Wing Attach Fitting Welded In Wrong Spot

[TonyG]

Hi all,

I'm working on my Stits SA6B project, and have turned my attention to the wings. This was obtained as a partial project, and I've just noticed a possibly serious error on the part of the original builder. The right rear wing attach fitting is located too far forward on the fuselage, and so isn't aligned with the rear spar as it should be. See pictures below.


4119
Correct (forward) fitting

4120
Incorrect (rear) fitting

I would say the discrepancy is 1/4-3/8". Mostly what concerns me is how far off the fitting the aft-most spar strap is. The forward one is adjacent to the fitting, albeit "inside" the bracket rather than in front of it, but at least the bolt is only loaded in shear through this part. The rear strap is cantilevered off the fitting by the bolt.

My thoughts are:
1. Cut and relocate the fitting, which is all but impossible without serious reconstructive surgery to a major cluster of structural members.
2. Weld a shim to the aft face of the fuselage fitting to relieve the bending on the bolt. This still leaves the forward strap of the rear spar on the wrong side of the bracket, but very close to it. Here is an illustration of the idea.

4122
Proposed fix.

Any thoughts? It's a deviation to a critical structural component, but I don't have a lot of options. I don't like that the fuselage-side bracket won't be "captured" by the spar straps as per design, but I don't know how big a deal that is. I could put a bushing on the bolt, captured in the middle of the fitting, which would effectively transfer any longitudinal loads into the rear half of the fitting without putting the loads on the nut.


-Tony

[Aaron Novak]

Hi all,

I'm working on my Stits SA6B project, and have turned my attention to the wings. This was obtained as a partial project, and I've just noticed a possibly serious error on the part of the original builder. The right rear wing attach fitting is located too far forward on the fuselage, and so isn't aligned with the rear spar as it should be. See pictures below.


4119
Correct (forward) fitting

4120
Incorrect (rear) fitting

I would say the discrepancy is 1/4-3/8". Mostly what concerns me is how far off the fitting the aft-most spar strap is. The forward one is adjacent to the fitting, albeit "inside" the bracket rather than in front of it, but at least the bolt is only loaded in shear through this part. The rear strap is cantilevered off the fitting by the bolt.

My thoughts are:
1. Cut and relocate the fitting, which is all but impossible without serious reconstructive surgery to a major cluster of structural members.
2. Weld a shim to the aft face of the fuselage fitting to relieve the bending on the bolt. This still leaves the forward strap of the rear spar on the wrong side of the bracket, but very close to it. Here is an illustration of the idea.

4122
Proposed fix.

Any thoughts? It's a deviation to a critical structural component, but I don't have a lot of options. I don't like that the fuselage-side bracket won't be "captured" by the spar straps as per design, but I don't know how big a deal that is. I could put a bushing on the bolt, captured in the middle of the fitting, which would effectively transfer any longitudinal loads into the rear half of the fitting without putting the loads on the nut.


-Tony

Tony,

I would advise against the shim idea as that would be causing an oblique loading of the attach fitting. Perhaps gusseting and reinforcement could be added to counteract this loading, however the best approach to me would be to correct the error in the fuselage (to match the plans). In the end its probably not as bad of a re-work as you are thinking.

[Frank Giger]

Duct tape. Lots of duct tape sandwiched a quarter inch thick.

Okay, more seriously:

I agree that while shimming the mount as you suggest will work, it's not optimal due to the weird loading on the joint - over time it could crack, and having a wing depart from the aircraft would have adverse effects to the flight characteristics (or so I've read).

If the fuselage is square as it is I understand your hesitation on entering the rabbit hole of unintended snowballing of corrections. Nothing makes for piles of junk like a fix to the fix needed to fix something caused by a repair.

One thing to consider is to make a gusseted mount and weld it to the fuselage. If one looks at the problem mount the start of the gusseting solution was actually started with the top mount coming out to the wing; it just wasn't pursued in the rear. To get a little beefier, one could oversleeve the fuselage tube to match the gusset.

I think you'd be fine with that since this is a low speed non-aerobatic aircraft.

I will, of course, defer to experts who will now post that I'm trying to kill you with bad advice.

