FAA / EASA's policy prohibits flying with known cracks in primary structure 2

[Ruag]

Hello everyone!
Does anybody know if exists a similar advisory circular to AC23-13a for large A/C also?
Thx.

 

[rb1957]

i understood flight with known cracks was acceptable if the structure demonstrated residual strength under ultimate load for the crack length (obviously factored) anticipated at the next inspection ... which is one of the exemptions, see AC23.13A,para6.3a(2).

For a large part25 a/c the situation should be dealt with by the maintenance manual. i'm sure most OEMs say "inspect the structure. if you find a crack, call us" there might be some provisions in the SRM to deal with specific situations.

Quando Omni Flunkus Moritati

 

[MNLiaison]

FAA Policy Memos;

http://www.airweb.faa.gov/Regulator...f30163d786257bac005bcbb7/$FILE/1986-00055.pdf

http://www.airweb.faa.gov/Regulator...b5fd1f2e86257ba20064bc7d/$FILE/1983-00043.pdf

http://www.airweb.faa.gov/Regulator...74b99e5686257b9f00600f9e/$FILE/1983-00101.pdf

Regards

 

[Ruag]

Thanks MNLiaison,
Interesting links.

Thx to rb1957 too.

 

[blakmax]

I am intrigued by this discussion from the FAA in the links, albeit that it is somewhat dated.

If the FAA does not accept the concept of flying with known cracks in primary structure, then what is the purpose of 2x.573 which mandates demonstration of damage tolerance? Surely if it has been demonstrated by testing or analysis that a PSE can sustain ultimate load in the presence of a defect of a known and demonstrated size, and analysis or testing can demonstrate that the defect will not propagate to a size which compromises ultimate load capability within a defined inspection regime, then surely the application of a blanket "no cracks" policy is contradictory to the intent of FAR 2x.573???

Regards

Blakmax

 

[rb1957]

the point of damage tolerance is to ensure that the a/c can sustain cracks that are expected to develop in service, and that a/c safety is assured by an inspection program, analytically derived and not based on service experience.

once you find a crack you need to repair it, as you now know that your structure deviates from the type design (which doesn't have the crack). if the crack can sustain ultimate load then the structure is still compliant. this is not the typical application of DT, whihc is more like "if you find a crack, repair it" ... kinda obvious.

btw it is 25.571 (25.573 used to exist, but it dealt with fatigue of landing gear) and 23.571, 572, 573, 574, and 575.

Quando Omni Flunkus Moritati

 

[blakmax]

OK, 2x.573 is specifically related to damage tolerance. 2x.571 actually calls out 2x.573 for this purpose. 2x.571 appears to relate to fatigue and fail-safety, but 2x.573 specifically deals with damage tolerance and hence discussion of the existence of cracks is more relevant to 2x.573.

I have a high level of interest in this topic in relation to the applicability of adhesive bonding for repair of cracks. There is a considerable amount of experimental and analytical data which shows that bonded repairs are extremely effective in reducing (or even arresting) crack growth in metallic structures. There is also some experimental and analytical data which shows that removal of cracks or stop-drilling prior to application of a bonded repair actually reduces the fatigue life of structures repaired using adhesive bonded patches compared to leaving the crack unmodified prior to repair. There are technical reasons which I can elaborate if there is interest.

In the past on military aircraft I have had to deal with the "blinkers on" attitude that cracks must be removed because certification was undertaken in the absence of cracks. The FAA memos quoted show that this is not entirely the case. Their requirement is that the cracks are repaired. It does NOT stipulate that the cracks are removed. Surely if it can be demonstrated by analysis that the stress intensity after repair is well below that which would risk failure at ultimate load and it can be demonstrated that crack growth can be managed on a safety by inspection basis, and there is data to show that bonded repairs without crack modification give the best fatigue outcomes, then why should the cracks be removed?

This assertion clearly depends strongly on the validity of the design and process certification for the adhesive bonds. I have addressed these issues in other postings on this forum, but I would be happy to elaborate if required.

Regards

Blakmax

 

[rb1957]

i know i'm being a "dick" about this , but i don't like the terminology 2x.57y ...
there is a 25.571, 25.573 was deleted lots of amendments ago.
there is a 23.571 (and 23.572, 573, 574, 575 !?).

i think 571 is a sufficient generic label for DTA requirements.

i agree with your comments about adhesive repairs, they are very good at repairing the damage but the quality of the cure is paramount. we did a patch on a fatigue test aircraft; we had an "expert" come in from HQ ... he misplaced the thermocouple, and baked the wing ! whilst this was a couple decades ago, and i'm sure the technology and the familiarity with it, has improved, it is still a valid cautionary tale.

in my previous post i was going to add the story about the C130. I know it flies with known cracks; th epoint is if the cracked structure can support ultimate load then it is still compliant.

Quando Omni Flunkus Moritati

 

[blakmax]

Errors in repair application technology usually involve contamination control, poor process performance, poor process selection and as you correctly state poor implementation of temperature measurement and control. (It wasn't a C-141 in the early 1990's was it?).

The worst example I have seen is a recent helicopter repair manual which uses one thermocouple and one blanket irrespective of sub-structural heat sinks. There is a very high probability of adhesive undercure or overheat damage of the structure.

Regards

Blakmax

 

[rb1957]

no it wasn't a C141 ... it was a DHC5 ... but it was like your helicopter repair ... a single thermocouple. fortunately it was at the end of the test so it wasn't that much of a problem.

Quando Omni Flunkus Moritati

 

[blakmax]

I know we are diverging from the subject but...

In the case of the C141 it was a deficiency with the controller, combined with bad thermocouple manufacture. The wires on the designated control t/c were only twisted and during the cure cycle adhesive flowed into the junction. An increase in resistance in a t/c results in a low voltage and hence a low temperature is reported and the controller ramps up the power, causing the actual temperature to increase dramatically. The operator noted that the other thermocouples were reporting higher than expected temperatures so he swapped the control t/c to another one by unplugging them both and swapping the connections. It only took a second but the controller was programmed to detect if the control thermocouple was open circuit, and then if so to swap to the next thermocouple, which unfortunately was the one which was malfunctioning and had just been swapped. They turned the system off in a condition where the deficient control thermocouple was reporting temperatures within acceptable limits but it was the only one doing so. The cure cycle temperature was set at 250F but the structure reached over 350F.

Regards

Blakmax

 

[AeroStructAnalyst]

I think EASA policy is quite straight forward with flying with known cracks in that provided that the cracks are proven to be benign and when the primary structure is subjected to loads up to the Ultimate loads to be experienced in-service it does not result in adverse performance of flight. An example of this is with the Airbus A380 Wing ribs cracking, for which Airbus successfully demonstrated that the cracks would have no adverse effect on flight performance and that repair would be effected before the cracks develop to critical stages.