747-8 certification

[Grunf]

Originally Posted by Porrohman

If the 737-400 had been required to undergo a more comprehensive set of certification trials before carrying fare-paying passengers could the Kegworth accident, and several other 737-400 incidents around that time, have been avoided?

I know there were other contributory factors to the Kegworth accident, but the chain of events began with an issue that, it could be argued, should have been detected during the certification process had "grandfather" rights not been applied.

Have the rules regarding "grandfather" rights altered since the Kegworth accident? If not, certain issues with the 747-8 might not be discovered until airline pilots and fare-paying passengers are on board, and one of the important lessons from the Kegworth accident will have been lost.

b.t.w. I've seen at first hand (fortunately in an environmental test facility) the effects that modifying a large structural component of an aircraft had on its resonant frequency and integrity. Our Client wanted us to significantly reduce the weight of the component, and this appeared to be feasible, however the weight reduction significantly altered the resonant frequency of the component. This resulted in all of the aircraft standard fasteners rapidly unwinding when the component was subjected to the expected vibration conditions. The fasteners unwound and the maintenance access panels fell open within seconds of the vibration test commencing. We tried beefier and beefier fasteners, but to no avail. It was amazing to see and surprised everyone except our chief mechanical engineer who was very experienced and had seen this sort of thing before. The solution to the problem was to offer less of a weight reduction, which resulted in a component that had a resonant frequency that was compatible with the rest of the airframe.

Porrohman.

Porroohman,

One of the certification issues mentioned above is the structure (which I have commented). Since Kegworth, as bad as it was, was an issue primarily with engines and systems design (fuel system to be more specific) it could be a completely separate topic.

As for the structure 'grandfathering" is not a big problem (see what flies above our heads, for that matter!). In case of new systems I think Boeing has learned its lesson with Kegworth, at least (as well as the rudder issues etc).
In case of vibrations, however, it depends if the piece you were working with was part of primary or secondary structure, if it was an SSI/PSE etc. In general and from what I've seen none of the OEMs (commercial) are having much concern over vibration issues. Sadly, but this is truth.

Of course they do verify maybe for the first 2 modes for the overall aircraft model (a very lumpy one, mind you) but I've never seen detailed work as it is necessary for military projects.

So, certification issues with "grandfathered" structures are usually with systems, not structure.

One gets to mind Bombardier's series of RJs (CRJ 200, 700, 900 and future "extended" 900) which are all DERIVATIVES of the original CL600 Challenger?! Talk about grandfathering, huh?

Cheers

[Tallbloke]

Is the 747 unique in having no forward exit for the front cabin?

[barit1]

Originally Posted by Jetstream Rider

The Ford model T was very successful, so why make it any better? Operations have been done without anasthetic for hundreds of years so we obviously don't need it?...

The reason the Model T was finally discontinued is that the market discovered newer cars to be more efficient, safer, more reliable --- and they quit buying the "T".

When the A380 demonstrates superior statistics in service, the falling 747 sales will tell Boeing it's time to start a fresh design. Until then, 1966 vs 2007 original cert standards will have little influence on the marketplace.

[Volume]

OK, it's only one instance so a very small survey sample, but didn't the incident with a UA Classic out of Honolulu go some way to indicating that a 747 could cope with a pretty big bit of damage (loss of a cargo door and surrounding skin), rapid decompression (which followed the loss of the cargo door) and sustained engine imbalance (3 & 4 out all the way back to HNL)?

