tanks with fuel versus having the tanks full with the plane in the hangar once finished.
Thanks for all the input. I am removing the tanks from the QB wings to check for an air leak. I plan to use soap bubbles. I plan to use low pressure. In general, I plan to follow VAn's process. Over the course of this project I have observed in general their recommendations are acceptable and when I deviate very much I am often sorry. After this step I install the avonics panel Aerotronics built for me and the engine and prop. I trust that when I place fuel in the tanks, fuel will only exit the tank through the fuel lines and into the cylinders and I will be a happy flyer.
I am really not interested in a debate of applications of laws of Hydraulics or Hydro-static pressure, I'm just trying to pass along some ideas of how I have successfully tested wing tanks for leaks, without damaging them. Again I would reiterate that the air method will not show you a leak of anything but air! The tank must be tested with a liquid for the small seepers to present themselves. Take six pounds of water in a gallon can, raise it the height of your light stand and attach a piece of tubing to the bottom of the can, and see if you can inflate a balloon with it. If it doesn't blow up in your face, then you can probably attach it to your tank and achieve a modest amount of pressure to test the tank.(actually 2 PSI) How do I know this? Because I have done it many times. (Spoiler alert, nothing blew). Use your still intact balloon as a relief valve. Believe the actual results, not miss-informed opinions and guesses.
here is a short video of the rig for all you doubters:
https://www.youtube.com/watch?v=ey8HTRJwSas&feature=youtu.be
here is a short video of the rig for all you doubters:
https://www.youtube.com/watch?v=ey8HTRJwSas&feature=youtu.be
