doubler damage tolerance analysis 2

[PierpaoloF]

I will be involved in the near future in the damage tolerance analysis applied to skin doublers for repairs and antenna installations, but I am not aware of specific techniques for this kind of applications.

Is there any common methodology, or specific basic principles, to tackle this problem and where can I find references, preferably available online, concerning this issue?

Thanks in advance for the help.

 

[rb1957]

g'day,

be careful about following AC43-13 (from FAA website) ... it's not wrong, just old (much like most of us).

i'll assume you're dealing with a pressurised fuselage (rather than an un-pressurised fuselage or a wing). I think your critical design requirements are fatigue loading (meet this and pretty much by default you'll meet the static requirements). Michael Niu's books are good technical sources.

First, a couple of design guidelines. doubler thickness about the pressure skin gauge. general practice is one gauge higher. allow two rows of load transfer rivets, as closely spaced as the stringer rivets (or spaced as general design practice would have them, 6D). I prefer to terminate the doubler at the adjacent stringers (as these support the out-of-plane bending due to the off-set load going into the doubler) some people would put the first row of fasteners on the other side of the adjacent stringers, some people would put them inside. In addition to the load transfer rivets, field rivets on a wide spacing, where they'll fit. Take so care with the fore/aft ends of the doubler. usually i'll go to the adjacent frames, again using the frame flange to support the secondary bending, but as the loads are about 1/2 this is less of a problem.

And so to the fatigue load. start with cabin delta p, and factor as required (!?). you're factoring to allow for longitudinal loading (that you're not explicitly accounting for) and in the rear fuselage (particularly near the wing Rear Spar at the roof) for gust/manoeuvre cycles. i'm guessing that you don't have this OEM information, but your local cert. people should appreciate some allowance for this effect; i'd suggest that a facotr on delta p of 1.5 is plenty conservative. ie a GAG cycle of 1.5 delta P hoop stress should be plenty conservative.

Then you've got the crack geometry to worry about. basically you have a crack at a hole, loaded by tension stress and pinloading. these geometry solutions are widely available (david broek's book, rooke and cratwright, JC newman (with NASA).

i'd suggest assuming a detectable crack of 1 pitch; this is easily detectable (but you will need to specify an NDI technique like low frequency eddy current to detect a crack growing in the airplane pressure skin under the doubler). so an initial crack length of 1 pitch + 1 diameter + 0.01", initiating little starter cracs at opposite sides of the detectable crack. grow this crack in steps, i use a spreadsheet, you can use NASGRO or FLAGRO (free !) for another pitch. i think you'll find a 3 pitch crack grows pretty quickly, so you don't get much return on continuing the analysis. simply make the threshold = repeat inspection interval, if this is not too penalising to the operator.

your repeat inspection interval is the crack life from detectable divided by factors, at least 2, 3 is reasonably conservative, 4 is very conservative. depends abit on where the doubler is, the belly of the plane tends to be more prone to corrosion, and so gets a higher factor.

if you need to determine the threshold, now you've got a 0.05" flaw at a hole loaded in tension and pin-loading. this is quite a bit more difficult to deal with, but the geometry solutions (and FLAGRO) are available.

if you teminate your crack growth at three pitches you won't have a problem with residual strength, but it's something you should check, assume limit stress is a high proportion of yield (100% is very conservative, 90% is probably conservative, 80% less so ...).

damage tolerance in a can,
good luck !

start with the hoop stress.

 

[WKTaylor]

PierpaoloF..

Go-to FAA circular website [below] and enter "fatigue" in search box. Goo start-point.

http://www.airweb.faa.gov/Regulator...rgAdvisoryCircular.nsf/MainFrame?OpenFrameSet

Regards, Wil Taylor

 

[PierpaoloF]

Thanks to both for the info, it's a very good starting point. The specific problem I have is that we often use doublers for blade antenna installation, so I have to consider the large overturning moment caused by aerodynamic forces. It's not a big deal to calculate the maximum stress for the static analysis, but which is the easiest way to take into consideration these loads for fatigue and crack growth? Consider that to keep the job economically profitable for my company (small DO), I can't spend more than few hours dealing with the whole stress analysis.

