Ethiopean 787 fire at Heathrow

[speedbump59]

On condition of anonymity, this engineer explained the likely repair process.
Boeing will cut out the skin damaged by the fire probably in a rectangular cut with rounded edges, he said.
It will cut the patch to the same size and shape and drop it into the space as a plug. The tiny gap around the patch will be filled with paintable sealant that will stretch and compress as the fuselage is pressurized and unpressurized.
Then mechanics will work on the inside, gluing a splice plate to the original skin and to the patch, overlapping both by about 4 inches.
The glue is a superstrong adhesive that is cured using heat blankets that are held under pressure by vacuum bags applied to the area.


The above was published in Seattle Times, 21 October. If correct, it should stop the discussion on scarfing the joint.



Speedbump

[MrSnuggles]

From speedbump59

It will cut the patch to the same size and shape and drop it into the space as a plug. The tiny gap around the patch will be filled with paintable sealant that will stretch and compress as the fuselage is pressurized and unpressurized.
Then mechanics will work on the inside, gluing a splice plate to the original skin and to the patch, overlapping both by about 4 inches.

Interesting... this is a somewhat unexpected approach. Sounds like a mix between metal repair and wood repair. My spontaneous(sp?) reaction is that I would want to invert the plug method (that is, inserting it from the inside), but if it works, it works. Would love to see a paper on this!

[joy ride]

Thanks for that, speedbump59, I am fascinated by all the talk here of how this repair will be done. That sounds like a surprisingly "primitive" way of repairing it compared to some suggestions here, but being familiar with wood and metal work it also sounds "good" to me! I am eager to find out exactly what is done and hope in the fullness of time we will know for sure.

[Lemain]

I am very uneasy about this. Long before any novel material is decided upon for production of civil aircraft repair should have been investigated and repairs exhaustively tested in lab and field conditions. For repairs carried out ex-factory or specialist repair workshop, the means of carrying out such repairs should have been tested at remote non-specialised workshops.

That it has taken almost four months to decide how to do it is deeply concerning and implies a serious oversight in the design and production programme. Certainly the method of repair should be in the public domain, including all test data taken, confidential proprietary information or not. It is only by publishing the full details to the wider engineering world that peers can comment. If there is a paper, it sounds as though it won't be peer-reviewed and won't have had any proper field trial.

[MrSnuggles]

From Lemain:

I am very uneasy about this. Long before any novel material is decided upon for production of civil aircraft repair should have been investigated and repairs exhaustively tested in lab and field conditions. For repairs carried out ex-factory or specialist repair workshop, the means of carrying out such repairs should have been tested at remote non-specialised workshops.

Carbon fiber composites are not new materials, although being a bit novel to the aircraft world. There are plenty of repair technology documentations.

That it has taken almost four months to decide how to do it is deeply concerning and implies a serious oversight in the design and production programme. Certainly the method of repair should be in the public domain, including all test data taken, confidential proprietary information or not. It is only by publishing the full details to the wider engineering world that peers can comment. If there is a paper, it sounds as though it won't be peer-reviewed and won't have had any proper field trial.

i) The lenghty process could very well have been due to insurance questions, responsibility division, contract agreements etc, as others have pointed out.

ii) Yes, a paper would be most interesting but I venture a guess that that won't happen. Sadly, because this is a very interesting case due to the placement of the damaged area.

[Lemain]

clandestine

Carbon fiber composites are not new materials, although being a bit novel to the aircraft world. There are plenty of repair technology documentations.

The materials used in the Comet 1 were not 'novel' in themselves. With a 200 tonne+ heavy jet flying at near the speed of sound, with 15 psi internal pressure, for fifteen hours a day in turbulence, taking off and landing, the airframe is in a very severe environment, hard to replicate in the laboratory. Forces, changes in force magnitude and direction, vibration, temperature and chemical/moisture attack are hard to replicate in any faithful manner together as a complete structure. With a new aircraft, tens of thousands of hours are flown in real conditions long before the a/c enters service.

Yes, a paper would be most interesting but I venture a guess that that won't happen. Sadly, because this is a very interesting case due to the placement of the damaged area.

I don't think that 'most interesting' is the appropriate term. It is vital that the proposals are peer reviewed by a range of materials and aviation specialists if the repair is going to be tested on the public (or line flight crew, come to that).

[MrSnuggles]

Lemain.

1) Yes, and...?

2) You are correct, that would be most helpful. I hope that a paper will be released, but not holding my breath over here... where there's money there always is an incentive to keep things proprietary...

[Lemain]

1) Yes, and...?

