A380 Broken Wings

[vpcaptain]

Just read this mornings Flight International and Noel Forgeard at EADS/Airbus is adamant that his ruptured A380 wing will not have to be retested even though it failed Below the ultimate Design Limit.

Any thoughts?

[GearDown&Locked]

Yes, I've been thinking... and... it felt funny!:}

[CargoOne]

If it fail again on a another test it would delay the first deliveries for another year or two. No test - no delays - no problems. Happy flying!

[mary_hinge]

http://thestar.com.my/news/story.asp?file=/2006/2/20/apworld/20060220202949&sec=apworld

Airbus official says A380's failed wing test will not delay delivery

SINGAPORE (AP) _ Airbus's chief operating officer for customers said Monday the first delivery of the world's largest commercial jet, the A380, would not be affected by the failure of the jet's wing in a stress test last week.

"It's not a big problem at all in fact. We hope and expect it to be relatively minor,'' John Leahy said on the sidelines of an Airbus briefing ahead of the Asian Aerospace exhibition.

Before certifying an airliner for commercial service, aviation authorities set its "limit loads'' - the maximum strain each part is likely to be placed under during extreme turbulence or hard landings. New planes are required to resist loads of 1.5 times the limit.

The A380 wing had ruptured sooner than expected during ground testing in Toulouse on Tuesday, reaching only 1.45 times its limit load before breaking - 3.3 percent short of target. The wing had been bent upward by 24.3 feet at the tip.

Leahy said there should not be any major modifications to the design of the plane. "We have enough data from that test to know what needs to be done,'' he said. "This should have no impact on the delivery of the aircraft.''

Seven airlines in the Asia-Pacific region have ordered a total of 49 A380s, accounting for 31 percent of 159 firm orders so far for the world's largest passenger aircraft - but deliveries have been pushed back six to eight months due to production delays.

The 555-seater is due to enter into service with Singapore Airlines Ltd. by the end of the year. The aircraft has a list price of US$292 million (euro243 million).

The A380 will overtake The Boeing Co.'s 747 as the world's largest commercial jet when it enters service. Boeing announced late last year that it is launching the 747-8, a more fuel-efficient version of the double-decker plane, to compete with the A380.

Separately, Singapore Airlines said Monday it was evaluating proposals from Boeing and Airbus to supply long range aircraft, but a decision is unlikely anytime soon.

"We are evaluating (the proposals). The early submissions didn't meet our requirements in terms of delivery schedule and operational economics,'' Chief Executive Chew Choon Seng said on the sidelines of an aviation conference.

Singapore Airlines asked Boeing and Airbus in August to submit proposals for ultra-long range aircraft, to meet its fleet renewal needs and future expansion plans.-AP

[blackmail]

hello every one,

it is as simple as that: the test article wing failed below certification limits. OK. So be it. that's what testing is for.the only solution is to test a new/modified/beefed up/production wing to destruction & do it as many times as required until the thing passes the requirements. lobbying the certification authorities is definitely only going to strengthen the design on paper. that it will cost a lot of money & induce further delays in certification is for shure, but totally irrelevant. if i were mr. noel foregard i might have a word with my chief structural/stress design engineer.
last remark: i do not buy the argument that the tested wing is not identical to production wings. it's like testing a volkswagen to destruction & then going to mercedes to tell them they have a good car.

[Jetstream Rider]

i do not buy the argument that the tested wing is not identical to production wings. it's like testing a volkswagen to destruction & then going to mercedes to tell them they have a good car.

Why not? These things are not made by someone spreading glue with a stick and "adding a bit more" for the test article. They are made in jigs, very accurately and with some pretty amazing machine tools as well as some hand work. What exactly will make a difference between a test specimen and a service specimen? A Volkswagen and a Mercedes are different shapes for a start! (insert irony mark).

You could restrict the A380 weight - which then means the aircraft is strong enough, as the load has then been reduced (which means no further testing, but I think this is unlikely). The mode of failure is also important (and I do not know how it failed), but if a part of the wing failed, leaving the main structure intact, then maybe only that part needs alteration. I would be interested to see the test results, but imagine these are kept fairly tightly under wraps.

[barit1]

The test was undoubtedly planned to demonstrate 150% of some Max ZFW limit. As it happens, it failed a few percent shy of that value.

Another way of stating this is that Airbus have demonstrated 150% of some lower ZFW value (~3 to 5% reduction). Assuming all other tests are met, the initial aircraft could be delivered with a reduced ZFW. This would imply some financial penalty for Airbus because of the reduced profit potential for the airline.

