[danielhv]

Ok. So I admit. It took me probably 4 months to do one fuel tank. And after that 2 of those 4 months, I just quit looking at the plans, and just built it. Everything went fine. *ahem* except one thing. I "forgot" (forgot to NOT get in a hurry) to dip the rivets for the skin to rear baffle in proseal. I did however, get a good bead on the inside where the baffle sits like the plans call for. And I did coat all the shop heads and seam with proseal to ensure there is no leak.

Well, I pressure tested today, and it went much better than it could have. I have 3 rivets leaking. And they are on the skin to rear baffle rivets. So here are the ideas I came up with:

1. Drill out the 3 rivets, drop some proseal in, and re-rivet. (only fear is disturbing the bead on the inside or lodging metal chips betweek the skin and baffle.

2. Try the whole vacuum the tank and suck proseal in through the 3 rivets. (Kinda confused on that one... if I suck it into one rivet, and pull a vacuum again to do the other 2, would the first one I did suck air and not seal. Here's some pics:

 

[N941WR]

If it were mine, I would try the Loctite trick first. If that doesn't work, then I would go after the rivets.

 

[RyanM]

For air to be leaking through the rivets, I'd have to think the skin/baffle seal isn't up to par. I did not use sealant on each rivets for the rear baffle, and so far 1 tank has no issues, the other one did have a leak, which I fixed last night.

Honestly, as much as I didn't want to, I ended up cutting a hole in my rear baffle to fix the leak. In the end, I feel (mentally) much better. It wasn't too big of a deal, and I'm much more confident in my fix.

Take a look at my blog entry.

You can try the locktite, but personally I'd much rather have a cover on the rear baffle and a nice layer of proseal in the tank. But that's me!

Let me know what you try, and the results

 

[Captain Avgas]

For air to be leaking through the rivets, I'd have to think the skin/baffle seal isn't up to par. I did not use sealant on each rivets for the rear baffle, and so far 1 tank has no issues, the other one did have a leak, which I fixed last night.

Honestly, as much as I didn't want to, I ended up cutting a hole in my rear baffle to fix the leak. In the end, I feel (mentally) much better. It wasn't too big of a deal, and I'm much more confident in my fix.

Take a look at my blog entry.

You can try the locktite, but personally I'd much rather have a cover on the rear baffle and a nice layer of proseal in the tank. But that's me!

Let me know what you try, and the results

Hi Ryan, I note that you had to split the reinforcement ring to get it inside the tank. From an engineering point of view the problem with incomplete stiffening of a surface under flexural loads is that forces will be hugely attracted to the surface that is not stiffened (ie where the ring is cut) and the propensity for fatigue cracking in that area is thereby greatly increased.

There might have been two ways to solve this problem. One might be to instal the complete reinforcement ring on the outside of the tank (with countersunk rivets for the platenuts of course). The other solution might be to bridge the internal split ring with aluminium of the same thickness as the reinforcement ring and pick up the bridge with the 2 platenut rivets either side of the gap.

The problem with your current detail is that if fatigue cracking propagates in that area it will be impossible to see it until it runs beyond your proseal by which time the tank will be leaking fuel. Of course that's an area that requires removal of the tank for an inspection in any case (which you will probably not be doing as part of normal periodical inspections) so it warrants a very considered modification if one is done.

Strange as it may seem, it may have been better structurally to have a smaller opening and no stiffening ring at all (as recommended by Vans as per your blog) than a large opening with incompete stiffening.

 

[Rick6a]

Daniel,

I'd be reluctant to go the locktite repair route on new fuel tank construction.....maybe if the plane were already flying but....in this situation, I'd always opt for the permanent long lasting fix. Kudo's are in order because after all, you completed both tanks and apparently only have these 3 small leaks to deal with. That is a heck of a lot better than many builders achieve first time out the gate.

We know you did your best to "get a good bead on the inside where the baffle sits like the plans call for." The bad news is that unfortunately and in all likelyhood that is exactly where the migration path exists allowing the fuel to seep from between the parts up the rivet shank and out the rivet head. A void or interruption in that bead coupled with a void in the proseal between the flange portion of the rear baffle and where it mates with the tank skin is all it takes. However slight that path is, fuel will always find it. The good news is the leaks you discovered could not be located in an any easier place to fix. I would not hesitate to drill out and remove those three rivets. If the removal process results in even the slightest degradation of hole quality, I would then carefully ream the holes up to #30 and WET install NAS1097AD4 rivets. After setting the rivets, avoid using solvent of any kind (at least until the proseal is cured) to remove the sealer inevitably smeared all around the manufactured head of the rivets. Finally, repair the encapsulation (which looks very good in your pics) around the shop heads. Avoid pressure testing for a day or two. Done properly, I'm sure your tank will achieve leak free status.

