B

bkthomps

Well Known Member
Well I leak tested the tank today, and I'm thinking now I did something incorrect when sealing the rear baffle- there are no less than 5 or 6 leaks on each side. I put a bead of sealant 3/16" from the rivet holes forward, set it down nicely and riveted away

So anyway, now what to do, it has been 48hrs, should I drill all of the rivets out and remove it? should I just drill the baffle->skin rivets out and spread sealant down in there, re-rivet

I'm thinking if drilling all the rivets out, z brackets have to come off too....I'm not really wanting to put up that much of a fight, and I can't find decent instructions on how to properly drill an access hole, do you use scrap metal to make a cover? which exact blind rivet do you use?? help!

Also, I don't think I squeezed the shop heads enough on the rear baffle, is it too late to go and do that now? This is what you get for working tired/late

Pics because some are visual:

DSC02259-1024.jpg

DSC02261-1024.jpg
 
Last edited:
Hi,
I'm unable to see the pictures so I don't know where the leaks are. If they are all in the same bay, the best fix is to cut an access hole to seal them. If the leaks are all along the skin/baffle joint...well, I really don't know which is the best way (more than one access hole?).

To fabricate a cover you can use some scrap aluminium sheet; as rivet I think you can use the same kind used to rivet the z-brackets (I don't remember the correct part number).

By the way, to seal my tank baffle, I spread the sealant forward of the rivet line but just in touch with the edge of the holes. Once dropped the baffle onto the tank I had a good fillet all along the contact zone between the skin and the baffle. Maybe you used not enough sealant or you spread it too forward...

Good luck.
 
I think you should have put the sealant bead right in front on the rivet line for the baffles, not 3/16" forward. When you push the baffle into place it should squeeze the sealant forward and create a solid bead. It also helps to butter up the baffle flanges and rib flanges with sealant so you know that there will be a good solid sealant coverage all along the baffle to tank skin contact areas.

If the sealant hasn't completely set up, any drilling of rivets will create one heck of a mess with all of the metal shavings. If it has set up, trying to pry the baffle off may be difficult.
 
bruceh said:
I think you should have put the sealant bead right in front on the rivet line for the baffles, not 3/16" forward. When you push the baffle into place it should squeeze the sealant forward and create a solid bead. It also helps to butter up the baffle flanges and rib flanges with sealant so you know that there will be a good solid sealant coverage all along the baffle to tank skin contact areas.

If the sealant hasn't completely set up, any drilling of rivets will create one heck of a mess with all of the metal shavings. If it has set up, trying to pry the baffle off may be difficult.
Click to expand...

yeah I think a series of access holes are in my future, I tried the MEK thinning/vaccum method, but i didn't see any movement on the thinned proseal, it sorta sat there mocking me, so when that dries up maybe tomorrow or sunday, i'll try again....assuming it will not fix it I will start to figure out how to do my access holes

I saw another builder who tried removing the rear baffle and well, i'm not really wanting to walk a mile in his shoes :)
 
Don't start drilling yet

It would be helpful if you could post some pictures to show exactly where the leaks are.
If the leaks are at the factory heads of the countersunk AN426 rivets between the skin and the baffle, then you probably didn't drive the rivets wet and you may have to drill them out and reseal them.
If the leaks are along the aft seam between the baffle and the skin, or around the shop heads of the AN426 rivets between the baffle and the skin, it doesn't look like you have sealed these seams after driving the rivets. I used a veterinary syringe with a homemade nozzle (3/8" aluminum 3003-0 tubing with a tip filed to the bead contour) to run a crescent shaped bead about 1/4" wide x 1/8" high (use a guide to get a straight bead) centered on the baffle flange to skin seam. Then I used another syringe to encapsulate each rivet shop head with about a 5/16" dia x 3/16" high blob.
No, I haven't yet done a leak test but I am pretty confident I won't have any.
See also Rick6a's post "Tip:proseal...The fact and the fiction". Shows these beads and blobs.
Also:
http://www.bergdahl.com/NeatStuff.htm
and click on Thickness and Contours of Sealant
 
Access holes

A friend of mine put an access hole on the baffel, between each rib. Use the same access flange and covers that you did on the end baffel. This may sound like a lot of work, but if you get future leaks you'll be glad you did. If you suspect more leaks after you start flying, you might think about not painting the tanks until after phase one, and the leaks have been fixed.