[Gillegan]

Given that weight is such an important design element of an aircraft, ideally you would want the wing to fail at 150.00001 % of its design load. If it fails at 160%, you are probably carrying excess weight. I wouldn't be surprised if Airbus was purposely cutting their margins very fine in order to save weight. If the wing failed within a few percent of the target and if the results did indeed validate their finite element computer model then I would think that while structural modifications would definitely be required, a retest would not.

[blackmail]

hello jetstream,

you ask:"why not," then i can ask:"why, yes?". if the load bearing stucture of the wing is not identical to production types, what are we talking about? and, yes, a VW has a different shape than a mercedes, but that was just the point i wanted to highlight. that airbus now, reluctantly, has to introduce weight penalties is ok, but in my opinion it still needs a full retesting & please that no computer guru/engineer tells me that recalculations only would be sufficient.

[Gillegan]

The certificating authorities are going to want to see their computer models. Presumably, Airbus will now refine their computer models based on the results of the static test. Those refined models should be able to predict the results that the static test produced. If they can, then the authorities (JAA/FAA) should be happy to use those models to validate the structural modifications that Airbus most definitely have to produce (with the resultant weight penalty that Airbus is conveniently leaving out of their press releases).

[blackmail]

hello gilligan,

if i were the certification authority, i would be happy to receive/review airbusses new computor models, but i would still insist they break a new wing in a full static test rig. amen.

[Grunf]

Hello all.

Blackmail:

In general CAA responsible for certification does not ask for additional tests unless there is a big problem. This problem can be resolved "on paper" or by adjusting the FEM.

It can result in some weight penalty but it might not since they (Airbus) can change their assumptions which will create a small change in loads which will result in a positive ore zero margin.

Anyway, we should not forget the impact of Airbus on EASA similar to the same FAA-Boeing connection. it is hard to say that this has nothing to do with lobbying etc.

Politics, politics:E

[Jetstream Rider]

Blackmail - I see what you mean, I thought you were saying that the test specimen would be stronger than a normal wing so breaking it doesn't prove much. Here's why yes!

As I said before it depends on the failure - for instance an aircraft may have a low Velocity never exceed (Vne) because the pitot tube breaks. Fix the pitot tube and the Vne can be increased - in other words a low Vne doesn't mean a weak aircraft. Similary a failure of an Airbus wing doesn't nescessarily mean a failure of the whole structure. It might be that the flaps broke off so the flap attachment needs further work - rather than the wing box itself - this would of course mean further flap tests, but not nescessarily further wing breaking tests. Without further info it is hard to judge.

Its is easy for Airbus not to incur a delay - bring the aircraft into service with a reduced ZFW, then fix it sometime later. It depends on the cost of delays, versus the cost of a lower ZFW.

[blackmail]

hello jetstream,

sorry, yes of course, if i am correct, airbus says production wings will be stronger than the test specimen that was broken. and again, if i can recall, the rupture was in significant, loadbearing structure.

[Jetstream Rider]

Fair enough, be assured though that failure of structures is a lot more complicated than just "it broke" - there could still be a lot of things that would mean further destructive testing is not required. conversely, there could also be a lot of things that mean it is required. I would be interested to get more details on this - do you have any idea where I might discover more?

[CargoOne]

Although I'm not a specialist in rocket science (or aircraft design) but generally speaking if certification authorities can accept the computer-generated changes into wing design as the proof of compliance to the standarts without any further actual tests why those tests have a place at all? Why not to accept the whole construction as "safe" based just on these computer models?

[kansasw]

Although I'm not a specialist in rocket science (or aircraft design) but generally speaking if certification authorities can accept the computer-generated changes into wing design as the proof of compliance to the standarts without any further actual tests why those tests have a place at all? Why not to accept the whole construction as "safe" based just on these computer models?

I am likewise neither of the above, but think I can offer at least part of the answer: The whole construction is a very complex system and the real-world/destructive testing may elicit flaws resulting from interaction of various parts of the whole, that the engineers did not anticipate. If destructive testing reveals a flaw that can be clearly identified as "simple," and confidently judged to have no interaction with other parts of the whole, and reliably reinforced to meet requirements by beefing up a single component or a few, then it seems to me that computerized or slide-rule engineering could supply a reliable result. If contrariwise the flaw cannot be identified as "simple," then it's back to the drawing boards. The more large-scale and complex the fix, the stronger the reason for a full-scale retest.

