Re: pressure hull design (Fwd)

[John Brownlee <jonnie@chronic.lpl.arizona.edu>]

	Hi Marsee,

	My point in suggesting that someone who did not know where to
find/derive such a formula probably needs a little more than a formula in
a book to be safe. For example, if you don't know a little about the
mechanics of materials and structures, the Nash book (which incidentally
DOES treat surface irregularities and tolerances, more later) will be
largely useless to you. It is wonderful, but very rigorous with a good
deal of multivariable calculus in the beginning. For those who don't need
this sort of rigorous analysis, don't bother. Desigining in a huge safety
margin is always a good insurance policy.
	The Nash book is VERY rigorous, and more or less shows how to set
up analyses for all the hull geometries you can name. More importantly, it
has an amazing bibliography, pointing to many papers on both the theory
and experimental trials on the buckling of imperfect shells under
hydrostatic pressure. It does not address tolerances per se, but presents
a general way to address geometric imperfections of many kinds. Indeed,
the chapters on spheres are pretty interesting that way; the emphasis is
not solely on the size of the imperfection (the tolerance), but the
-nature- of it. For example, considering the crush depth for a sphere
which has been flattened by a clamp on opposing sides is very different
than for one which has a dimple on one side. The bits on reinforcing rings
for cylinders and cones are also pretty cool. 
	Don't get me wrong, the Nash book is NOT for everyone. But, if you
suffered through senior-level analysis or calculus, it's a chance to
exercise the math muscles and learn a lot about the mechanics of pressure
hulls. It is largely geared towards someone with a strong engineering, or
at least mathematical, background. If you never took much calculus, or
don't want to get your 'hands dirty' in the hull design process, don't
bother. Hire a licensed engineer to do the analysis for you, etc. Also,
much of his emphasis on high pressure regimes (like hundreds of PSI) is
serious overkill for the 100' boat. 
	With regard to whether or not forged vessels intended for holding
compressed gasses and the like are good enough, I must confess I don't
know exactly. There's a lot of unknowns there, but I suppose with the
proper bent you could find out. I'd just rather build the thing from
square one using known steels, etc. myself, but that's how I've operated
all along. I am sure that there are several COTS products which work fine,
but somehow the ol' propane tank leaves me a little on edge, as I'd like
to dive deeper than 200' in the best of all worlds. 
	I should qualify all this by saying that I have probably in one
form or another designed and built everything BUT the hull for my own
boat. Being a computer/optics wonk I started with what I knew and went
back to school for what I didn't. Started with life support, of all
things, and built from there. Of course, working at a university helps;
when I knew I needed to know more about statics, I could sit in
on a class over at the engineering college and pick it up. 
	This raises an interesting question: in designing, you obviously
start with the hull (well, you -START- with a spec, but that's different
than what I mean) but in BUILDING, does everyone start with the hull and
move inwards? Has anyone done it 'backwards,' like I have? Moreover, does
anyone test their components hydrostatically during construction?

							John

John Brownlee
Lunar and Planetary Lab
University of Arizona
jonnie @ lpl . arizona . edu