RE: Unmanned test dives

[protek@shreve.net]

Michael,
    I made some comments about window cycle testing the other day.  I did
not test any of my windows that are currently on my sub.  I fabricated them
via a cook-book approach lined out by Jerry Stachiw in his book.  He is an
authority on the subject.  He has tested all the basic designs in their lab
environment.  Why spend a lot of money and time cycle testing windows that
you will have to throw away and remanufacture, when you can use proven
designs.  The book is "Acrylic Plastic Viewports", by Jerry D. Stachiw.
The publisher is Marcel Dekker in New York.  It is about $100, or was when
I bought mine years ago.  It covers all thicknesses and unsupported areas
that you can use in a sub window design.  It even shows you where to put
the seals and how thick etc. etc.  The book was well worth it to me.  There
is one thing in window design that has not been mentioned on here.  If you
machine a metal seat for the window (concrete, or plywood for some) and
then weld it to the conning tower some warping will occur.  You can not
weld a piece of metal without having some warpage.  There is a tolorance
for this explained in the book.  I welded round rings into my conning tower
before I welded the conning tower to the sub body.  I then had the rings
milled before welding to the sub.  This minimized the warpage at the window
locations.  I machined separate seats for all my windows and attached them
to the machined portions of the conning tower with O-ring seals and bolts
(not welding).  This is probably too elaborate, but the windows seats are
difficult to machine as an installed part of the sub.

Gary Boucher



At 01:10 PM 3/1/99 -0800, you wrote:
>To all,
>On saving the hull, I believe someone suggested using a fishing net to
capture the big pieces and a winch which can handle the weight plus the
jerk when it tries to sink (after being neutrally bouyant).
>
>For the viewports I would suggest facing the port into a pipe with a cap
on the end and a machined flange on the other similar to your sub's port
receiving hole.  The pipe can then be flooded and pressurized to the depth
you want to test to.  Pressure cycling and/or long term exposure is then
very trivial, it could even be automatic with a little ingenuity.  In
addition, a dial indicator can be positioned on the port to measure
deflections.  This may help predict failures or show non-elastic
deformations.  A 3000psi scuba cylinder with a regulator can provide the
pressure for an air over water charge line to the pressure vessel good to
any depth up to 6000'.  Using a small orifice can prevent most of the
explosive problems with this setup.  Make sure your pipe and the flanges
have the capacity for the pressure you will be using.  You can even make
this double ended to allow testing of two ports at the same time.  I would
recommend a blast shield of some sort around the whole t!
>!
>hing just to be safe!
>
>> ----------
>> From: 	Michael B Holt[SMTP:michaelbholt@juno.com]
>> Sent: 	Monday, March 01, 1999 1:56 PM
>> To: 	personal_submersibles@psubs.org
>> Subject: 	Re: Unmanned test dives
>> 
>> On Mon, 1 Mar 1999 10:09:20 -0800 (PST) Ray Keefer writes:
>> >> How does someone dive their submersible, unmanned, for testing
>> purposes?  
>> >
>> >Better yet. If the hull crushes how do you get the remains to the 
>> >surface?
>> 
>> How easy is it to find a crane to do the lifting?
>> 
>> Along this same line, perhaps it's time to contemplate a test rig
>> for viewports.   I can visualize a cylinder with viewports on either end.
>> One simply lowers the cylinder into a known depth fo water until 
>> something, uh, changes.
>> 
>> 
>> 
>> Michael B. Holt
>> Oregon Hill, Richmond, Virginia, U.S.A.
>> --
>> 
>> ___________________________________________________________________
>> You don't need to buy Internet access to use free Internet e-mail.
>> Get completely free e-mail from Juno at http://www.juno.com/getjuno.html
>> or call Juno at (800) 654-JUNO [654-5866]
>> 
>
>