[Atlanta-Driver]

Was that it could not perform as published causing Singapore Airlines cancel the order and go for A340's instead. 3.3% does not sound much but it is a huge number. Deliver aircraft with lower than specified ZFW will have effects on it's performance and WILL cost Airbus money. What ever Airbus management says it is definately a problem. I am having a feeling of Dejavu.

AD

[Sunfish]

Oh dear! Pilots talking structural engineering. The A380 failure is immaterial because we are talking about failure within 3 % of limit load. This is supposed to be 150% of design load. If we assume that the A380's positive G rating is +4, then to even reach the design load would require an aircraft to do something like flying through a thunderstorm at absolute max gross.

To put it yet another way. If an A380 ever reached its limit load in flight, failure of a bit of wing structure (which is by the way fail safe, with redundant load paths) is likely to be the very least of your problems.

Pleases top bagging Airbus.

[ironbutt57]

Was wondering when somebody was going to clarify that sunfish, the question in my mind is, is it a certification requirement to reach 150% of design load for certification, or is this just an "industry standard" so to speak??

[DoNotFeed]

If you cant bend the wing - bend the rule!
@Sunfish - sorry for beeing a skipper, turbulence hits sometimes bcause there are some fency turbs you cant see or simply on long haul they are of considerable extension.
Some jockeys dont like to s****t their pants, PAX too.
best regards and fix:ok: the ship

[blackmail]

hello sunfish,

tell me why pilots may not talk about structural engineering matters? after all, we sit upfront & have first hand experience how the aircraft behaves in the real world. i agree engineers should design safe aircraft & pilots should fly the the thing safely as well. problems arise if engineers want to fly aircraft & vice versa for the pilots. but that doesn't mean everyone might not express his/her opinion about the matter

for info : designload for public transport aircraft= -1/+2.5g
ultimate load = +50% of designload= -1.5/+3.75g


and fully laden aircraft have & will fly through TS(damn fool pilots that is)

[Mariner]

Quote: "If we assume that the A380's positive G rating is +4, then to even reach the design load would require an aircraft to do something like flying through a thunderstorm at absolute max gross."

Sunfish, A380's will fly thru thunderstorms, sooner or later all planes do. Not on purpose, but for one reason or other.
And about wing-bending moment; I'd rather to be at "absolute max gross". All that fuel in the wings will help reduce the wingbending moment.

[superpilut]

This world is absolutely nuts.
If my pass score for a test is 75% I fail if I score 74%.
Not only would I be axpected to read a few more pages which -in theory- should give me the required 75%; I would be forced to redo said test!

So tell why it is more important to score a theoretical test within limits then to have a structure of an aircraft keep it up to the limits?

:confused: Explain that one to the "Daily Mail" after the first failure of a wing with pax on board..:sad:

[blackmail]

hello every one


i agree 100% & more with donotfeed & superpilut

[Charly]

It´s all too theoretical :8 .... This 150 % bending aplied smoothly and with a careful increase in intensity is surely a have to have test, but it isn´t the proper simulation of flying through a TS.

I guess that other tests like constant vibration with short & harsh jolts over longer periods of time have been made and were succesful. Tests like these will determine the actual durability of the wing.

But does anyone know if these 150 % are an industry standard by Airbus, or if it is an actual JAA/FAA requirement?

[vapilot2004]

Will the FAA require them to crack another or force a reduction in ZFW, being as the failure point was outside the allowed margin ?


And I Just wanted to say...........

Many headlines make it seem as though the wing test was a complete failure and the stories don't mention the fact that the wing was supposed to break.

Even though Airbus is so :mad: secretive, when official news does reach the light of day, the media always makes such a sensation with grossly misleading headlines.

This kind of crap reporting doesn't happen with Boeing in the development stages - they tend to get picked on later. (for whatever reason):hmm:

[Jetstream Rider]

Sunfish - whats wrong with Engineers talking about structual failure? I am one. I happen to be a Pilot too.

Airliners are not designed to 4g, the usual limits are what has been stated above.

The fact is that the 150% is an aim and the wing failed that aim. Certification requirements are there for many reasons - and this wing didn't make it. As I said above though, it could be a simple fix, or indeed something much more serious. We are not saying the A380 wings will fail in light turbulence, or even in a TS. What we are saying is that the aircraft hasn't met the requirements - the reason for this is unclear and so the ramifications of that are unknown. Could it be that the aircraft will fail in a TS one day after 20 years of service, or could it be that the failure is easily fixable and nothing to worry about? For your info, severe turbulence can be encountered without warning, in clear air and have a very sudden onset - its not just about TS.