[Jeff Point]

You are looking only at fixing the fuselage half of the joint. How about the wing? Could you build a new wing attach fitting in which the front plate is shimmed forward that 1/4" in order to fit on the outside of the fuselage fitting? You'd have to shim the new wing fitting to fit the wing spar (ala aluminum spar Cub type wings) and you'd still have the shim behind the rear part of the fitting (as per your idea) but now the wing fittings would both fit outside of the fuse fittings as designed, and that should eliminate the oblique load issue that Aaron mentioned. (I'm not an engineer, so Aaron please check me on this if I'm wrong.)

This way you don't have to mess with the fuselage structure at all. Building a new wing fitting will be some work, but far less work than your idea and you don't mess with a critical piece of structure (ie- if you screw up the new wing fitting, toss it and build another one.)

[martymayes]

Tony, I don't see a problem with adding a spacer like you suggest. Might even consider fitting it to the longeron and welding there as well. Will certainly add weight. The wing is externally strut braced so in the overall scheme, it won't make much difference. Perhaps you can find someone who can verify the loads at that attachment?

Neat project BTW.

[TonyG]

Aaron, that's the answer I was afraid of! Frank, you understand my mentality on the issue as well. My concern is just how integrated that fitting is. The fitting is welded into place and then a strap is formed and welded over it to capture it against the tubing (see plans excerpt below). That thing ain't coming off without cutting out the whole cluster, I fear. I do agree (and appreciate the confirmation) that it seems a little sketchy to load that fitting obliquely when it was designed to take vertical loads symmetrically across the two flanges. Jeff, I thought about modifying the wing as well, at least adding a spacer between the spar and the forward strap so that it rests on the correct side of the fuselage fitting, but the load will still act through the same plane and result in torque on the fitting. It's just transferred slightly differently through the bolt. I still like the idea of getting the strap to the correct side of the fitting though.


4124


Marty, after posting the original post I thought about extending/welding the shim all the way to the longeron and should have suggested that to begin with. Anybody else have any input on that? It results in a very beefy flange, but would help move the load further back and closer to where it should be transferred into the fuselage and (hopefully) eliminate/reduce the torque on the fitting. It greatly changes the rigidity of the flanges relative to each other (which could be remedied by doing the same to the forward flange), but I don't expect that to be a huge issue in this case. Also, a rough back-of-the-envelope calculation suggests that the rear spar fitting might only take about 1/4 the normal load of the forward spar fitting, which is built exactly the same and suggests quite a bit of margin in the rear fitting that is incorrect.

I'm ready and willing to take advice, and if that means I have to do some serious surgery then so be it - I just want to exhaust all my options before committing to that!


-Tony

[martymayes]

Marty, after posting the original post I thought about extending/welding the shim all the way to the longeron and should have suggested that to begin with. Anybody else have any input on that? It results in a very beefy flange


I wouldn't keep it the same thickness all the way to the longeron but instead bevel it so it would be ~1/16" thick where it butts the tube. This would make for easier welding as you wouldn't have that thick steel wicking heat away. In fact, I'd bevel it everywhere 1/4" thickness wasn't required.

Tony I know of another type plane where if you use later style wings on the earlier style fuselage, the wing fittings don't line up and it's common to install spacers to make it work. That type of fitting is very common and it's doubtful Ray Stits designed it as a "zero tolerance" fitting. A fractional spacer will have minimal impact on overall strength.

[TonyG]

I wouldn't keep it the same thickness all the way to the longeron but instead bevel it so it would be ~1/16" thick where it butts the tube. This would make for easier welding as you wouldn't have that thick steel wicking heat away. In fact, I'd bevel it everywhere 1/4" thickness wasn't required.

Tony I know of another type plane where if you use later style wings on the earlier style fuselage, the wing fittings don't line up and it's common to install spacers to make it work. That type of fitting is very common and it's doubtful Ray Stits designed it as a "zero tolerance" fitting. A fractional spacer will have minimal impact on overall strength.


Marty, I was thinking about the challenge of welding 1/4" steel to 0.035" tube as well. Not super easy, but doable with practice and careful heat management. The benefit to maintaining the thickness all the way to the longeron would be that you would also be transferring the load into the longeron further back, closer to the design. Agreed, though - this isn't a precision instrument, it's an aircraft designed with margins and an understanding that an unskilled fabricator would be forging and assembling components by hand.


