"matt weber"
verizon.net> wrote in message
news:fh4hk351skmhtf54boinkedouj363k3nkj@4ax.com...> On Sat, 24 Nov 2007 20:09:55 +1000, "RT"
nowhere.com.au>
> wrote:
>>>
>>
gmail.com> wrote in message news:326d2a51-8102-4d2f-ab1f-
>>>>> This model of rudder does not include any design features in the
>>> sandwich panels to mechanically arrest the growth of disbond damage or
>>> in-plane core failure before the damaged area reaches critical size
>>> (such a feature was not specifically demanded for certification)."
>>
>>Ouch........ That's a bit scary. I hope such features were required
>>for
>>the Dreamliner.
>
> Dreamliner has strain gauges incorporated into the structure, so it is
> likely you will be able to see the forces that lead to failure, long
> before failure occurs. I.E. de-bonds will cause stress increases in
> some areas, and decrease loads in other areas. Putting these devices
> into the structure is kind of admission that we don't have good,
> simple, easy to use Non Destructive Test methods for parts made of
> these materials, so hopefully the strain gauges will allow you to see
> the loads that lead to failures well before the failures actually
> occur.
I don't agree with your conclusion. A strain gauge on the spot would
negate very expensive (as in difficult/time) access for any other
inspection. An analogy is the huge dragline booms at the coal mines. They
are made of welded tubular steel and then pressurised (nitrogen, I think).
Any drop in pressure is indicative of a crack and the rate in pressure drop
gives a first estimate of the size of the crack. Can you imagine the cost
of inspecting every single tube and cluster to look for possible cracks
otherwise?
Coop and Stealth - you prolly should do that with your Austers. Mmm - on
second thoughts you'd prolly have to carry a nitrogen bottle with you to
keep the tubes pressurised and the whistle of escaping gas would tend to
blot out the sounds of engine misfiring. Better forget it...
It's off charter & I've likely mentioned it before, so sue me:
Did a major job on a big Bucyrus Eyrie boom in the past. They were getting
consistent cracking at the same 3 tube clusters (where a bunch of tubes are
welded together - a joint, ie) and couldn't find out why.
They supplied the engineering drawing for the boom and I used it to do up a
full-sized Finite Element model (FEM). Usually you use symmetry to model
only part of the actual structure - it's quicker. Because I didn't know
what I was looking for I decided to play it safe :-) The boom has different
diameter tubes of different lengths and at different angles with different
wall thicknesses - there was a bloody great table of the dimensions.
Anyway, obviously the boom had been designed using a FEM because the
stresses in the individual tube material throughout the boom was effectively
constant under normal loading - a beautiful design job.
So I tried dynamic loading. Normally you have to deal with the first mode
of vibration, maybe the second and perhaps the third mode in some cases.
None of those affected the clusters I was interested in. In
desperation/curiosity I ran the analysis for higher modes. The 5th mode
affected ONLY the three clusters of interest and was a bending/torsional
vibration of the boom where the boom tip was deflected sideways and at the
same time rotated.
The operators had already specifically been warned not to "sweep the floor"
with the bucket........ :-)
> In metal, you can use Eddy current, and dye tests to spot small
> cracks. In a composite materiall, if the problem isn't on the inside
> or outside surface, it can be very hard to find, and there are
> circumstances where ultasonic inspection won't show it either.
Although I haven't actually used it for such, I'd expect ultrasonic to be
extremely effective in detecting any debonding