Superpilut - your analogy is a bit too simplistic. Structural Engineering is a bit more complicated than answering a few questions on a piece of paper. It depends why you failed the test. A good examiner will write an exam that tests all your areas of knowledge - if got your 74% by demonstrating good knowledge of all but one of your subject areas, you may only be re-tested on the one you failed. Do you see the difference?

Daily Mail readers - well you just can't explain to them, that's half the problem!

[blackmail]

hello everyone

it is the first time since long that a new project of that size failed such an important & spectacular structural test. but, then, when was it the last time? i sadly remember the comet story.

[Old Aero Guy]

It´s all too theoretical :8 .... This 150 % bending aplied smoothly and with a careful increase in intensity is surely a have to have test, but it isn´t the proper simulation of flying through a TS.
I guess that other tests like constant vibration with short & harsh jolts over longer periods of time have been made and were succesful. Tests like these will determine the actual durability of the wing.
But does anyone know if these 150 % are an industry standard by Airbus, or if it is an actual JAA/FAA requirement?

The 1.5 factor is an EASA/FAA certification requirement.

[lomapaseo]

I'm in partial support of Sunfish on this. This is a complex subject which goes far beyond simple structural engineering discussions and involves judgements within the whole basis of the actual regulations themselves, of how much, how often compared to all the other competing risks to flight.

Most of the preceeding discussion are of a brevity that leads for lots of misunderstanding between posters.

In my opinion, you really can't do justice to a reasoned discussion simply using a keyboard driven forum.

I don't mean to abort this discussion but only to caution against simple conclusions for or against since clearly all the facts (rule intent and design intent) are not in evidence.

[blackmail]

iomapaseo,

it is simple: the thing failed a mandatory certification hurdle. noel says he will not break a new wing, therefore lobbying, computor modelling & untried beefing up solutions will be in the order of the day. american alligators(lawyers) are already in the starting blocks for the day that it will suffer a major break up(heaven forbid, i hope, if there is one).

[SLF3]

I remember seeing a series of television programmes on the 777, in one of which they showed the wing tested to destruction by application of hydraulic rams across the surface. From memory, the test criteria was 1.5 x design load, so this is a standard number. The load was applied slowly.

I would suggest the issue of tolerances, live versus static loads, accuracy of computer models etc. is the reason for the 1.5 factor. This is a considerably higher factor of safety than is applied to other structures which are not tested (such as support structures for deepwater oil production platforms).

[Curious Pax]

From time to time you see reports about various airliners being offered with an incrreased ZFW, as part of a package of improvements to its performance. Does the discussion here suggest that:

a) they test a new wing to destruction to satisfy the 150% rule, or
b) the original wing was tested to a higher level (say 160%), and so the new ZFW still falls within range?

Apologies if I am merely demonstrating that a little knowledge is a dangerous thing, but it's been bugging me!

[Grunf]

Hello all, Sunfish especially.

OK, let us clarify some things related to certification:

Limit loads = the maximum loads to be expected in service - FAR 25.301 (a)

Ultimate loads = limit loads multiplied by prescribed factors of safety - FAR 25.301 (a)

Factor of safety = 1.5 (unless stated otherwise) - FAR 25.303

therefore:

Ultimate load = 1.5 x limit load

There are no Design Loads defined in certification documents.

In order to release the first FTA (Flight Test Article) to start flying, usually, CAA responsible for certification asks for completion of a batch of limit loads tests.

It is up to CAA to define, according to the relevant regulations and its judgment, which tests are required.

For the completion of certification process all the ultimate loads tests have to be satisfactorily completed. It is up to CAA, again, and the responsible party (OEM) to agree on what has to be done if there is some unacceptable test result (Margin of Safety < 0).

That can be a repeated test, a change in FEM, a weight penalty or whatever they agree on. It is FLEXIBLE.

As for the statement that structure is fail-safe this is wrong since this concept was left in early 70s with introduction of Damage tolerant structure.

Last fail safe model was 747.

Cheers

[BenThere]

I don't like the logic of 'When will you ever go through a thunderstorm at max gross weight?'

These aircraft will fly for decades and will age through the process of thousands of pressurization cycles, firm landings, and wing flexes. Their initial structural strength will deteriorate over time. Initial certification testing's purpose is to assure structural integrity over the design life of the aircraft.