Indeed. That's the issue. Even if you wrote-off this hull you are going to face another ding before long. They happen and when they do you don't expect the Mfr to sit in smoke-filled rooms deciding what to do for the next four months - and then come up with a solution that is essentially un-tested. You know what I'm saying but nobody wants to come out with it. The only way for damage limitation in this product (and loss of public confidence in the Industry) is for the Industry as a whole to address the problem and not treat it as 'proprietary'. I can't see a safe alternative.

Heaven forbid they'll be a bad outcome and I'm sure many of the best minds have looked at it. Out of curiosity, I wonder if the decision-makers have indemnity insurance covering this?

[joy ride]

I think it was closer to 3 months to prepare for the repair a five weeks estimated repair time. Perhaps a good deal of the three months was putting into action a well-researched operation, but Boeing did apparently build a whole new rear fuselage barrel to cut the repair patch from. I presume this would have taken quite a while, and in the mean time they had to plan/hire/build the shelter and heating/ventilation equipment. There is quite a lot of "stuff" there; sourcing, making and/or hiring all of that would take a while, especially as composites need controlled stable environments, and winter at LHR is hardly stable!

[Lemain]

I think it was closer to 3 months to prepare for the repair a five weeks estimated repair time. Perhaps a good deal of the three months was putting into action a well-researched operation, but Boeing did apparently build a whole new rear fuselage barrel to cut the repair patch from. I presume this would have taken quite a while,

Presumably the need to cut out of a new whole section is to preserve the pre-stressing inherent in the manufacture of a whole section? Obviously they wouldn't have done so if a section could have been laid by itself, so it must presumably must be in order to match the original. I presume that the problem is strain rather than stress and that the composite is anisotropic - i.e. the Young's modulus is not the same in all directions. This would affect the natural frequency of the section and it's possible that unexpected resonances would show up after completion of the repair. Resonances are likely to be a greater risk than absolute strength.

I wonder what this tells us about the cost and time to repair less sensational cases such as collision damage on the ground. Could be very expensive for the insurers.

....and in the mean time they had to plan/hire/build the shelter and heating/ventilation equipment. There is quite a lot of "stuff" there; sourcing, making and/or hiring all of that would take a while, especially as composites need controlled stable environments, and winter at LHR is hardly stable!

Many composites need stable environments when curing and until the final seal is applied (to prevent ingress of water) but a panel from the factory should be fine to handle. Obviously the outside and inside surfaces are completely safe as they are designed to resist moisture, ice, glycol, salt and common cleaning fluids and I presume that the edges will have been treated with a suitable sealant. The prepared edges of the a/c hull section are susceptible to ingress of water though environmental control beyond a normal maintenance hangar is unlikely to be needed. A plastic tent around a given section is a well-established technique and can be set up over a weekend. If it needs more than that I would be surprised.

[A and C]

Metalcentric repair ?

While not being in a position to say exactly why Boeing have decided on this plug repair it seems like it has not been driven by conventional composite repair techniques but it has rather more in common with metal repair techniques.

I would be very interested to see the compleated repair and find out quite why Boeing went down a road that seems to abandon the advantages of composite construction.

[MrSnuggles]

A and C:

While not being in a position to say exactly why Boeing have decided on this plug repair it seems like it has not been driven by conventional composite repair techniques but it has rather more in common with metal repair techniques.

Been pondering over this during the day and one thing I realised... MAYBE it is because the damage is so close to the tail fin. This would MAYBE make the scarf joint unreliable (because there would be less material for the angle on one side) regardless of pegs, nibs and hooks, and thus they opted for this intriguing solution. Very interesting indeed.

[gas path]

Speedbump... that's fairly close.
joy ride etal... The Boeing repair teams never turn up on site until they have absolutely everything in place and ready to go.

[NWSRG]

It's important to remember that no-one could have anticipated a damage of this nature.

Had this been a metal aircraft, it would surely have been damaged beyond repair. But with the 787, the composite structure gave Boeing a change to salvage an aircraft that otherwise would have been written off. And while Boeing have no doubt developed detailed plans for dealing with normal 'ramp rash', this repair has probably been developed from first principles.

So 5 months is not unreasonable, in fact, I think its pretty impressive to think that this frame will be flying again in a few weeks.

This could turn out to be a remarkable testimony to the ability of a composite aircraft to sustain major, yet repairable, damage. Boeing will gain invaluable data on such a repair.

And hopefully a paper will come out to tell the story...who knows, maybe the Discovery Channel is in there with them!

[Old Engineer]

Well, from an engineering standpoint, as I see it:

The replacement section is already on site;
The interior finish reeks of smoke and hazardous particulates, and so is likely already out...

Just disconnect the various lines, unbolt the damaged section, and bolt the replacement section back in, shim to align as required. Replace or renew the lines. Cover with new interior.

Whisk the damaged section back to the factory; make whatever repair might have been contemplated at some time, and test at leisure to see if it would have worked.

