Equivalent Strength Repair / Analysis 2

[ssouthworth]

Does anyone have any documented information on equivalent strength repairs / analysis? I’m currently working an issue in which an engineer has stated that equivalent durability (I take this to mean fatigue durability) is achieved through equivalent strength repair design / analysis. I have not been able to find any books, research paper, conference paper, or anything else to support this claim. During this search I became curious about the origins of the equivalent strength methods. I haven’t been able to find anything really documented on the method either (I have discussed this with several DERs). Can anyone point me towards some solid references?

Scott S

 

[SWComposites]

What type of structure? material?

Most equivalent strength/durability "analyses" are based on wishful thinking and fast arm waving.

 

[rb1957]

at a high level, equivalent strength reapairs should return the structure to it's previous performance.

However, there are a couple of things that aren't, that will be new to the repaired sructure.

1) is it a Part 25 a/c (ie Damage Tolerant) or a Part 23 a/c (more fatigue based) ?

2) you probably added an external doubler. this probably affects the inspection of the underlying structure (eg if the pressure skin is inspected externally visually, then the doubler means the the skin under the dblr must be inspected by NDI).

3) you probably attached the dblr to the structure with rivets. These localise the load transfer into the dblr and create fatigue hot-spots. These are usually the critical fatigue site in a repair.

for references, maybe AC43-13 (though i really dislike their DT discussion) or Boeing Liaison stress Analysis (U.Kansas)

 

[ssouthworth]

The original material was 7050-T7451. The repair material is 7075-T7651 due to availability. The member being repaired is an I-beam shaped member acting as a beam-column. Repair is essentially a double-shear splice with flat strap on exterior of the flange and angles on both sides beneath the flange fastened back to both the flange and the pad-up area of the web. This is a military aircraft so FAA/FARs do not have to govern although ACs can be used as acceptable tech data. I’m looking for any references that would support this engineer’s statement. I’ve search far and wide and can find nothing.

Scott S

 

[rb1957]

ok, i'd express the question as compare the fatigue/durability properties of 7050T74 and 7075T76.

but first i'd ask ... if this is a beam colmun, emphasis on column, why would i be concerned about fatigue ?

another consideration would be this is now potentially multi-loadpath.

 

[ssouthworth]

The aircraft is a CH-47 helicopter. This beam-column directly supports the main transmission and rotor system. The beam is a machined frame. The concern then comes that if cracking initiates there are no natural crack stopping edges like that of a built-up / composite frame. (Some refer to them as re-initiating sites as the crack has to jump from the web to the flange or vice versa)

The assumption is that the rotor system acts as forcing function on the frame (high cycle fatigue). I’m not arguing this point.

The only point I’m interested in at the moment is do we the engineering community have any scientific data to support a statement that an equivalent strength repair, based on ultimate tension, directly equals a fatigue durable repair? It seems like a rather bold statement that so far cannot be supported based on the lack of data.

 

[rb1957]

of course it doesn't.

it puts back the same (or equivalent) area, so the same of equivalent stresses should occurr from the same external loads. so far so good.

but the loadpaths have changed. you have introduced fatigue hot-spots at the fasteners. in the type design (unrepaired structure) you had a beam, happily carrying this stress. now in the repaired structure the beam has to transfer the load into the repair (and out again). IMHO you need to look at this. it could well be that the fasteners have interference fits, and so beneficial compression stresses, to off-set the fatigue effects of load transfer. This is actually a bigger problem in the parent beam ... the repair splices could well be over-sized ('cause its a repair) but the parent beam is now doing this secondary load transfer job as well as it's day-time job of carrying the primary loads. This is one reason why you need to be very careful designing repairs (not to make them too stiff). Working in the parent beam's favour is that this laod transfer is happening away from the critical stress region of the beam (you'd hope) ... again off-setting minors (reduction in working stress vs increased fatigue due to load transfer) but something that shouldn't be hand waved (particularly in the structure you're talking about).

mind you, the repair would probably have very limited multi-loadpath capability ... consider how the load will redistribute if part of the repair is gone ... unless the repair has been carefully designed so that any two elements (for example) can carry limit load.

of course, your designer could well say "well that's the way we've designed these repairs for years and nothing's happened", in which case "i come from Missouri, show me".

we used to make similar statements regarding FAR25 aircraft mods ... "standard of type" before we had a full time DT guy ... luckily we fixed these before Part26 came around.

 

[ssouthworth]

Well that's interesting. This other engineer is also a DER. He is pushing this notion, and has even written it down for this repair, that equivalent strength (unltimate tension case)equals equivalent durability. And some people are swallowing it hook, line, and sinker.

I on the other hand come from your state (of mind) ... Show Me. So far he's all talk and NO show.

 

[rb1957]

what s/n should i stay away from ?

he may be a DER but what're his credentials from fatigue analysis ? does the repair look well designed for fatigue ... several steps (rather than large changes in thickness) ? off-setting dblrs (so they don't all start at the same location), interference fit fasteners (Hi-Lites ?).

In my opinion your two materials are reasonably equivalent in fatigue, and static props too. Your repair could very well be adequate, but for such an important piece of structure i'd like more than someone's say-so.