 

[Rick6a]

Hi Ryan, I note that you had to split the reinforcement ring to get it inside the tank. From an engineering point of view the problem with incomplete stiffening of a surface under flexural loads is that forces will be hugely attracted to the surface that is not stiffened (ie where the ring is cut) and the propensity for fatigue cracking in that area is thereby greatly increased.

Bob,

Others may well disagree with your assessment.....from an engineering POV.

Upon the drawings and direction of a highly qualified DER-IA (Designated Engineering Representative) I replaced the badly rusted filler flange on his personally owned Cherokee fuel tank and used the split ring method per his direction to close up the access hole I had to cut into the side of the fuel tank. I must conclude the multiple degreed (retired) chief flight test engineer for McDonnell-Douglas had little concern for the propensity for fatigue cracking when he drew up the plans.

 

[Jamie]

Daniel:

For what it's worth, I found a fuel leak the day after I filled my tanks for the first time. The leak was on one of those same rivets, but on the bottom.

I drilled the rivet out, dipped a pop rivet in pro-seal and riveted it into place. It's been fine since.

If I were in your situation though at this point I would probably just bite the bullet and cut a hole in the rear baffle. I know it probably seems like a step back, but I think you'll spend longer thinking about it than actually doing it. The only challenge will be finding a way to neatly cut a hole in that rear baffle.

 

[wingtime]

Had the same problem

Dan,

I had the same problem on my first tank. The problem is probably not enough pro seal between the tank skin and the rear baffle. Van's instructions at the time said to use a proseal sparingly at this joint to avoid "pillowing". I guess I was to "sparing" and mine leaked.

I had started off finding one or two leaky rivets. Then I fixed them and more started leaking and so on. The only solution is getting proseal in that skin to rear baffle joint forward of that row of rivets.

I thought of cutting access holes in the rear baffle in each bay however I figured that would be the last resort.

I ended up drilling out all the skin to rear baffle rivets (top and bottom) cleaning off the proseal with a scotch brite wheel in a die grinder, and scraping the rest out of the joint with a scraper I made out of scrap aluminum. I was able to keep the joint pried open with popsicle sticks while I squirted proseal in the joint with a syringe. It was hard work but I have no leaks.

My second tank received a good buttering up on this joint with no pillows or leaks. (be careful not to block the cut outs in the back of each rib)

PM me if you would like to call me about it.

 

[Captain Avgas]

Bob,

Others may well disagree with your assessment.....from an engineering POV.

Upon the drawings and direction of a highly qualified DER-IA (Designated Engineering Representative) I replaced the badly rusted filler flange on his personally owned Cherokee fuel tank and used the split ring method per his direction to close up the access hole I had to cut into the side of the fuel tank. I must conclude the multiple degreed (retired) chief flight test engineer for McDonnell-Douglas had little concern for the propensity for fatigue cracking when he drew up the plans.

Hi Rick, your post was most interesting and certainly intriguing. But it may in fact raise more questions than it answers.

The access hole that you have shown in the tank of a Cherokee is certainly not of any Piper design and therefore for that modification to be legal there would presumably need to be an Engineering Order approved.

Conventional wisdom would dictate that in normal circumstances obtaining an Engineering Order to gain temporary internal access to the outboard end of a Piper Cherokee tank would not be worth the trouble. Typically rivets would be drilled out and the skin would be removed and replaced if internal access to the tank was required.

I suppose the question that must be asked therefore is whether this mod is actually a legal modification or something that has been conjured up by the aircraft owner simply to avoid the necessity of having to repaint the tank.

Incidentally, the aircraft owner may be, as you stated, a "highly qualified DER " but that in itself may not necessarily be relevent in this matter. The FAA designates 9 categories of DER (Designated Engineering Representative) including Acoustical Engineering, Radio Engineering and Flight Test Pilot among many others. Only one category requires any formal training in structural engineering, namely Structural Engineer.

It is highly unlikely that any certificated aircraft ever produced has a split ring reinforcement at a fuel inspection port. That means that there is no substantially known precedence or track record for this mod. That is something worth considering.

In the final analysis my goal is simply to alert RV Experimental builders that they need to be VERY cautious when they deviate in ANY way structurally from Vans drawings. They may not understand the adverse structural ramifications of modifications (even very minor mods) that are based purely on "intuitive" decisions.

Of course this is just terrible thread creep and probably not of interest to anyone at all.

 

[RyanM]

In the final analysis my goal is simply to alert RV Experimental builders that they need to be VERY cautious when they deviate in ANY way structurally from Vans drawings. They may not understand the adverse structural ramifications of modifications (even very minor mods) that are based purely on "intuitive" decisions.

I did speak with Van's on my split ring approach, and they didn't have any issues with it.