Why take any chances? So let people think that a marvelously cheap repair has been devised, since that seems to be the drift right now. But just do the sensible thing.

We used to sometimes say in the engineering office that sometimes it's cheaper just to do the job. And I'd say, it always was.

OE

[barit1]

The Boeing repair teams never turn up on site until they have absolutely everything in place and ready to go.

Harr-UMPH!

About 25 years ago I was assigned to a project across the pond. I had never been associated with this hardware set to that point, but was assured it was "just like the last job".

So when I stepped off the plane, I was immediately hit with "Where is xxx? And we're still missing yyy and zzz!" about three IPC pages' worth of hardware was nowhere in sight, because of configuration differences.

After ten days of followup from the logistics gurus, who had been gathering bits and pieces from three countries on two continents, the customer said I might as well go back home for two-three weeks while they worked on production schedule. I was left holding the bag; my only excuse was that I never should have taken the job under false pretense.

But when I finally returned, the job went smoothly, per plan!

[joy ride]

Lemain, post 1108, your comments about the need to make a whole section and cut a panel from it make sense to me. On reflection it seems odd to me that Boeing did not make a complete sacrificial shell, either now, or in the past as part of the development plan.
Airliners do get all sorts of knocks - just recently a BA 787 was slightly damaged by a set of flying steps borne by high winds which had been predicted at least 3 days earlier!
Perhaps Boeing did contemplate a donor hull but this was by-passed due to project delays.

[speedbump59]

If indeed the Seattle Times article is correct and Boeing will use a metalcentric type repair with a lap joint strip, then maybe there are a few more things to think about:

The patch will have a large pressure on it from the inside. Thus the joint between the patch and the 8 inch wide lap joint strip will be in tension (as well as shear). I think it is absolutely impossible that Boeing would trust a resin glued joint in tension. Thus I think the lapped joints must have rivets or the composite equivalent. This makes it even more of a metalcentric type repair.

If they use this method, I assume this means that the cut out and repaired area will not go across any barrel joins. In other words the whole repair area will be confined to within one barrel section. Also the Seattle Times article says they made a barrel section to cut the patch, not that they made two or three barrel sections for patches. But if this repair does have to go across a barrel join, it would add immensely to the complexity. Hard to believe that it would be possible. Anybody know the locations of the barrel joins with respect to the external damage that we could see in the photos?

Thus it seems that the size of the repair is relatively large (thus needing the specially made cut out patch) but limited in size to being within one barrel. Did the fire actually penetrate the barrel (make a hole in the fuselage?) Are there any better photos or newer information? I haven't seen anything that shows a hole, just heat damage (charring) on the outside.

If there is no significant hole, I am now surprised that they will cut out the fire damaged section. Why bother? Given that carbon fibre is pretty heat resistant, even the damaged section probably still has a fair amount of strength remaining. Just leave the damaged section in place. For the patch design, assume the damaged area has zero strength, but we get some extra strength benefit by knowing that it still has some.

Take your lovely patch, glue it on the inside of the barrel, bang in whatever rivets are needed, and put a bit of bondo on the outside of the airplane to smooth out the surface damage. Yes, some modest weight penalty to do this way (anyone have an idea what a 1 m x 2 m sized piece of the fuselage would weigh?). This method seems more secure than cutting away all the carbon fibre material that, although damaged by the fire, almost certainly still has some significant strength remaining.

Speedbump

[MrSnuggles]

To Speedbump:

Yes, I have been thinking along the same lines as you, that is what makes this such an interesting case.

I really would love to know how they decided upon this approach. I am especially curious about the "plug from outside" technology. There are many pros to have a plug inserted from inside - especially in an airplane.

Plug from inside would require a sharp angle scarf joint (or something with these pegs, nibs, hooks etc) though, otherwise the material might weaken in the edges, enough to MAYBE compromise the plug enough to make it fail catastrophically. A blunt angle scarf joint (which is, I believe, the case with this plane) could theoretically vibrate the glue enough over time to make it unstable.

Now they plugged it from outside. The theoretical case with the glue still stands but they removed the feature for the plug to "plug itself" into the hole. Speculating here, maybe that is to reduce compression stress on the glue. This would correlate well with the patch from inside that supposedly could reduce the effects of pressurisation (which would cause a tension stress on the glue), thus trying to keep glue properties stable. Just speculation from my side.

Thoughts on this?

[A and C]

Speed bump

You make the assumption that the plug would be fitted from the outside, there is no technical reason with a well executed scarf joint that this should be any weaker than the original structure.

There seems to be the assumption on these pages that a scarf joint is weaker than the original structure....... It is not weaker or stronger it is exactly the same.

My preferred option would have been to fit the new section as a plug from the inside as this would be less disruptive to the outside of the aircraft.

All this assumes than there is enough room to get the required surface area to perform a scarf joint.