[Date Prev][Date Next]
[Chronological]
[Thread]
[Top]
Re: New ideas
I just finished reading in the book 'Hydrodynamically Loaded
Structures' which describes an interesting hull design. The whole hull
looks like it's accordion folded, with hemi-heads on each end. Turns out
in testing that it's actually comparable in strength to a comparable
cyllinder IF MADE WITHIN TIGHT TOLERANCES. That's an important disclaimer,
remember that different geometries tolerate (or in most cases don't
tolerate) deviations from perfect volumes of revolution in very different
ways. The cross-section of the "cyllindrical" wall looks like this:
/~~\ /~~\
/ /\ \ / /\ \
/ / \ \ / / \ \
\ \/ / \ \
\__/ \ \
I mean, the damned thing looks like it would fold up along the
revolution axis like a piece of collapsible Tupperware! But, all the
finite element models and pressure tests showed that it held up if it's
made exactly right. Bizarre-O.
Now, to read most of your minds and answer your concerns, this is
not necessarily a great idea, what it is supposed to illustrate is that
sometimes math and physics join together to confound common sense in
engineering. However, and this is the REAL point, unless you have amazing
manufacturing skills, your common sense should always over-ride such
counter-intuitive ideas. Often such elegant and beautiful juxtapositions
of opposing forces (such as the accordion fold hull above) are intolerant
of imperfections which real-world structures are going to have.
So unorthodox ideas need to always be approached with caution,
because chances are someone else already had them and didn't do it for a
good reason. That's a long way to come to get that out. Sheesh!
Back to wood, it's been done before rather sucessfully. That means
that it is possible to create a viewport that handles the cycling fine. In
fact, if I remember my reading right, there's a lot of good qualities
attributed to plywood, ease of manufacture not the least among them. Just
be prepared to have to re-create in wood the solutions people with steel
subs had solved for them in the 50's, like viewports and penetrations.
John
John Brownlee
Lunar and Planetary Lab
University of Arizona
jonnie @ lpl . arizona . edu