The Aircraft in question was a 747-100, which means a lower load level anyway (lower MTOW) and much smaller engines than the -400 or -8, thus smaller loads due to sustained engine imbalance for a shorter time period (lower range).
The door took away a big part of the main deck sidewall, which was a good thing, as the main cabin decompression was through that hole, and not through the cargo door opening. otherwise the main deck floor would have most probably collapsed. (Not THAT bad for a 747, as the flight controls do not run in the main deck floor, but in the main cabin crown area, so not that fatal as for the THY DC-10 in Paris, the accident which forced the rules to be changed wrt. rapid decompression). Rapid decompression is most critical for the smallest volume (in that case for the cockpit), as the small air volume escapes quite fast for a given hole size, imposing a greater pressure difference end hence pressure load on the structure separating that area from the main cabin. A big hole in the cockpit roof may make the cockpit floor collapse into the cockpit, killing the crew and severing the flight controls running there. The 747-400 with the old european "4-country-certification" (it was pre JAA certification times...) has a reinforced upper deck floor to cope with the JAR decompression rule, which required a bigger hole to be taken into account compared to the FAR requirements those days.
For the damage tolerant structure, the China Airlines crash clearly demonstrated, that the rear fuselage is not able to cope with a certain crack size, the frames are not able to stop or bridge a crack growing over a certain length in the longitudinal direction. With respect to the widespred fatigue damage criteria taken into account today, the 747 will also have some problems demonstrating compliance.
To be fair, the 747 is not "1960Žs safety standards" today, as some measures have been taken retroactively to all aircraft, for example the fatigue and damage tolerance issues past the aloha-accident. So Boeing demonstrated, that the 747 can take a little more, than demonstrated according to the initial rules. But demonstrating that it is 2006 safety standard, looks impossible to me.

[Taildragger67]

Volume,

Thanks. Full & complete, QED!

Cheers
TD67

Tallbloke,
Yes AFAIK amongst current civils.

[Grunf]

Volume,

I quite agree with your statements, especially the China Airlines stuff. Again, this was covered with several SBs and consequently appropriate analysis (DTA including) was completed with new inspection intervals.

As for meeting the 2006 safety standards if we go back I would be ready curious to see how is Airbus going to certify the A330-200F as a derivative since it will be non-compliant to the new requirements, as well. Not on the level of 747 of course but still not meeting all the requirements (come to think the new EASA requirement of 2G Up for the cargo floors etc).

Also, a good example would be the pain Lockheed is seeing when trying to sell more Hercules' as they were never certified by FAA (nor designed per FARs). Someone these days requires that for the new sales and bang, a big problem is generated.

Grandfathering is a slippery road but there is always a way to go through, at the end. We can like it or not but IMO one needs to know the relations between Boeing and FAA and Airbus and vice versa EASA/JAA.

One would assume the impartiality in both cases but then if you work for some other OEM (Bombardier, Embraer) then you find out how these relations work . I had a painful experience with both JAA and FAA related to that and nowadays I can again see how it looks when you're on the "right side".

Cheers

[alexban]

As far as I remember Kegworth is a case of mistakenly identifying the faulty engine,so what has to do with 'grandfathering' ?
Are you talking about the fact that a blade ruptured in one engine? So..?
As I remember,they incorrectly checked the vibration indicators (low is good as opposite to all other instruments ) ,and also helped by the fact that the vibrations reduced with reduced thrust,as they level off and desended, they decided to shut-down the good engine. After resuming thrust ,the bad engine failed,so they remained without engines,and landed short.
So ,again,what's with the 'grandfathering'..?

[Porrohman]

Alexban;

Due to "grandfathering", Boeing did not fully test the new engine / airframe combination. Here's an extract from Wikipedia;

"Analysis of the engine from the crash determined that the fan blades (LP Stage 1 compressor) of the uprated CFM56 engine used on the 737-400 were subject to abnormal amounts of vibration when operating at high power settings above 25,000 ft. As it was an upgrade to an existing engine, in-flight testing was not mandatory, and the engine had only ever been tested in the laboratory. Upon this discovery all 737-400s (around 100 at the time) were grounded and the engines modified. Following the crash, it is now mandatory to test all newly designed and signficiantly redesigned turbofan engines under representative flight conditions."

