Model Forum / General / Models / July 2008

Actually, I was wondering the same thing. Is it really that overbuilt?
This plane was designed completely on computers, no wind tunnel stuff
and the like first. The first plane of this magnitude to be done this
way. I'll bet that by the time my friends told me about it the facts
had been filtered through so many people that the truth is somewhere
unknown. I do know that the wings were flexed was past what aluminum
wings can take and they did not fail. And that they did exceed design
strength. At the same time they are apparently stiffer than aluminum
wings.
ERS

....except that, if it's 'new technology', they'll perhaps be anxious
to keep it a bit on the over-engineered side until it's been proved in
service. They wouldn't want a Comet-style series of failures.
I thought I had read somewhere, a few months ago, that the project was
a bit behind schedule because they were having to sort out issues with
the attachment of wings to fuselage.

Tim

Tony Jeffree Wrote:

It's a little more complicated than that.

Both the Boeing and the Airbus wings have been tested far beyond th
maximum stress expected on the wings, more or less the point wher
either the wing, the fuselage or the passengers would have failed.

If I remember correctly the safety factor is Maximum *1.5. The Airbu
designs  failed at around the critical point (actually slightly below i
but within 3%)  The Airbus wing flexed 7.4 metres before rupturing.
http://www.cs.york.ac.uk/hise/safety-critical-archive/2006/0031.html

I remember reading somewhere (High Performance Composites?) that
Boeing design had had a similar failure.

The design problem is enormous, whereas one can look as a piece o
aluminium as a single piece, the equivalent composite part may b
mutiple layer of a material in a matrix . Each layer may be laid at
different angle and where the material is woven, there is a stress poin
whereever one thread crosses another.

So instead of the fairly precise data that is available to the meta
engineer, one has to work with approximations of behaviour.

Another problem is that effectively you 'create' the material ever
time you make a component part.  Imagine if you had to mix, melt, rol
and treat and test every piece of aluminium instead of relying on th
manufacturer to do so. Indeed the Americans have the AGATE databas
which defines how to put the various materials and resins together t
make a standard material and the various figures that one can use fo
that material to avoid some of the problems.

So 'overengineered' is a relative term, given the large variable
possible and the many modes of failure that have been identified

--
rss

Cheaper here...

http://www.frost.co.uk/productList.asp?catID=29&frostCat=Fillers

John