Boeing new issue with 777X engine support

[Ronen E]

My comment is about speculation about what Boing might have done wrong (shortcuts, etc.) there is way too much we don’t know. My point was in fact to redirect the discussion to the (physical) causal investigation process.

Exactly. We don't know. Currently, there is no hard evidence that ANYTHING in the design process was done wrong, leading to the cracked link**.

This is not to say that Boeing is okay or not; I'm hardly the person to comment on that. Regardless, from an outsider mechanical design engineer (with hands on experience in product development) perspective, I don't see what's wrong with the DESIGN DEVELOPMENT process here. Validation is in progress; an issue was identified (IMO, in an appropriate manner - we could always hope for earlier discovery, but realistically there are objective limits); now the design/engineering team* needs to work on the solution. That's it.

Let's please also not forget that the plane landed safely (with the cracked link). It's not like the engine dropped out of the sky. To me that's an indication that the design - overall - had enough safety headroom (whether that was achieved through redundancy, or otherwise). The design team should actually be commended for that.

*) IMO, the original one.

**) Pre-empting @Tidge - my practical engineering experience shows that cracking (contrary to ductile failure) can be very difficult to predict. The fact a crack was found is NOT evidence that the designers/engineers did not do their work properly. It just means it can be really difficult to get it right, and more than one iteration might be required. I also strongly agree with @Bev D that simulations and calculations can sometimes only take us so far; and sometimes real-world testing is the only way to find out (which implies sometimes there will be breakdowns during testing).

 

[Ronen E]

In my opinion, speculation about errors of commission and/or omission are valid subjects as PART of a root cause investigation. Some examples I have seen or found in some.cases:
Reliance on test results in metallurgical samples from vendor labs labs later found to be easily discoverable with random exams of duplicate analysis reports;
Omission of steps in written assembly procedure;
Incidents of installation of parts in wrong orientation (missed in design FMEA);
Concentrating on material spec failure when true problem was error in load factor in design.

If the root cause investigation is genuine, and held in an org culture where getting to the bottom of things (physics/engineering wise, not politics-wise) is paramount, all of these will trickle down (as relevant) from studying the failed part - first, as installed (still on the plane); then in the labs/office.

If this is not the culture, however, the results will be random, and so will the effect on future safety.

Errors of engineering/manufacturing commission/omission are certainly a legitimate interest in root cause investigations. However, I think what @Bev D was saying was that speculations around whether those were "natural" errors (which still require addressing, in a constructive manner) or the result of a shortcutting / cover-up atmosphere under commercial pressures, is besides the root cause investigation point and only distract from the engineering essence.

 

[Miner]

The fact a crack was found is NOT evidence that the designers/engineers did not do their work properly.

True. In analyzing many broken motor shafts, we would find a stress riser caused during the machining of the shaft. These were often a missing or undersized fillet radius that concentrated the normal stresses causing a crack to start leading to a fatigue failure. There was nothing wrong with the design.

@Bev D Here's your physics!

 

[Bev D]

I had a transfer case part that was in pre-production and everyone of the things cracked during the final heat treat. turns out the forging was contaminated (dirty) and every crack initiated at a contamination particle. The contamination was due to procuring the material on the ‘grey market’ as this particular material was constrained world wide. (The constraint was expected to clear before job1 manufacturing began). There was nothing wrong with the design, nothing wrong with the heat treat parameters, nothing wrong with the machining processes. A simple cleaning process was implemented and we passed the preproduction qualifications.

in the 777X crack there may be unexpected stresses induced by the geometry and material properties of the plane itself. And on and on.

Each of these scenarios changes “who is to blame”. So any discussion regarding “who is to blame” is mere speculation (finger pointing) until the actual causal mechanism is known. Boing needs to focus on the causal mechanism investigation. the strategy and tactics they select now are critical as the number of failures may be low…

 

[Sidney Vianna]

There are a few places where a discussion is being held about this issue with people much more knowledgeable on the specifics of the parts in question. Interesting to note that some speculate that the cracks would not be on the rod, but on the eyes that connect them - apparently there are weldments, which by nature are much more prone to crack initiation and propagation than a forged titanium rod.

https://www.reddit.com/r/aviation/comments/1ewfcpv

For another discussion with pictures of the engines and thrust links, peruse https://www.airliners.net/forum/viewtopic.php?t=1489789&start=650

 

[Tidge]

I had a transfer case part that was in pre-production and everyone of the things cracked during the final heat treat. turns out the forging was contaminated (dirty) and every crack initiated at a contamination particle. The contamination was due to procuring the material on the ‘grey market’ as this particular material was constrained world wide. (The constraint was expected to clear before job1 manufacturing began). There was nothing wrong with the design, nothing wrong with the heat treat parameters, nothing wrong with the machining processes. A simple cleaning process was implemented and we passed the preproduction qualifications.

...and this is why I suggested that the original design team be directed towards investigating the sourcing and assembly of their design output. At no point did I suggest kicking them off the project. I did imply that the original team may not be a good fit for re-reviewing their original design outputs.

 

[Ronen E]

...and this is why I suggested that the original design team be directed towards investigating the sourcing and assembly of their design output. At no point did I suggest kicking them off the project. I did imply that the original team may not be a good fit for re-reviewing their original design outputs.

Duely noted and respectfully disagreed.

BTW, I don't know your background, but typically mechanical design engineering and, on the other hand, sourcing and setting and managing assembly are two distinct and quite diffetent specialties. This is especially true for something as complex as a modern jet engine.

 

[Sidney Vianna]

So many good posts on the Reddit thread I linked above. One of the posters stated

Test flights are going to result in findings and design revisions, that’s what they’re for. That’s the whole point.

I wholeheartedly agree with that. And will repeat, test flights subject the aircraft to extreme conditions which would (normally) never be encountered during a regular flight (with passengers). On another "paywalled" article I read the flight immediately prior to the cracks being found subjected the craft to a series of aggressive thrust maneuvers at 38,000 ft.

 

[Bev D]

OK so this is a thrust link on the engine (designed by Boeing not GE) and there are 3 planes with the stress crack. But there is really not much that we actually know for sure at this point.

But a certification flight isn’t necessarily as stressful as early test flights. Certification flights are intended to validate that the plane operates safely for its intended purpose. While I am a big proponent of worst case (but not foolish) test conditions it is not clear that all of the planes with cracks (if indeed there are 3 planes with cracked links) were subjected to ‘aggressive’s stresses that would never be encountered in real life.

 

[Wes Bucey]

I am heartened that other professionals with access to some facts about the situation are discussing. It and are apparently monitoring the progress.

Thanks Sydney, for this input.