NOTE.
Just to complicate the picture.
RE: ANSI H35.2 Dimensional Tolerances for Aluminum Mill Products
Perfect flatness of raw material [sheet, plate, etc] is a fairytale. Per H35.2, aluminum sheet is 'allowed' an as-manufactured out-of-flatness tolerance that is substantial for large flat skins, webs, etc... but perfectly adequate for small flat, flat-formed [bent or hydroformed] parts such as frames, ribs, intercostals, formed-spar channels, etc. Just look at the average wide/long sheet of clad aluminum from a shallow angle, and this phenomena is easy to observe!
Just about ever GA aircraft out there has waviness induced by the natural-out-of-flat condition of the raw sheet aluminum stock. Perhaps these tolerances can be reduced a bit for 'premium sheet-stock' with extra flatness requirements imposed [$$]; however this is not the rule from most raw sheet/plate aluminum stock. Same goes for any 'hard' alloy made from titanium, CRES, Inconel, etc.
When assembled to rigid structure, there is a tendency for large skins/webs to form shallow waves or pillows between rigid members, such as stiffeners, frames, fittings, etc. Often adjacent open bays may have opposite tendencies: one bay pillowed-out, adjacent bay pillowed in... and natural wrinkles and large pillows may actually pass-thru these rigid edge attachments. Most GA Acft, when taxied at low speed and low power [reduced noise levels] can produce some awful popping/snapping sounds... clearly indicating that the skin/web is in a state of dynamic flex in these 'oil-canned bays'.
NOTE. The technical term for 'oil-canning' is [gasp!!!] 'snap-thru elastic buckling', which describes the nature of the beast: the member snaps from one-state [EX bowed-out] to the opposite-state [bowed-in] elastically, and almost instantly, with very little force [transverse to the sheet plane]. The problem with oil-canning is [generally speaking], fatigue cracking at periphery fastener holes due to the large number of oil canning cycles per flight [which are not usually heard]. Even though flight loading can mitigate the problem for awhile, still experiences various stress states in non-static-flight conditions; and eventually the structure still has-to return to it's normal static state when tied-up on the flight line. NOTE: the poor riveting practices noted in a previous post [excessive localized swelling due to rivet installation pressure] is a great aggravating factor for this phenomena!
NOTE.
I have encountered a phenomena where localized overstress due to poor fastener-fit [and also poor riveting practices - another story], mechanical or flight loads has induce permanent distortions resembling natural oil-canned sheet, that make it 'easier' to 'oil-can buckle' with less/less energy...' at lower and lower outside loading. In worse case instances, fasteners in holes can also loosen/smoke/crack... and worst possible case is where flutter [natural vibration] frequencies can change... usually for the worst [IE flutter at progressively lower airspeeds, etc].
NOTE. The aircraft I work on was designed with secondary wing rib webs allowed to diagonal-tension-wrinkle at/above limit load: rivet-hole cracking is a major problem. I suspect that, even though the thin web sheet used was perfectly adequate by analysis/test, it had enough inherent natural flatness tolerance issues to be a 'set-up' for oil-canning 'as-riveted into position'; hence premature failure.
Regards, Wil Taylor
o Trust - But Verify!
o We believe to be true what we prefer to be true.
o For those who believe, no proof is required; for those who cannot believe, no proof is possible.
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion"]
o Learn the rules like a pro, so you can break them like an artist. [Picasso]