[comp.sys.att] Opening the 3B1

wilber@alice.UUCP (12/08/87)

I managed to open my 3B1 one day.  After removing all screws I basically just
grabbed the edges and pulled/teared/stretched/munged until the damn top came
off, ripping off half the skin on my fingers in the process.  The plastic is
molded in such a way that the top latches onto the bottom as though superglued,
even when all screws are removed.  If there's a really slick way to open
the case I'd like to hear about it, too.

Also, I'd like to know if you really can add a 68881 FPU.  (I didn't think
you could, but it looks like I was wrong, since AT&T seems to be selling
68881's for the 3B1.)

Bob Wilber    wilber@research.att.com

alex@umbc3.UMD.EDU (Alex S. Crain) (12/09/87)

In article <7517@alice.UUCP> wilber@alice.UUCP writes:
>I managed to open my 3B1 one day.  After removing all screws I basically just
>grabbed the edges and pulled/teared/stretched/munged until the damn top came
>off, ripping off half the skin on my fingers in the process.  The plastic is
>molded in such a way that the top latches onto the bottom as though superglued
>even when all screws are removed.  If there's a really slick way to open
>the case I'd like to hear about it, too.

The AT&T repairperson who replaced my motherboard did what you describe 
(yank, pull, snort!) until the top came loose with a loud Snap! She said
"Customers hate it when it makes that noise." so I think you have the 
procedure right. I think that the case is molded on AFTER the unit is 
assembled, using the frame as a mold.

>Also, I'd like to know if you really can add a 68881 FPU.  (I didn't think
>you could, but it looks like I was wrong, since AT&T seems to be selling
>68881's for the 3B1.)

Mee Too! There isn't a socket on my mother-board for one, so if it exists, its
a card. Has anyone actually Seen on or gotten a firm price for the thing? I'd
love to do slick graphics on the thing if I could!

						:alex.


alex@umbc3.umd.edu

jhc@mtune.ATT.COM (Jonathan Clark) (12/09/87)

In article <7517@alice.UUCP> wilber@alice.UUCP writes:
: I managed to open my 3B1 one day.  After removing all screws I basically just
: grabbed the edges and pulled/teared/stretched/munged until the damn top came
: off, ripping off half the skin on my fingers in the process.  The plastic is
: molded in such a way that the top latches onto the bottom as though superglued
: even when all screws are removed.  If there's a really slick way to open
: the case I'd like to hear about it, too.

Well, it's not *that* bad. There is a tiny sort of hook-over thing at 3 (?)
places along the bottom of the front, under where the keyboard is
supposed to sit. Not that I can draw, but from the side it would look
something like this:

		  metal	-> || <- plastic
		   pan  -> // <- cover

What we generally do is to undo all the screws, then stick a
screwdriver between the metal bottom pan and the plastic cover, and
lever the plastic out until it pops off the metal. Then move the
screwdriver along and repeat the process as necessary. You don't need
to pop the plastic out very far, and it's fairly bendy.

: Also, I'd like to know if you really can add a 68881 FPU.  (I didn't think
: you could, but it looks like I was wrong, since AT&T seems to be selling
: 68881's for the 3B1.)

No, you can't. This project did exist, but it got cancelled, hence the
misleading mentions in some of the catalogues. The FPU chip and
associated hardware would have sat on an expansion board, taking up
one slot.
-- 
Jonathan Clark
jhc@mtune.att.com, attmail!jonathan

The Englishman never enjoys himself except for some noble purpose.

sbw@naucse.UUCP (Steve Wampler) (12/09/87)

In article <7517@alice.UUCP>, wilber@alice.UUCP writes:
> Also, I'd like to know if you really can add a 68881 FPU.  (I didn't think
> you could, but it looks like I was wrong, since AT&T seems to be selling
> 68881's for the 3B1.)
> 
> Bob Wilber    wilber@research.att.com

I tried to track the "Math CoProcessor" found in the 3B1 parts list about
6 months ago.  After about a dozen phone calls (with 2 cycles) I found
someone who gave me the list price: $2520.00.  Sigh.  If they were correct,
that's simply too much for poor folks like me to put into a system selling
for less than that! I wonder if CT markets an FPU?  I'd love to have one,
but heck, I can do without one at that price.

As an aside, NO one else had heard of the Math Coprocessor, so that price
may be for a FPU for the 3B2 line, and the MC68881 Math Coprocessor in the
parts list might just be vaporware.

Speaking of dreams - I just read that ATT was working on building a
version of its VISTA board for MAC's (rumor?).  Since that's also a
68000-line, I wonder if ATT would consider spinning a few boards off
for their own 3B1's?  I must be sleeping, 'cuz I sure am dreaming.

rosalia@mozart.UUCP (Mark Galassi) (12/11/87)

>>Also, I'd like to know if you really can add a 68881 FPU.
>
>Mee Too! There isn't a socket on my mother-board for one, so if it exists, its
>a card. Has anyone actually Seen on or gotten a firm price for the thing? I'd
>love to do slick graphics on the thing if I could!

    Back in September, when I was trying to get someone to sell me a
3B1 at the fire-sale price, I talked to hundreds of AT&T hot-lin people.
Most of them were twits, and didn't know what on earth a floating
point processor was, but one finally told me the tragic story.  I
sort of quote him (I don't remember exactly, but close).

    "Oh, that was really a tragic story:  we had a board with a 68881
designed and built and it worked beautifully, but then someone in
'upper management' decided that there wasn't a marketplace for it."

    If the manager at AT&T who made that decision reads this newsgroup
(probably not), I would like to tell you that you had a bad idea.

    In any case, even though there *IS* a part number for it (!), AT&T
does not sell it.  Maybe someone who works for AT&T could find a few
thousand of them thrown away in some corner and distribute them to
the rest of us :-).
-- 
						Mark Galassi
					    ...!mozart!rosalia
{ These opinions are mine and should be everybody else's :-) }

krohn@u1100a.UUCP (Eric Krohn) (12/12/87)

In article <629@umbc3.UMD.EDU> alex@umbc3.UMD.EDU (Alex S. Crain) writes:
] In article <7517@alice.UUCP> wilber@alice.UUCP writes:
] >  [ ... description of brute force technique for taking the cover off
] >  of the UNIX PC. ]
] >If there's a really slick way to open
] >the case I'd like to hear about it, too.
] 
] The AT&T repairperson who replaced my motherboard did what you describe 
] (yank, pull, snort!) until the top came loose with a loud Snap! She said
] "Customers hate it when it makes that noise." so I think you have the 
] procedure right. I think that the case is molded on AFTER the unit is 
] assembled, using the frame as a mold.
] 

Try slipping a knife or thin bladed screwdriver between the plastic molding
and the metal case along the front, then twist the knife/screwdriver while
pulling up on the plastic.  The molding has three catches (left, right, and
middle) that you need to loose.  It's really not that hard.

-- 
--
Eric J. Krohn
krohn@ctt.ctt.bellcore.com  or  {allegra,bellcore,ihnp4,seismo}!u1100a!krohn
Bell Communications Research,	444 Hoes Ln,    Piscataway, NJ 08854