The full article is here; http://en.wikipedia.org/wiki/Kegworth_air_disaster

In the Kegworth example, the application of "grandfathering" resulted in the engine problem that started the chain of events that led to the accident. The rules regarding engine certification were changed following Kegworth. What I'm not sure of is whether rules concerning other aspects of "grandfathering" have also been changed. If not, there is potential for "granfathering" to result in future problems.

Porrohman.

[lomapaseo]

Originally Posted by Porrohman

Alexban;
Due to "grandfathering", Boeing did not fully test the new engine / airframe combination. Here's an extract from Wikipedia;
"Analysis of the engine from the crash determined that the fan blades (LP Stage 1 compressor) of the uprated CFM56 engine used on the 737-400 were subject to abnormal amounts of vibration when operating at high power settings above 25,000 ft. As it was an upgrade to an existing engine, in-flight testing was not mandatory, and the engine had only ever been tested in the laboratory. Upon this discovery all 737-400s (around 100 at the time) were grounded and the engines modified. Following the crash, it is now mandatory to test all newly designed and signficiantly redesigned turbofan engines under representative flight conditions."
The full article is here; http://en.wikipedia.org/wiki/Kegworth_air_disaster
In the Kegworth example, the application of "grandfathering" resulted in the engine problem that started the chain of events that led to the accident. The rules regarding engine certification were changed following Kegworth. What I'm not sure of is whether rules concerning other aspects of "grandfathering" have also been changed. If not, there is potential for "granfathering" to result in future problems.
Porrohman.

The discussion is OK but the inference varies greatly with the subjectivity and the relavance to the thread title.
The words 'fully test" have a very broad meaning among certification engineers. Thus the word "garndfathering" has an equally broad and subjective meaning. The objective is to define the product in engineering terms such as vibratory stress and frequency over the complete expected operating environment. Then to provide for variations in tolerance (both mechanical and operational) to ensure that it is unlikely that the product will fail in this environment.
The certification standards provide boundries to the level of mechanical tolerance within defined levels of environment. The certification testing provides data for analysis to show that the expected variations in mechanical conditions in context with the expected environmental conditions will not result in a failure condition.
We have long since progressed way beyond the build-m and bust-m stage and that includes such margin tests as the blade-off and ingestion tests. Instead, all test data is considered as sucessful and applied to analysis. This is then taken together with past experience or history and can be interpolated in a far broader range than any affordable testing of the one product/environment example at hand could demonstrate.
So, in the lessons learned sense, the only useful tweaking of the meaning of "mandantory" is that all parts of the certification process need be examined when there is a shortfall and not necesarily just a simple test requirement.
OK, now to get back to the thread subject at hand.
Is there a shortfall? (if it ain't broke, don't mess with it)
Has the total industry wide experience indicated a change in the environment?
Is the previous test data and analysis still credible? or has the environment substantially changed?
For the questions above, what level of new data and analysis are necessary for validation?
Notice that I always use the term data and analysis together if one is to make any worthwhile subjective opinion.

[Porrohman]

lomapaseo,

I agree with your comments. I see merits in "grandfathering" where it's appropriate and I see merits in "fully test" where the data from previous tests might not be relevant to the modified aircraft.

The manufacturer and the certification authority need to apply very careful judgement in relation to where they apply "grandfathering" and where they decide to "fully test" to obtain new data for analysis. I expect that this will involve educated, well intentioned but subjective judgements in relation to certain issues. In the case of Kegworth, these educated, well intentioned but subjective judgements were, with hindsight, flawed. Some might argue that the different vibration characteristics of the new airframe / engine combination should have been foreseen and that new data for analysis should have been obtained before releasing the 737-400 to airline pilots and the travelling public.

We rely upon the certification authorities and the manufacturer to make the correct decisions in relation to this matter and we trust they will have learned from the Kegworth accident.

Porrohman.

[alexban]

If you have some time it may prove a very interesting reading of the kegworth accident report (he 638 kb pdf file ) ,found here ..http://www.aaib.gov.uk/sites/aaib/pu..._90_502831.cfm
121 very interesting pages