-Tony

[martymayes]

Tony, with what you have demonstrated in this thread, I have complete confidence in your ability to solve the issue. Press on!

[TonyG]

Tony, with what you have demonstrated in this thread, I have complete confidence in your ability to solve the issue. Press on!

Ah jeez, people always have more confidence in me than I do, myself!

[Aaron Novak]

Ah jeez, people always have more confidence in me than I do, myself!

That should be a good thing right? I am betting that some careful use of a dremel cutoff wheel will have that strap and mount taken off in no time. I would slice off the mount, then gently cut while peeling off the strap so as not to cut into the tubing underneath. Once the strap is off, clean up any remaining weld then start over with a new mount and new strap like nothing ever happened. In the end I think you will have more confidence this way, and it will go a lot faster than you think. Since you are not touching the base cluster, I am very doubtful that there will be any distortion. Good Luck!

[martymayes]

Tony, how does the other side fit up?

[TonyG]

Aaron, I'm definitely going to need to stare at the fitting for a while to work up my courage to do it, but you may be right. It's a heck of a thing to pull the trigger on, that's for sure.

Marty, I didn't get the opportunity to check the fit of the other side last weekend, but the spacing measured correctly when I checked it a while back. I'm planning to go out to the hangar again this weekend and will see how good or bad that is. I will say that the rear fitting on the left side is noticeably aft of the fitting on the right side. It should be in the right spot.

[Aaron Novak]

Aaron, I'm definitely going to need to stare at the fitting for a while to work up my courage to do it, but you may be right. It's a heck of a thing to pull the trigger on, that's for sure.

Marty, I didn't get the opportunity to check the fit of the other side last weekend, but the spacing measured correctly when I checked it a while back. I'm planning to go out to the hangar again this weekend and will see how good or bad that is. I will say that the rear fitting on the left side is noticeably aft of the fitting on the right side. It should be in the right spot.


Ten to Twenty minutes to remove it and prep the area for welding, another hour to form up a mount and strap, welding it on. Most stumbling blocks in projects are purely a figment of our imaginations, and once we get over that and start working, we suddenly realize our own ability. Go for it :)

[TonyG]

It will be next weekend before I can think of tackling it, but I'll try to convince myself it won't be that bad in the mean time. Seems to me like the cleanup will be pretty lousy. Those weld beads go over some pretty strange shapes. My main worry is taking off too much or otherwise compromising the good material which may or may not be visible or fixed when I re-weld the fitting and strap. Maybe I'll make up a test coupon to see how easy and controllable it is to do such an operation...

[Frank Giger]

Aw, it's just another "deep breath" moment in building, where we all hem and haw and look at it and think and then hem and haw some more - and then at some point we just dive in and do it.

And it works out okay.

Oh, and your comment about others having more confidence in your abilities than you do? That's pretty much the easiest way to spot a quality leader who doesn't realize he's leading.

[MCD]

Hope this isn't to late...Looking at the incorrect fitting arrangement this is what I would do and it should be easy. 1) Make a new tab which you will weld onto the fuselage side (same thickness as the original tabs. 2) Weld it to the fuselage tube outside of the lower wing tab (per the photo) 3) Use a spacer between the original tabs and...4) if space allows use two bolts so that the upper two tabs can be compressed and likewise the lower three tabs and spacer can be compressed. From the phto t looks like there is plenty of room.

[TonyG]

Thanks for the encouragement, Frank, though I think I will take your last sentence with a grain of salt!

MCD, not sure if I've interpreted your proposal correctly, but I've drawn it up as I understood it. Is this what you're going for? I guess my concern with this arrangement would be transferring the load from the new rear tab into the fuselage truss - it would probably be fine, but would add a (limited) concentrated loading point right into a tube whereas the original design spread it over a larger area with the finger plate (not shown).

4136

[TonyG]

Aaron, I headed out to the hangar this afternoon and, among other things, inspected my fittings to see what I was up against. Of course you're right, as it looks like it won't be nearly that bad. Inspection (and verification with the plans) shows a decent gap between the wing fittings and the fuselage, welded only at the top and bottom edges. Should have plenty of room to cut without needing the hand of a brain surgeon. I think I can get away with just cutting out the fittings themselves* and leaving the straps intact. This will save me the big concern of possibly grinding into the tubing to peel away the conformal 0.065" straps. Then I can just put a new strap on top of the old strap ends after cleaning up the weld beads so my new strap fits down snug.