 

[rb1957]

g'day,

we do that alot too. most often, usless the blade is very big or the plane very fast, aero. loads aren't that significant, even with a CL of 1. in any case, we consider the aero. loads as static, and usually the driving static load is ground handling, say 200 lbs in any direction (usually across the blade) applied at the tip. we find that this usually requires some sort of channel under the antenna, distributing the loads to the adjacent frames. it'll be interesting to see the comments on that !

the fatigue loads are primarily the loads in the fuselage, mostly pressure, that are affected by the doubler load transfer.

to do the job with the least amount of work, use typical design rules (dblr thk one gauge higher than the skin, LZ rivets (you don't want to knife-edge the dblr, i'm guessing skin < .08"), plenty of rivets, at least two rows for load transfer) then you'll need an LFEC inspection (to inspect the skin under the dblr, hopefully you can apply an existing OEM procedure for similar geometry) then you'll need to plead with your cert. people to accept an arbitary repeat inspection (1 year?) ... 'cause if you have to do DTA without experience you'll blow the bank !! if the cert people won't buy the arbitary (conservative) inspection, then try to get an analysis shop to help out ... if you've used good design practice you won't get surprised.

good luck

 

[crackman]

You might want to check out thread2-94480 in the aircraft engineering forum. We have discussed this topic to some extent in the past.
Good luck.

 

[Sparweb]

An FAA engineer named Bob Eastin wrote a presentation of guidelines that he expects used in these kinds of installations. It's a simplified method of finding fuselage bending stresses and combining them with the pressurization stress.

http://sparweb.squarespace.com/storage/articles/FAA-AIR-90-01.PDF

http://sparweb.squarespace.com/storage/articles/Eastin_DTA_Slides.ppt

Steven Fahey, CET

 

[crackman]

The approach in the presentation for addressing fuselage bending effects is certainly a valid one but very conservative and I believe was only presented as a simple approach. It does obviously meet the FAA requirements but generally results in very very short inspection intervals which is not very good for the operators.

As always, you get what you pay for. If a conservative, short and inexpensive FAA compliant DTA is required, then certainly one could use the approach. However, the operator ends up paying in the end when he ends up doing a LFEC every 1500 hours on a simple antenna installation. So its really "Pay me now, or pay me later". This approach may be acceptable for GA but on an airliner, this would be tough indeed on the operators where aircraft accumulate 2000 to 3000 hours a year. The typical goal for an inspection interval for a commercial airliner should be at the very least a C check. Not sure the method presented can ever achieve this.

Just for example, consider a fuselage barrel with 2024 skin, 8.6 psi internal pressure, 74 inch fuselage radius and 0.05 inch thick skin. Per the presentation, here is the resulting 1G stress = 14.8 ksi. Now, assume a typical delta Nz of 0.25(very typical for a once per flight basis) and add in pressure. The total stress is now 24.85 ksi! How many cycles does one think this will really last at 24.85 ksi per flight? Particularly when a normal DSG is around at least 30,000 cycles.

The best approach that I have found is also the tougher one but the proper one. If one is to embark on performing DTA's, they should also be experienced in developing fatigue loads as well as performing crack growth and this should be expected of them. One without the other results in a lobsided approach. In this vain, I have seen many good engineers go to great lengths to select their crack growth methodology, the specific material data used, speciliazed stress intensity solutions to only end up making a globalized general assumption about the most important of all effects --- stress ---. If an antenna/repair/mod is placed in an area that is affected by fuselage bending, then a fuselage bending spectrum must be developed. This is not difficult if the engineer is an experienced DT analyst. The FAA has published many load history exceedance tables for taxi, landing, gust, maneuver, etc. for numerous commercial aircraft. In turn, any well experienced DTA engineer should be able to develop balanced aircraft fatigue loads and correlate them. With these two, a fairly representative fatigue spectrum can be developed, it only requires experience and time. No quick boot-strap solutions here. DTA is not something that is learned over nite and requires good training and mentoring from the old gray beards. As a quick aside and obviously in my own opinion, here are the key areas which an experienced DTA engineer should be experienced in:

Static Strength Analysis (strong understanding of primary structure)
Fatigue Loads Development(internal and external)
Spectrum Development (mission profiles, block spectra,flightbyflight spectra,randomization,etc.)
Stress Life and Strain Life Methods
Material Characterization (dadN,toughness,Rcurves,etc.)
Stress Concentrations and Stress Intensity Solutions
Retardation Effects
Crack Growth Rate Methodologies
Inspection Methods
Probabilities of Detection and Detectable Flaws
NDT Procedures

Ok, I am now officially stepping down from my soapbox.
Best of luck.