Steve Barnes "The Builders Coach"
 
I used almost one can per tank(RV-10)...

you may end up cutting the access holes to fix properly. I would not recommend baffle removal. Only use toluene(max. 15% by weight) to thin sealant not mek. Mek for cleaning surfaces before/after applying sealant. Also if you have not built second tank yet, allow at least 4 or 5 days cure time. Higher temp/humidity=quicker cure.
 
terrye said:
It would be helpful if you could post some pictures to show exactly where the leaks are.
If the leaks are at the factory heads of the countersunk AN426 rivets between the skin and the baffle, then you probably didn't drive the rivets wet and you may have to drill them out and reseal them.
If the leaks are along the aft seam between the baffle and the skin, or around the shop heads of the AN426 rivets between the baffle and the skin, it doesn't look like you have sealed these seams after driving the rivets. I used a veterinary syringe with a homemade nozzle (3/8" aluminum 3003-0 tubing with a tip filed to the bead contour) to run a crescent shaped bead about 1/4" wide x 1/8" high (use a guide to get a straight bead) centered on the baffle flange to skin seam. Then I used another syringe to encapsulate each rivet shop head with about a 5/16" dia x 3/16" high blob.
No, I haven't yet done a leak test but I am pretty confident I won't have any.
See also Rick6a's post "Tip:proseal...The fact and the fiction". Shows these beads and blobs.
Also:
http://www.bergdahl.com/NeatStuff.htm
and click on Thickness and Contours of Sealant
Click to expand...

I will get some pictures, it is mostly along the skin to baffle seam, I'm guessing there isn't enough pressure held in yet to stress the rivet holes enough to determine if they are leaking, the balloon will not even inflate!
 
It's not hard to access the inside of an assembled tank. An ordinary fly cutter does fine, IF you turn it with a very slow (60 RPM or less), very high torque drill motor.



Think of it as an opportunity to seal your tanks really well, per the Bergdahl link Terry supplied.

 
Last edited:
DanH said:
It's not hard to access the inside of an assembled tank. An ordinary fly cutter does fine, IF you turn it with a very slow (60 RPM or less), very high torque drill motor.



Think of it as an opportunity to seal your tanks really well, per the Bergdahl link Terry supplied.

Click to expand...

did you use a hand drill for this? I'm also considering trying the slosh method w/ the different sealant before I start to drill
 
Last edited:
well i went out today and retested it since I tried the mek thinned proseal/vaccum trick, the thinned sealant has had about 12hrs to harden

other than one rivet on the top inboard side, which I can cover with my thumb to eliminate (i'll fix it, but that's easy) the balloon inflates now!!

I'm nervous about how well i've sealed it though, because what if this shows up again down the road.....so I scraped off all visible sealant between the rear baffle edge and the skin, so there is no "bead" that could be providing me with a false sense of security

it still holds air!!!

And for those of you who aren't here yet or don't understand, I didn't wet-rivet this (seal each rivet on baffle) so putting a bead between baffle/skin is pointless, the rivets will still weeep......must be done the hard way
 
I've had the same problem with the baffle rivets. I drilled some leaking ones out, wet-sealed them, and all was well. So now I'm going to drill them all out, and wet seal all the baffle rivets.
 
spoke too soon, the balloon deflated after 2 days, I'm going to drill out all of the rivets in the rear baffle, pry back the rear baffle/skin seal and do the vaccum/proseal technique, then wet rivet all of the rivets back into place.


good news is i mounted the tank and all of the zbrackets lined up and the W-423 joint plate to the outboard leading edge matched up perfectly
 
Cut the holes

I'd cut access holes (because I did it) before I'd drill out a bazillion rivets and attempt to pry off the rear baffle.
That's why Vans sells the AS6x12x063 aluminum. Split in half, it's just the right size to cover 5" holes in 2 bays. ProSeal and poprivet in place.
 
yeah, i'm not planning to remove the zbrackets or the rib->baffle connections, i'm just going to reseal the baffle->skin joint and wetseal the rivets
 
Don't drill yet! The balloon is not there to determine is you have a leak --it's there to be a pressure relief valve in case you overpressurize the tank during the leak check. The real check is either A) soapy water applied to every rivet and seam looking for bubbles or B) a manometer.