*plural - gonna have to do the same to the left side, but not as bad.

Picture showing gap and (lack of) weld, for reference:
4155


-Tony

[martymayes]

Tony, just curious: Do the plans show the wing to fuselage fittings mating up like your photo and sketches? That is, one tab inside, one tab outside the tabs on fuselage fittings? I would think they would fit with both wing tabs inside or outside the fuselage tabs but I could be wrong (again). Also, are those the actual size bolts that will be used?

I think I would just install spacers and let that give it some character. Chalk it up as builder tolerance. You and aeromike have some pretty cool Ray Stits projects going.

[TonyG]

Marty, just for fun the plans don't actually tell you how the wings attach! However, the outside dimension of the fuselage tabs is called out as being (conveniently) the same as the inside spacing between each pair of spar straps. Spacing between front and back fittings is kind of ambiguous, as "by the numbers" the wings don't fit correctly. I noticed this while building my CAD model of the project to help visualize various bits of work. In any case, the idea seems to be to capture each spar fully without having to put any axial load on the bolt. The bolt holes are 5/16", so I assume an AN5 bolt is the tool for the job (the bolt is also not called out in the plans, unless I'm missing it).

I'm still open to the spacer idea since that's what I wanted to do in the first place due to ease, but since that wasn't met with resounding agreement my confidence in that method is kinda weak. Would it be fine? Probably. Is it worth me worrying about for the next XX hours of flight? Ehhh... I know I should probably just do it right to begin with and not think about it ever again.

-Tony

[TonyG]

Well, I'm committed now. I spent Saturday night out at the hangar cutting and grinding the offending wing fitting off. I started with a Dremel and an abrasive cut off wheel, but it proved to not be stout enough for the job. Mainly, I think the Dremel just didn't have the torque I really needed. A newer one might be better in that regard.

In any case, I moved to my angle grinder with cutoff and grinding wheels. The welds were (reassuringly) pretty stout, so it took some work. I never got that moment of satisfaction of pulling/peeling the fitting away, instead having to basically grind the whole thing down to the welds. But I did it, and now I have to fix it. There is still some cleanup to do before installing a new fitting, but it was past midnight and I had been grinding for over an hour. My hands were killing me. I might remove the upper portion of the strap still in place since it's attached to a non-critical piece of structure. The lower portion is welded to a critical member and will also be out of sight in the long run.

I was worried the angle grinder would be like trying to cut wing ribs with a chainsaw, but I was able to keep precise control for the most part. The fuselage gave me a good stable platform to rest against. I put a few shallow gouges in the neighboring tubing, but they're small and not worrying. If I change my mind, I might weld a 0.035" or thinner patch over the gouges and call it done. The patches would be hidden once the airplane is covered.

4166
Original fitting.



4168
And butchered.


-Tony

[Aaron Novak]

Well, I'm committed now. I spent Saturday night out at the hangar cutting and grinding the offending wing fitting off. I started with a Dremel and an abrasive cut off wheel, but it proved to not be stout enough for the job. Mainly, I think the Dremel just didn't have the torque I really needed. A newer one might be better in that regard.

In any case, I moved to my angle grinder with cutoff and grinding wheels. The welds were (reassuringly) pretty stout, so it took some work. I never got that moment of satisfaction of pulling/peeling the fitting away, instead having to basically grind the whole thing down to the welds. But I did it, and now I have to fix it. There is still some cleanup to do before installing a new fitting, but it was past midnight and I had been grinding for over an hour. My hands were killing me. I might remove the upper portion of the strap still in place since it's attached to a non-critical piece of structure. The lower portion is welded to a critical member and will also be out of sight in the long run.

I was worried the angle grinder would be like trying to cut wing ribs with a chainsaw, but I was able to keep precise control for the most part. The fuselage gave me a good stable platform to rest against. I put a few shallow gouges in the neighboring tubing, but they're small and not worrying. If I change my mind, I might weld a 0.035" or thinner patch over the gouges and call it done. The patches would be hidden once the airplane is covered.