Design tests are mandated to prove the design concept under actual physical stresses. This isn't a Boeing vs. Airbus issue, but one of aviation safety.

[pax2908]

Please pardon (again) my ignorance. I understand that the +50% factor
imposed by the regulations may be somewhat arbitrary. What is actually done to confirm or revisit this number as a reasonable safety margin?
How close are real dynamic situations (turbulence encounters) to the static loads which are imposed during the ground tests?

[hobie]

Safety Factor, is a multiplier applied to the calculated maximum load to which a component or assembly will be subjected. Thus, by effectively "over engineering" the design by strengthening components etc …..

Aerospace engineering where safety factors can be as low as 1.10 or 1.25 (in extreme cases like unmanned vehicles) or more typically 1.50, are relatively low because the costs associated with structural weight are so high. This low safety factor is why Aerospace parts and materials are subject to much more stringent testing and quality control …..

One can easily design an Aircraft with for example, a safety factor of 5.00 ….. PAS loads would be amazingly low ….. fuel consumption extremely high and range well below needs …. Probably impractical …….

The 380 tests being so close to the design objective of 1.50 is rather impressive if one thinks about it ….. it’s just a pity it is slightly below target rather than on or just above ,,,,

[TightSlot]

Please excuse a question from an unqualified person, but I'm struggling with the basic concepts of this thread. If something is tested, against an established pass mark, and then fails that test, then it has failed to pass. The test can be re-taken, in the hope of meeting or exceeding the pass mark, but until the pass mark is achieved, it is not a pass: Surely, what shouldn't happen is that a negotiation takes place as to the validity of the pass mark, or a pass is awarded on the basis of a promise of future work and improvement?

Clearly I'm missing something - maybe my view is simplistic? I would appreciate clarification...

[Grunf]

TightSlot:

It does go more or less along the lines I have already posted.

You do get a result based on the initial set of assumptions.

If you do not have the good result you change the assumptions.

One thing you can change is your model (FEM = finite element model - the "wire" model that in specialized software (FE) simulates the structure exposed to conditions it will see in flight).

The closer are your simulated results to tests the better it is. If you are close on the lower side that gives you few options:

1) Change the model so it can better represent the real conditions

2) Declare a weight penalty (ouch!) and pay the price

3) Reinforce the failed part of the structure and redo the test (tooooooo expensive)

Also a certain level of "convincing" and political approach is always present. All this rules can be interpreted in different way. What I quoted before from CFAR 25 is just a small portion of all that has to be satisfied in order for an aircraft to get its CofA (certificate of airworthiness).

it is up to a local CAA (regulatory body) to make that decision.

Therefore, it is not a straight yes or no answer.

[TightSlot]

Thanks Grunf

[Packsonflight]

When Douglas did the load test on the C17 .the wing failed a bit short of the ultimate load limit. I do not remember the excact number but after a few weeks the company decleared that the wing had actually passed and the test was wrong

[hobie]

and do let us remember the wing was tested to destruction ...... it failed at a stress level 45% greater than anything it will endure under all operating conditions ...... That's why the Airbus Engineers are not jumping off high buildings ... :)

[Sunfish]

Thank you for explaining the latest details Grunf. I watched Boeing build the first 767 and haven't been involved in a new aircraft program since.

When watching what happens on one of these tests, the wing doesn't "break off" some part of it fails and the wing starts flexing again, relieveing the load. All you hear is a loud bang, then get out the calculator and find whats busted and where.

I wouldn't know the changes Airbus will need to make to fix this, but it could be as simple as adding a few pounds by making a component a fraction of a millimeter thicker, or changing a simple design detail like moving a hole or changing a radius. It is even possible that the strength may be improved by removing material if the failure was caused by an unsuspected stress concentration, as was necessary on the old F27 landing gear. I guess i'm showing my age......

[Grunf]

Hello.
Sunfish, hats down to you. I always respect the "old" guard. Where would we be without you guys ?

As for the possible changes I don't think it would go that far as adding or removing something. Updating FEM would do the math. I would be surprised to see the weight penalization.

But still all this is hypothetical since we do not know what was the failure and under which condition.

I do not think that Airbus will release that info.

Cheers

[wiggy]

Just to add - if the guys flying the 380 South of Toulouse yesterday are reading this (perhaps someone can pass it on): The big beast sure looked impressive, especially with the bank wound up to what looked like at least 60 degrees...I had to stop the car to admire the "show", keep up the good work.