The balloon can deflate without leaks for a number of reasons:

- The balloon skin itself is permeable allowing air to escape over time
- Changes in atmospheric pressure
- Changes in temp (cool the tank down and the balloon will deflate)
- Leaks around the balloon/vent interface itself or the gas cap, both of which are non-issues.

Bottomline, the balloon deflating after 2 days doesn't mean you have a baffle leak unless a real leak check confirms it. My $0.02.
 
Auburntsts said:
Don't drill yet! The balloon is not there to determine is you have a leak --it's there to be a pressure relief valve in case you overpressurize the tank during the leak check. The real check is either A) soapy water applied to every rivet and seam looking for bubbles or B) a manometer.

The balloon can deflate without leaks for a number of reasons:

- The balloon skin itself is permeable allowing air to escape over time
- Changes in atmospheric pressure
- Changes in temp (cool the tank down and the balloon will deflate)
- Leaks around the balloon/vent interface itself or the gas cap, both of which are non-issues.

Bottomline, the balloon deflating after 2 days doesn't mean you have a baffle leak unless a real leak check confirms it. My $0.02.
Click to expand...

excellent point, maybe i will do the soapy water trick one more time, I think resealing the rear would just make me feel better....:D
 
Hey, understand completely. I've scraped more than one part that was probably OK, but that I just didn't have a warm fuzzy about or just generally wasn't thrilled with my effort.
 
gimp2x said:
did you use a hand drill for this?
Click to expand...

Yes, an ordinary DeWalt 7 amp 1/2"

<<I'm going to drill out all of the rivets in the rear baffle, pry back the rear baffle/skin seal and do the vaccum/proseal technique, then wet rivet all of the rivets back into place.>>

No way I'd remove and replace the rear baffle, not when I could install access holes. Working through the access holes allows the installation of a proper 250 mil filet along the rear baffle/tank skin line, something you cannot do using the standard assembly proceedure. Sliding the baffle into place gives you something like one of the three bad examples, and sealing becomes a mix of technique and luck.



BTW, with a filet joint in place it doesn't matter if the skin-to-baffle rivets are set wet or not.
 
Last edited:
So in taking the advice above, I rested the balloon method with the soapy water trick outside, I could not find one leak. I believe the balloons are just, well, cheap. To make myself feel better, and to also further take the advice above, I did not drill out any rivets, nor did I cut any access holes, I did slap on a dab of proseal on the baffle to skin edge as well as the shop head of each rivet. This leaves the only possibly place for a leak to be the rivet heads directly.

What really put me off was that the final fix was the locktite 290 in a very tiny pinsize hole combined with the vaccum method to suck it into the crevise, now that is covered with proseal, i feel better and warm/fuzzy.
 
Loctite!

I also had a minor leak - found it with soapy water.
It was also on the rear baffle.
Loctite sorted it as well.
125 odd hours later, no sign of any leaks.
Glad you didnt have to resort to major surgery to rectify.
Cheers
Richard
RV7
Sydney, AUS
 
If you can handle one more piece of advice (from someone who also had a leaking tank problem), try filling the tanks about two-thirds full with gasoline and see if it leaks.

Put the tank inot the cradle, and just wait a few days.

And if negative, then lie the tank flat and wait a few more days.

Now if the tank does not leak after all that, then you know for sure that the tank will not leak. Please note, that on some rare occasions, the air test will show negative, but there still may be a leak.
 
that's my next step, i'm itchin for an excuse to go get some 100LL and bring it home, after the tank test, it's going in my motorcycle :D:D


.....before I sell it to fund my fuselage order
 
gimp2x said:
that's my next step, i'm itchin for an excuse to go get some 100LL and bring it home, after the tank test, it's going in my motorcycle :D:D


.....before I sell it to fund my fuselage order
Click to expand...

Good thought!

However, you can save a few bucks by not using 100LL as the test fuel. Just use the fuel that you would normally use in your bike will be fine since all of it will be used for that purpose anyway.
 
You must log in or register to reply here.