 

[Sparweb]

If it helps (I'm not sure it will), you can go all "chapter and verse" on this DER. Well it may help you be informed at any rate. Anyway, you could point out that just because it's a military (or ex-military I guess) heli then that doesn't mean that civilian safety regulations aren't at least a reasonable starting point for evaluating structural repairs, even if they don't strictly apply. You can look up FAR 29 for the airworthiness regulations for transport category rotorcraft yourself, and then start citing these ones:

FAR 29.305, Strength and Deformation
FAR 29.307, Proof of Structure
FAR 29.549, Fuselage and Rotor Pylon Structures
FAR 29.562, Fatigue Evaluation of Flight Structure

The FAA publishes all of the FAR regulations on their website. That 307 can seem very daunting, and it should. It takes a lot of facts to show PROOF. The burden of providing that proof is on the DER.

Anyway, I haven't really answered your question, yet. RB1957 has raised enough questions, I can't add any better ones. The "I said so" stuff just set me off. Primary structure repairs don't get by without substantiation, military or civvy.


STF

 

[WKTaylor]

Run Forrest Run...

Hmmmm... this sounds VERY much like a curved frame beam we had to replace [NOT repair] at depot ~1990. Jacked/stabilized the SH* and did so for the rotor systems also. Removed the beam, reinstalled it, removed-it and then reinstalled it [with original fasteners] a few times to convince us we had the area in ~NO-Load situation/NO-drift situation [also, special hangar, limited access only as needed, etc to avoid disturbing acft alignment.].... then immediately inserted/match-drilled/fastened a new frame into-position. Everyone was happy and I did NOT have-sweat the frame repair.

Now, RE Your concerns...

I have several Liaison/field engineering documents, training courses and handbooks that discuss fatigue critical repairs. Also had a ton of field experience, with many acft types, same problem.

This notion of equivalent strength is only for starters. IE: developing the rough repair concept. Repair part thicknesses, min fasteners, general shape of parts, etc.

The next phase ensures that stress concentrations due to enlarged or new holes and the gradual or abrupt step-out of repair materials [thickness, fastening, etc] are accounted for. This is where the repair parts are tapered/staggered-out beyond the repair areas to minimized end fastener strain/stress: and allow visual inspection of these critical fasteners/surfaces in the coming years. OH YES recurring NDI will be required, so make it easy on the troops: do NOT bury critical areas such that You are forced to remove fasteners: stagger repair parts on the existing beam to allow access from at least one side for inspections.

CAUTION. Watch-out for hidden asymmetries, also: these can generate unanticipated/disastrous loads [obvious asymmetry can be designed/analyzed for].

NOW comes the devil-in-the-details.

WARNING. Best of all world is a well-designed repair executed with high quality workmanship. A poorly designed repair executed with good workmanship is usually as good-as or better-than a well-designed repair executed with poor quality workmanship. A poorly designed [thought-out] repair executed with poor workmanship is my worst nightmare.

CAUTION. Analysis/consideration is needed for the possible thermal and sonic stresses in the area of repair: SH*s scream and shake a lot.

CAUTION. If one side is cracked/broken, then the opposite side may be in process of cracking failing… just not evident. Suggest You consider examining the opp side beam-frame also.

CAUTION. repair stiffness can alter structural vibration modes… especially when asymmetry is introduced.
Details…

NDI and inspection for damage in structure that will be covered by [over-laid-by] the repair parts; and restoration of damaged finishes [etc...]. This includes fit/form/function changes that may affect other components [mechanical, structural].

Ensure base part and repair part have adequate mechanical allowables; and that other critical factors such as fracture toughness, crack propagation rates, SCC and EXCO margins are significant before beginning [IE: MMPDS stuff].

Base-part damage removal methods : ensure damage is carefully cut/trimmed/blended/ stop-drilled etc]. A picture of damage removal BEFORE repair parts are installed is helpful to the repair techs.

Repair part configuration and fit to the base-part has to be carefully detailed. Any omitted details will probably be “guessed” by the repair techs.
Fastener selection is obvious… but difficult: tension and/or shear, male [pin] and female [collar/nut] types, nominal or oversized diameters [nom = new holes, OS mandatory for old-holes matching to new parts], grip-lengths, hole sizes [clearance, transition, interference, etc].

Surface finishes all parts: shot peening, low RA machined finishes, corrosion protective finishes, sealant. Also, IF surface roughness, waviness or gapping between parts is significant, then epoxy metal filled "liquid shims" [“Aluminum putty” or harder still “Titanium Putty”, etc] in these areas might be advisable [minimize effects of poor-fit-up of parts... especially important in high sonic areas].

The repair sequence [for match-drilling/reaming, deburring, fastener installation, etc] is also very important.

Etc.. Etc…

Two good references for damage assessment and fatigue critical repairs [other than proprietary data] are:

FAA-AIR-90-01 REPAIRS TO DAMAGE TOLERANT AIRCRAFT [BY TOM SWIFT]

http://www.dtic.mil/dtic/

Failure Analysis and Prevention for the Air Logistics Center Engineer: CAStLE Course Development Summary

http://www.dtic.mil/dtic/

------
N’uff for now… gotta get back to work. Hope this makes sense.

SH* = Sh*t-Hook = CH-47 Chinook … but don’t tell Boeing…




Regards, Wil Taylor

Trust - But Verify!

We believe to be true what we prefer to be true.

For those who believe, no proof is required; for those who cannot believe, no proof is possible.