Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site asgb.UUCP
Path: utzoo!watmath!clyde!bonnie!akgua!whuxlm!harpo!decvax!genrad!panda!talcott!harvard!seismo!hao!asgb!tomm
From: tomm@asgb.UUCP (Tom Mackey)
Newsgroups: net.columbia
Subject: Re: Columbia's Tiled Damaged by Rain
Message-ID: <759@asgb.UUCP>
Date: Mon, 19-Aug-85 10:07:05 EDT
Article-I.D.: asgb.759
Posted: Mon Aug 19 10:07:05 1985
Date-Received: Sat, 24-Aug-85 15:20:05 EDT
References: <4010@alice.UUCP> <199@mot.UUCP> <521@calmasd.UUCP> <47@darwin.UUCP> <1400@cbosgd.UUCP> <542@calmasd.UUCP>
Reply-To: tomm@asgb.UUCP (Tom Mackey)
Organization: Burroughs Corp. ASG, Boulder Colo.
Lines: 39

Previous postings referring to the tile rain damage note that the tiles are
of a fragile nature.  One author questioned just how fragile they are.  I
worked at Lockheed Missiles and Space Company (LMSC) from 1977 until 1979,
during which time we produced the first ship set of >25000 tiles.  Last night
I came across a scrap tile that managed to "follow me home" towards the end
of my stay at LMSC.  The raw material feels like a very fine grained poly-
urethane foam, with a glass ceramic coating only a few thousands of an inch
thick.  This coating is what produces the smooth aerodynamic surface and
protects the interior from abrasion.  

It is interesting to note that eventhough the raw (meaning uncoated...the
tiles have been "cooked" several times by the time they are ready for the
final coating processes) tile material may be rubbed off like powder with
the finger, they required diamond coated end mills to machine them.  Yes...
each of the tiles are individually machined.  I understand the programming
for the N/C (Numerically Controlled) machines has been greatly automated by
now, but when we first started, MANY of the tiles were individual efforts.
The most that could be expected was to run the program in mirror image so
as to produce right and left hand tiles.  Of course, even that was no help
for the many tiles surrounding portals and sensors.

There was a southern Calif. company that was experimenting with a differant
technique to provide thermal protection.  The tiles (on large areas of the
ship) were produced to be installed in groups of approx. 24, called an AFA
(Array Frame Assembly).  This other company was looking at producing a bonded
titanium/graphite(?) assembly that would replace about 4 of the AFA's at once.
I don't know where that effort went.  It sounded promissing at the time, since
one of the most serious problems with the first several ship sets was too
small a footprint.  In other words, the aerodynamic forces would tend to fly
the tiles off the bonding surface.  A lot of redesign went into reshaping
some of the tiles so the bonding footprint could be made larger.

This posting, unfortunately, has turned into a random core dump, so forgive me
if I got a little off the subject.  I welcome comments and questions, and
would be glad to further contribute within the guidelines of proprietary
constraints.

Tom (Gee, I miss aerospace!) Mackey
....bmcg!asgb!tomm