23
$\begingroup$

There have been many accidents involving metal fatigue. How do the engineers, maintenance team, etc. know if an aircraft has metal fatigue? What is done to "minimize" metal fatigue or repair it?

$\endgroup$
  • 6
    $\begingroup$ On a side note: I used to work in the field of nuclear safety. We used a piece of very sophisticated and very expensive software for predicting an outcome of a ultrasound test of a piece of piping. A significant portion of that software comes from Boeing (not a software company). There are very interesting and hugely sophisticated methods available and are continuously developed within the industry, that are not visible to us passengers, but contribute a huge deal to the overall safety which we do care about. $\endgroup$ – Pavel Apr 7 '16 at 17:59
  • $\begingroup$ I know that you were asking specifically about metals, but if this was a question of curiosity you should look into new techniques for composites. These are being used in increasingly large percentages of components in aircraft, and need a whole new set of hardware to assess fatigue. I found carbon fibres pretty interesting- give it a look! $\endgroup$ – Jihyun Feb 4 at 7:57
  • $\begingroup$ @Pavel Hmm.. do Boeing have an in-house software team then? $\endgroup$ – Cloud Feb 25 at 14:29
  • $\begingroup$ @Cloud Can't say, really. I remember seeing Boeing in the copyrights and acknowledgements. $\endgroup$ – Pavel Feb 25 at 15:24
24
$\begingroup$

How do the engineers, maintenance team, etc know if an aircraft has metal fatigue?

First of all, is not an "aircraft" that has fatigue, but a component.

To the naked eye, there is no way to detect metal fatigue until it starts being too late: you can only see cracks that are already forming and/or propagating.

There are tools available to allow for early detection, my bachelor's thesis was about one of these: the photoelasticity of certain materials is used. A light coat of one such material is applied on the new component and the photoelastic properties and exams done during the lifetime of the component tell the maintenance people what is the current status of the inner layers of the piece.

Another method uses electric resistance, as metal fatigue, by cracking the metal, increases the resistance.

Depending on the scenario, X-rays could also be used.

What do they do to "minimize" metal fatigue or repair it?

You cannot minimize fatigue, once it starts setting in, it is an irreversible process. You can delay the likely catastrophic failure by applying strengthening elements around the affected one, but the safest course of action is to replace the component and recycle the old one.

$\endgroup$
  • 5
    $\begingroup$ And, of-course, they do stress tests of the separate components giving the manufacturer an indication about when metal fatigue will start to become a problem. They know when to look and where to look based partially on those tests. $\endgroup$ – Mast Apr 7 '16 at 15:57
  • 8
    $\begingroup$ First of all, is not an "aircraft" that has fatigue, but a component. - While this is true it's notable that sometimes the "component" in question is in effect the whole airframe. Consider Aloha Airlines flight 243, where widespread metal fatigue in the skin was a key factor. We can of course view the skin as a component, or even a collection of components (the individual sheets of aluminum), but for practical purposes the whole aircraft is affected as a unit. $\endgroup$ – voretaq7 Apr 7 '16 at 19:39
  • $\begingroup$ @voretaq7 eh, this is where I let my engineering take the best of me. Yes, as you say I would still consider the affected frames or skin panels as components, since, if caught early, you could still think about replacing them. Plus, it drives the idea that you have to observe each component on its own, no fatigue on the skin is no guarantee of no fatigue anywhere else. $\endgroup$ – Federico Apr 7 '16 at 19:46
  • 2
    $\begingroup$ Years ago, I used to work for an aircraft manufacturer who started getting fatigue cracks in components that according to all the data, should not have been occurring that early.this was in a test airframe that they cycled with load for just such unexpected circumstances. After much investigation they found it was due to two components that were bolted together of different materials having slightly different resonating frequencies. At certain engine speeds, the difference in vibration set up a harmonic which added way more load than was expected and caused the fatigue. $\endgroup$ – PabloInNZ Apr 8 '16 at 1:05
  • 1
    $\begingroup$ Also notably, damage tolerance plays an important factor here: Most cracks are not per se catastrophic, and actually act as useful fatigue indicators (that need to be watched carefully!). Depending on crack position/length/propagation speed etc. different actions are prescribed. $\endgroup$ – yankeekilo Apr 8 '16 at 16:13
13
$\begingroup$

Engineers have a variety of ways to detect metal fatigue in aircraft:

  • Visual inspection: sometimes cracks are visible to the naked eye, or can be seen under magnification
  • Auditory inspection: sometimes a thump sounding wrong can give a clue that there's an issue. It's not a primary means of checking for fatigue, but something that may clue in an astute engineer that something needs to be looked into
  • Ultrasound: Ultrasound uses very high pitched sound waves to image materials and can be used to find cracks inside materials
  • Radiology: xrays and other types of radio imaging can be used to find sub-surface cracks
  • Visible dyes: flourescent dyes in penetrating oils show up cracks which make it to the surface
  • Magnetic powders: these only work on iron-based parts which are rare on an aircraft but are used

Here's a good article that goes into some details on non-destructive fatigue testing.

$\endgroup$
  • 2
    $\begingroup$ In addition to that article Chapter 8 of the FAA Aviation maintenance Technician's Handbook (AMT) could well be titled "More than anyone ever wanted to know about inspecting for fatigue/damage"; the good stuff begins around page 8-18 (Nondestructive Inspection/Testing). $\endgroup$ – voretaq7 Apr 7 '16 at 19:42
  • 1
    $\begingroup$ IMO, this is a much better answer than the accepted version. $\endgroup$ – rmayer06 Apr 17 '18 at 15:01
3
$\begingroup$

Some metal fatigue can be spotted visually as an visible crack or as a collection of tiny cracks that can give the metal a 'frosty' look but most is done via non-visual non destructive testing techniques such as radiography (basically X-raying) or Ultrasonic testing to search for microfractures before they reach the problem stage. Visual checks are no use for platform critical damage as by the time you see it it is well into the failed zone which is why NDT techniques are used to detect an impending failure well before it causes a problem.

There are other techniques such as Dye Penetrant tests where a dye in a carrier fluid is painted into the surface and then wiped off. The fluid is light enough to penetrate and 'wick' into any fine cracks and will highlight cracks that are invisible to the eye. The dye is often fluorescent.

As for what can be done it very much depends in the extent and location of the cracking. Extensive damage and/or damage to critical area will be repaired by replacement, minor damage can be gouged out and welded up. The best method of minimising fatigue it to prevent it and so most manufacturers will feed back fatiguing data from surveys to the design sections who will design out the weak point if one exists. Real life data trumps even the most extensive FEA predictions.

$\endgroup$
1
$\begingroup$

The pilot listens and notify the technicians that the plane is giving out strange noises.

And now seriously: For each plane type there are cards which parts need to be replaced or inspected after certain flight hours, start count or time. Together with required quality control at production and maintenance it should prevent metal fatigue incidents as much as possible.

And for the rest - @AndyW already mentioned techniques used during the inspections

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.