From novak@clark.phys.nwu.edu Wed Jul 31 16:01 CDT 1996
Return-Path: novak@clark.phys.nwu.edu
Received: from relay.acns.nwu.edu by belmont.astro.nwu.edu. (SMI-8.6/SMI-SVR4)
	id QAA19973; Wed, 31 Jul 1996 16:01:43 -0500
Received: from clark.phys.nwu.edu by relay.acns.nwu.edu with SMTP
	(1.37.109.18/20.4) id AA047977307; Wed, 31 Jul 1996 16:08:27 -0500
Received: by clark.phys.nwu.edu (4.1/SMI-4.1)
	id AA23514; Wed, 31 Jul 96 16:16:06 CDT
Date: Wed, 31 Jul 96 16:16:06 CDT
From: novak@clark.phys.nwu.edu (Giles Novak)
Message-Id: 9607312116.AA23514@clark.phys.nwu.edu
To: jdotson@belmont.astro.nwu.edu
Subject: memo update
Content-Type: text
Content-Length: 3074
Status: RO

Hi Jessie,

Thanks again for the info on the Web page.  It is now much more clear to
me what is the best approach here.

I finally got around to fixing the typo on the necktube strength
calcualtion I did.  Updated memo follows.
_______________________________________________________________
Jessie:

I've done some calculations using standard beam theory to determine the
stresses and strains that the necktube will be under if we
use it as the sole support at the "top" of the dewar.  

(1) The total gravitational deflection that I calculate is 1.0 mils
worst case.  This is the amount by which the top of the
He-4 can sags when the dewar is moved from vertical to horizontal.
It results from the individual sags of the 3 sections of necktube.

(2) The maximim stress in the neck tube is 3,000 psi.  This 
stress occurs at two points along the length of the necktube:
the solder joint to the piece that is at 300K, and, following
the necktube down from this solder joint, at the point where
the necktube meets the OVCS copper thermal link.

Values for the tesile strength of stainless steel in the Goodfellow
catalog range from 65,000 to over 100,000 psi.

(3) The following assumptions were used in the above calculations:

Each section of necktube is 3 inches long.  (3 sections)

The wall of the necktube is .040 inches

.625 inch OD necktube

The insides of the dewar weigh 55 lbs.

The center of mass of the dewar insides is half way between
the two support points.

The 55 lbs. of weight is equally distributed among the OVCS, IVCS and
4 K components.  (this last assumption only affects the
calculation of the deflection, not the maximum stress, which
occurs on the outermost segment of the necktube.)

I used room temperature values for the elasticity and strength of
stainless steel.  Stainless steel (at least 304 stainless steel) gets
slightly stronger and stiffer as you cool it.  It apparently does not
go into a brittle phase like some high carbon steels do.  I got this
info from Barron (Cryogenic Engineering).

I would estimate that my calculations are good to within a factor of
2.  The formulae themselves are very simple and given in
many texts on mechanical design.  I had to think pretty carefully about
which formula was applicable to our case: where a simple cantilevered
beam supports a load at the far end of the beam and that load is also
supported on the opposite end by the Kevlar.  But in the end the
calculation is straightforward.

The maximum stress is inversely proportional to the square of the
necktube OD and inversely proportional to the wall thickness, and
directly proportional to the length of the necktube segment and the
weight.  I assumed that the weight is shared equally between the two
ends of the support structure, when in reality the ratio of the weights
shared by each end is the inverse of the ratio of the distance of the
center of mass from each end.

The conclusion is that we don't need lateral support
of the necktube provided that we don't make major changes
in the dimensions of the necktube. 

Cheers,
Giles