[blackmail]

hello grunf,

quote from grunf:" last fail safe model was 747".

grunf, in that case, well... , give me a 747 forever!

[Grunf]

Blackmail,

well yes structural analysis advanced since the late 60s (747) so nowadays in application is damage tolerant structure.

How good it is? Well, the aircraft are lighter, definitely, since you do not need to put all the redundant load path (additional structure that will carry the load if something fails).

On the other hand it is so hard to do a good prediction with damage tolerant structure on when and where it is going to fail so it can be prevented. If you don't have a good MRO it can go bad, really bad.

Anyway, inspection intervals for structure are set in a conservative manner in order to cover for all eventualities.

Still...:hmm:

This methodology was always prone to different interpretations.

Let's hope that 380 will get out of flight test without big problems...

[Hunter58]

Strange enough nobody ever said what the basis for that test is, and therefore what the 150% represent.

Considering that Airbus has a freighter variant on the line ans some people have been talking about a stretch one could actually consider that the tested load limit was based on a highter weight than the 'standard' aircraft has.

[Grunf]

Hunter58,

you might be right.

However since freighter version usually has different floor structure plus a completely different loading it is possible they will certify it later as a derivative.

One way would be to follow what you have said (for stretch it is the same thing).

They do talk a lot about that which is unusual. Again, they do not share what the real reasons are so at the end, I believe, they will certify the aircraft without problems.

[ElectroVlasic]

TightSlot:
It does go more or less along the lines I have already posted.
You do get a result based on the initial set of assumptions.
If you do not have the good result you change the assumptions.
One thing you can change is your model (FEM = finite element model - the "wire" model that in specialized software (FE) simulates the structure exposed to conditions it will see in flight).
The closer are your simulated results to tests the better it is. If you are close on the lower side that gives you few options:
1) Change the model so it can better represent the real conditions
2) Declare a weight penalty (ouch!) and pay the price
3) Reinforce the failed part of the structure and redo the test (tooooooo expensive)
Also a certain level of "convincing" and political approach is always present. All this rules can be interpreted in different way. What I quoted before from CFAR 25 is just a small portion of all that has to be satisfied in order for an aircraft to get its CofA (certificate of airworthiness).
it is up to a local CAA (regulatory body) to make that decision.
Therefore, it is not a straight yes or no answer.
I thought option 4) was

Update your software model using the data you gathered while breaking the wing
Now that you have a good model, replace the part that broke with one that will not break based on the model
Re-run the model and prove the part won't break
Ask FAA/JAA to sign off on the test

[Grunf]

ElectroVlasicu,

It is OK what you suggest but it is costly. Namely item 2 (replacing the part that broke) is acceptable only if it is the SAME part (same geometry and specifications i.e. materials).

One more comment, of a general nature:

There is no wing braking before the whole certification testing. If and when the test is finalized and all certification data completed and accepted by a local CAA, then it is possible to go through with wing braking.

This is not a part of required ultimate loads test.

So, Vlasicu, your suggestion is already somewhere in between the options 1, 2 and 3.

Cheers

[Snigs]

As a stress engineer who has performed lots of FEA, most of which has been in support of certification and some of which has been in support of failed tested items, I can confirm that ElectroVlasic's option 4 is basically how it works.

If you take the example of the wing, if you can prove the load path and boundary conditions of the area surrounding the failed component (for example by test data from strain gauges and the like) then you can generate a finite element model of another component of whatever design, and providing the interfaces are the same apply the known loads and boundary conditions and bob's your uncle. The certifying authorities need to be convinced that your models are correct, and if they are then no further physical testing is required.

[lomapaseo]

I thought option 4) was

Update your software model using the data you gathered while breaking the wing
Now that you have a good model, replace the part that broke with one that will not break based on the model
Re-run the model and prove the part won't break
Ask FAA/JAA to sign off on the test



I also have lots of experience in certification and failed test items and would offer the more likely path as:


Update your software model using data gathered just before the wing broke

Now show that the model was misapplied to the end conditions (the exterior cells to your FEM) ala the points of loads including pull wires, weights etc.

recalibrate your model by adjusting these end conditions until the model now predicts a margin beyond where the initial demo test failed

convince the authorities JAA etc. that the initial loading condition was misapplied and that since you are now so finely callibrated via the intial results up to the failure condition that it's a waste of time and money just to repeat the test and scrap another wing under overload.

BTDT