4166
Original fitting.



4168
And butchered.


-Tony

Looking good. So how many people welded on the original fuselage?

[TonyG]

Looking good. So how many people welded on the original fuselage?

In my expert opinion (i.e. complete guess), it kind of looks like there were two welders on the project. One might have welded up the basic structure and another came in later to add fittings and other odds and ends. The wing fittings were a little rougher looking than the rest of the structure, which is beautifully gas welded. Then there's me, a monkey with a torch.

I actually have recently been able to track down the original builder (or at least AN original builder), who had signed his name and location on the front of one of the spars. He's up there in years, and this project is who knows how old, but he's still around. An EAA member in the area was able to provide me his phone number and I just need to pick up the phone. It's on my short list of things to do soon. I'm hoping he can give me some information about the project.

-Tony

[Aaron Novak]

In my expert opinion (i.e. complete guess), it kind of looks like there were two welders on the project. One might have welded up the basic structure and another came in later to add fittings and other odds and ends. The wing fittings were a little rougher looking than the rest of the structure, which is beautifully gas welded. Then there's me, a monkey with a torch.

I actually have recently been able to track down the original builder (or at least AN original builder), who had signed his name and location on the front of one of the spars. He's up there in years, and this project is who knows how old, but he's still around. An EAA member in the area was able to provide me his phone number and I just need to pick up the phone. It's on my short list of things to do soon. I'm hoping he can give me some information about the project.

-Tony

Kind of figured that, there are 2 very distinct “signatures”in the welding on that fuselage from the pictures you showed.

[mrbarry]

i would not call that butchered , i would say +offending fitting removed +

im no welder however as a retarded [excuse me i mean retired ] inspector
i think you need to work on that upper join until the +teeth + are gone

to quote the lemon heads .. i might be right but i might be wrong

and at the proper and correct angles to receive the new piece , taking care to remove minimum amounts of the longitudinal structure

[TonyG]

Aaron, you got that right. Learning to weld has taught me a sincere appreciation for really nice welds and really challenging welds. There are some very pretty welds on this project. Those straps were certainly more utilitarian, but still not bad.

Mrbarry, don't worry, I'm definitely going to do some more cleanup. The teeth you point out are on the short list to be cleaned up but survived to see another day on account of it being 12:30am by the time I reached the stage shown in the picture! What I cannot remove safely will probably become filler in the weld of the new fitting. I plan to work the lower half some more also, as I recall there was still some of the original fitting hidden in there. It at least needs to be removed to a point that the new fitting will fit correctly like you say.


-Tony

[mrbarry]

a wing to fuselage fitting ,,
here is an image from earlier in the string should not he area circled be filled with a very pretty weld bead
perhaps some one else remarked and i missed it , perhaps i am wrong ,,but my gut requires me to speak my piece
let the experienced weld tube builders step forth and make it clear.

4171

[TonyG]

Ah, I see what you're saying. A valid concern and I would have wondered the same thing, but the plans actually specify welding only across the top and bottom of the fitting before forming and welding the strap over it.

-Tony

[nrpetersen]

Ah, I see what you're saying. A valid concern and I would have wondered the same thing, but the plans actually specify welding only across the top and bottom of the fitting before forming and welding the strap over it.

-TonyGood point. The problem with a fillet weld is that it is already precracked on the invisible side. That crack will propagate if there is flexing from tension-compression. Worse, the cracked side can't be inspected.

A fillet weld backside crack however won't seriously propagate if the weld is loaded in pure shear.

[Aaron Novak]

That is pretty typical of welded design. Adding a fillet in that areawould add no material strength and in fact could reduce the strength of thefitting by thinning the area in the bend. EAA’s welding handbook is excellent at explaining the engineering ofjoint design aimed at the homebuilder

[martymayes]

Ah, I see what you're saying. A valid concern and I would have wondered the same thing, but the plans actually specify welding only across the top and bottom of the fitting before forming and welding the strap over it.

-Tony

The plans are the correct way to weld the fitting. While welding all the way around a fitting seems like the strongest from a logical standpoint, it may end up being weaker than a couple of properly located welds.

Great progress Tony!