? Leak testing tanks ?

RudiGreyling · #1 12-27-2005, 10:35 AM

Hi Guys,

I am going to finish sealing both my tanks within the next day or two and have some queries on leak testing them. I was hoping someone that has done it before me can give me some assistance.

1) How soon after I sealed it can I test them? This pro-seal can take a while to set, and I don?t want to test to soon, whatever too soon is.
2) I?ve got the Van?s leak test kit, they suggest 2 methods, which is better pipe or balloon test and why?
3) With both tests they suggest testing with soapy water, this is the first time I read about this, normally I only read the guys just let is stand overnight and see if they get a deviation in pressure reading. Why, isn?t the overnight test a better test?

Thanks in advance?

Kind Regards
Rudi

tinman · #2 12-27-2005, 10:51 AM

Rudi,
I am going to suggest a better, but more time-consuming method of testing the tanks. Go ahead and do the leak test with air pressure. I used a rubber bladder to serve as the pressure regulator and sprayed the tank with soapy water to hunt for obvious leaks. You need to do one more test. Go to an automotive part place and purchase some ultraviolet leak tracing dye. Mix it up in a few gallons of mogas and pour it in the tank. Let the tank sit for a few days, then turn it so that the fuel is wetting another portion of the tank. Do this until you have tested the entire tank. At this point, you sould be able to see any leaks because the dye is quite noticable, but more importantly, you should be any "Weeps". Weeps are much more difficult to locate with the air bubble test. Weeps are often the source of the gas smell that seem to hang in the air around some planes. In many cases, the weep is so slow that it will never form a drop to "leak". If your tank can pass this test, you will be good to go.
Tinman

LettersFromFlyoverCountry · #3 12-27-2005, 10:52 AM

I used the manometer method, mostly because I read stuff with people who tried the balloon and there often seemed to be cases where the leak was a the balloon attachment.

How long to wait? I usually waited a week, although I did my tanks in cold weather. ProSeal doesn't harden but I believe the test is it shouldn't hold a fingernail mark.
As for the soapy water thing, it isn't an either-or and it's up to you. I set up the manometer but squirted soapy water around anyway to see if anything showed up immediately figuring there's no sense waiting if a problem might be obvious. (I also used the ear-to-the-tank skin method (g))

On the first tank, nothing was obvious, nothing leaked and I left the manometer system set up for three days.

On the second tank, it only took a few minutes to see I had a problem. The soapy water revealed a baffle leak. Tons of fun there. After fixing it and sticking a ton of ProSeal inside, it tested fine although I again used the manometer setup and pounded air in accordingly. I didn't let it sit as long this time (it passed the soapy water test) because when I looked at the tank -- and I didn't notice this the first time -- and the bays seemed to expand around the ribs. The manometer seemed set correctly so I figured the psi was correct, but I didn't want to leave it like that too long and risk blowing some seals.

JimWoo50 · #4 12-27-2005, 11:03 AM

I waited a week and than poured a gallon and a half of mogas in and let it sit for a couple of days. Than I turned it upside down and all around and let it sit. I also tried testing with air pressure but was afraid I would pump it up too much and burst something. If you clean the tank before filling it with gas you would be able to see a leak easily but putting the dye in as mentioned by a previous poster wouldn't hurt. I had one tank test leak free and the other had a slow seep in the baffle which I repaired by globbing more proseal in the area and than pouring in a small amt of Randolphs slosh seal and letting it flow down the baffle seam. No more seep ( I hope). Happy New Year and good luck.

RV10Man · #5 12-27-2005, 01:17 PM

I set each tank up in the craddle before installing the rear baffle & filled it level to the rear of the ribs with mogas. I let them set for about 3-4 days. The left one was fine but the right one leaked at the vent bulkhead fitting in the root rib. After riveting in the rear baffles, I fill them up again, set them on the rear side & let them set again for 3-4 days. So far everything looks fine. As far as how long to wait, I let the sealer sit for at least a week before testing.
Before anyone jumps on me for letting gas sit, I'm building in my own personal hanger at my home, unattached from my house. I did lay the baffle on the rear of the ribs while they were sitting. I didn't see much difference in this than a 100 gallons of gas sitting in a tank, inside.

Marshall Alexander
RV10
wings, tanks almost finished

w1curtis · #6 12-27-2005, 02:06 PM

Quote:

Originally Posted by greylingr

How soon after I sealed it can I test them? This pro-seal can take a while to set, and I don?t want to test to soon, whatever too soon is.
I?ve got the Van?s leak test kit, they suggest 2 methods, which is better pipe or balloon test and why?
With both tests they suggest testing with soapy water, this is the first time I read about this, normally I only read the guys just let is stand overnight and see if they get a deviation in pressure reading. Why, isn?t the overnight test a better test?

Rudi,

Have you checked the contents of the Van's test kit to determine if it will be usable on the -10. In the kit I received earlier this year (two leftmost items in the picture below) was totally useless for the RV-10.

I guess the Schrader valve could be of use, however I found it much easier to add air to the tank by just blowing into the drain valve with my mouth, rather than installing the Schrader valve. The cap in the middle is just fine for the "other" RVs however, the RV-10 tank outlet is a female 3/8 NPT pipe fitting. To plug it for testing you need an AN913-6D plug pictured on the right. Then you can attach a Manometer or balloon to the vent port and you are all set for testing.

Methods of adding fuel to the tank to test, to me, seems needlessly dangerous when other safer methods exist. I guess I've heard too many horror stories with bad outcomes while folks attempt to test tanks with actual fuel of any type. YMMV.

http://wcurtis.nerv10.com/07Wings/wings59a.html

w1curtis · #7 12-27-2005, 02:10 PM

Quote:

Originally Posted by tinman

Rudi,
I am going to suggest a better, but more time-consuming method of testing the tanks. Go ahead and do the leak test with air pressure.

What make this better? It is certainly more dangerous, and admittedly time-consuming; so how better?

RudiGreyling · #8 12-28-2005, 12:11 AM

Hi Guys,

Thanks for the quick and detailed responses:

1) Looks like it is best to wait at least 1 week before testing.
2) Looks like I will use the air pressure with mano meter pipe test.
3) I like the idea of the additional test with a little mogas and dye in the sealed tank afterwards just to be safe. This is a cheap, safe and easy final test.

Bob Collins: ?I also used the ear-to-the-tank skin method (g)?...Bob can you hear the air leak, does it hiss, or what do you mean?

Thanks again,

I will let you guys know here how it goes in week and a bit.

Kind Regards
Rudi

szicree · #9 12-28-2005, 12:16 AM

You may already know, but be aware that the manometer reading will vary with the temp in your shop. It's not a big deal, but it freaked me out when I tested mine. I initially thought I had a leak, but the reading came back up the next day.

tinman · #10 12-28-2005, 05:00 AM

Quoted from William:

"What make this better? It is certainly more dangerous, and admittedly time-consuming; so how better?"

William,
What makes it a better test is that it will show minute passages which may not become evident when using the air test. The fuel tends to wick its way through tiny holes over time. The dye test is good at finding these voids.
Tinman