 

[PierpaoloF]

Crackman,

the reason why I was asking is that I have a good academic knowledge of most of the fields you have listed, but no experience on this kind of applications, and I had the feeling that a DT analysis properly done was going to be misproportioned compared to the size of the whole job, as would probably double or treble the time needed to complete it, and consequently the final cost for our customers.

Is DT analysis required only if the skin is listed as a Structural Significant Item in the area of the fuselage where the antenna is installed?

 

[rb1957]

PierpaoloF

maybe it's an "english usage" issue but i think you're incorrect in saying "misproportioned" and "2x or 3x the time needed". i think you are referring to your company's budgetry estimates for the job. if you are inexperienced in DTA it will take longer (as i mentioned above), but then how can you estimate something you're not experienced in. i guess its probably correct to say yyour surprised by the extent of this work !

in any case, to answer your second question ... YES, in fact because the structure is an SSI strengthens the case for DTA. There are design solutions that minimise the impact (if you can maintain the OEM inspection program (intervals and technique) then you won't have a AirWorhiness Limitation (mandatory inspection) to add to the plane.

Most likely you won't be able to do this, instead either ...
a) if your adding an external doubler, you'll probably require a change in inspection techique (to NDI) but the intervals should improve over the existing inspection; or
b) if you use an internal doubler, which is trickier to do, you probably don't have the information to fine fune the DTA to keep the inspection intervals (and you'll probably have to add an inspection for the dblr.

it's motherhood i know, but designing in primary strucutre is tricky. Just for the DT aspects alone you need the existing inspection program, understand how your design is impacting this, analyze this, then determine the AWLs and create the inspection task cards and MMSs. The DTA is probably the shortest part of this, tho' it requires the most knowledge.

one thing that might be in your favour ... the term SSI is used but only a few companies. if you're working on a Dash8 if you can locate your feedthru hole (<1" dia) is a double thickness (skin+waffle = 0.06") you can probably avoid the external doubler (as the pressure stress is pretty low in these bays).

 

[crackman]

Pierpaolo

Per the FARs, any aircraft whose Type Certificate Data Sheet (these are available in pdf format on the FAA website) lists FAR 25.571 amendment level 45 or later as the cert basis as well as any aircraft having an AD mandated SSID which uses AC91-56 as its basis (search AD listing) and any aircraft listed under FAR 121.370, must have repairs which meet damage tolerance requirements.

As far as the time required to do a DTA, it is wholly dependant on the complexity of the repair/mod and the experience of the DT analyst and the tools and methods he/she has available. If he/she has smartly developed analytical methods and procedures which include spectrum generation (this takes some time to set up and obviously the experience to do so), a "small" antenna installation (within 1 bay, no external dblr - uses internal dblr, low height ie 4 inches or less) should not take more than a day. However, I caviat this with the fact that it must have proper load path, good load transfer - ie proper fastener sizing and spacing, use good fatigue resistant materials, etc. otherwise it can take much longer. This is usually why its best to be both the stress and fatigue engineer evaluating the design.

So in other words, it is always best to sit down and develop all of your methods before attempting to do DTA on a repair/mod otherwise you could end up taking much longer and thereby impacting the customer. Unfortunately, there is not quick way to learn it or brain dump available, just takes time and good mentoring from older more experienced engineers - a luxury in today's industry which we all realize I am sure. Anyways, a properly accomplished DTA can provide the necessary continued airworthiness instructions as well as be economical to the customer - it just depends on how experienced and prepared the DT analyst is to get the job done.

Hope this helps and good luck.