 

[AX-O]

For what is worth. I had a leak in my fuel tank. The leak was in the inner circle of the BNC connector. After figuring out what to do and if I should use the green Loctite, I went a different route. I mixed a batch of Proseal and thinned it out with acetone. The mixture was really runny kind of like warmed syrup. I placed a few drops on the connector and then put a vacuum on the tank by sucking some of the air out with my mouth. Once the proseal was gone I repeated the procedure. Did it 3 times and left it there for a week before testing it. The leak was gone and the fuel sender checks out good May want to try this method prior to cutting anything out. It was simple. Total work time was about 20 minutes.

 

[Rick6a]

.....The access hole that you have shown in the tank of a Cherokee is certainly not of any Piper design and therefore for that modification to be legal there would presumably need to be an Engineering Order approved....Incidentally, the aircraft owner may be, as you stated, a "highly qualified DER " but that in itself may not necessarily be relevent in this matter.

Bob,

Thread creep is a terrible thing and this is the last time I will publicly comment on this particular issue.

I suppose the question that must be asked therefore is whether this mod is actually a legal modification or something that has been conjured up by the aircraft owner simply to avoid the necessity of having to repaint the tank.

You are quite right that the hole in the fuel tank is not a Piper design, however, the modification work I did is documented and quite legal, as is the baggage door opening I helped cut into the side of his Cherokee 160...the very first in the world.

There was no need to open up the entire fuel tank as you so casually suggest when in this case one could more easily cut in a simple access hole to reach the underside of the fuel cap flange to buck new rivets. Not having to "repaint the tank" as you put it is a bonus, not some clever shortcut taken to avoid dissassembling the fuel tank.

Remember, Van's suggests this very technique (access plate) as an aid to repairing gross leaks in RV fuel tanks.

Incidentally, the aircraft owner may be, as you stated, a "highly qualified DER " but that in itself may not necessarily be relevent in this matter. The FAA designates 9 categories of DER (Designated Engineering Representative) including Acoustical Engineering, Radio Engineering and Flight Test Pilot among many others. Only one category requires any formal training in structural engineering, namely Structural Engineer.

It was imminently relevant when I stated the DER is "highly qualified." That was shorthand for: Chief Flight Test Engineer for McDonnell-Douglas. A condition of employment required a DER-IA designation. He is a Structural Engineer, has had and continues an ongoing working relationship with the engineering staff at the FAA, has developed and is a holder of over 12 STC's, the latest STC for the aforementioned baggage door, and he often works with various air carriers and the "Wings of Hope" organization to implement one-off structural modifications. If you doubt his technical expertise, I am certain he would be happy to assuage your concerns. Shoot me a private message and I'd be happy to put you into contact with him. Fair enough?

It is highly unlikely that any certificated aircraft ever produced has a split ring reinforcement at a fuel inspection port.

You may be right. But you are attempting to compare NEW construction with its inherently open access to a field repair that typically affords limited access.

In any event, Van's does indeed endorse the split ring method and for purposes of our RV's you can be certain Van has his and our best interests at heart. In the end, if Van's says it is okay, it is irrelevant what you and I may think. We do know that energies and efficiencies are better served if some of us builders put more distance between ourselves and the keyboard to focus much more on completing and flying our long unfinished RV project as an alternative to doubting, challenging and second guessing every detail of minutia.

Of course this is just terrible thread creep and probably not of interest to anyone at all.

Agreed, and I too consider this matter closed.

 

[Captain Avgas]

In any event, Van's does indeed endorse the split ring method and for purposes of our RV's you can be certain Van has his and our best interests at heart. In the end, if Van's says it is okay, it is irrelevant what you and I may think.

Rick, No....Vans don't "endorse" it at all. I can't let that one slide.

Vans understands (even if many builders do not understand) that very minor structural modifications to their plans can have potentially profound consequences given enough flight hours.

Vans also understands that they would become insolvent if they attempted to structurally analyse every mod that is thrown at them by builders.

In the end they will respond with something akin to: "That MIGHT work (or that PROBABLY will work) but any changes are your responsibility because we can't know if your mod is structurally satisfactory and we are not about to analyse it to find out." And the response will go on to add the inevitable: "We recommend you build the aircraft according to the plans".

Unfortunately that logical non-commital response, as in this case, is interpreted by the layman builder as being: "Vans didn't have any issues with it".

The next thing you know some-one else on the internet is parroting this misinformation as: "Vans endorses it".

In an age of incessant litigation this sort of thing must drive Dick VanG to the point of a nervous breakdown.

I've commented on this matter purely because I'm considering installing an access port in the outboard end of my completed main tanks. If I do so I will be installing the reinforcement ring on the outside of the tank which will obviate the need to split it as per my Post #4 on this thread.

Actually I'm quite surprised that your friend with the Piper didn't do that as well. Why split the ring and lose continuity of the stiffening element if there is absolutely no need. Perhaps you might like to ask him that question.