Question about Apple ///

Jim Manley jim.manley at gmail.com
Fri Jul 5 18:28:09 CDT 2019


The first 15,000, or so, Apple ///s had a problem where the very large,
dense, poorly mechanically-supported motherboard would warp as it heated up
after power-on.  That resulted in ICs popping up in sockets enough to break
contact, as well as micro-cracks in printed circuit board traces.  The
warping would occur after about 10 - 15 minutes badly enough to cause the
problems.  Apple quietly replaced those systems when people brought them
in, including mine.

If any of the ICs on your board have popped up enough to cause problems,
reinserting them may not be enough to fix the problem.  High humidity and
fine dust can cause socket contacts to become corroded or dirty, making
operation of the computer flaky.

Micro-cracks occur as the board warps while heating up and stretches the
traces on the expanding side of the board.  They open when expansion is
sufficient and close when the board cools down enough, generally causing
operation to cease and resume after a fairly constant amount of of time
being powered on and powered off, dependent on the ambient temperature.
The warmer it is, the shorter the time, and vice versa when it’s cooler.

One potential solution, if the warping is occurring, is to remove the board
and drill the mounting holes out to slightly enlarge them.  Then, when
remounting the board, don’t fully tighten the screws, allowing the board to
“float” and expand without pushing against the screws and warping.

It’s possible that at least one hole is being used as a connection to the
metal base as a frame ground, which helps reduce RF emissions that might
interfere with very sensitive devices, especially nearby and/or on the same
AC power circuit.  Not tightening associated screws might prevent that
interference-reduction from occurring, but, that’s secondary to solving the
failure to boot.

ICs can be swapped out one at a time with those in a known working board,
if one IC has failed.  However, extreme care in not bending any pins when
reinserting them must be performed.  If more than one has failed, the
combinatorics of the number of swaps needed increase exponentially with the
number of failed components.  Use of an o scope, logic analyzer, etc.,
along with extensive digital signal troubleshooting knowledge (especially
without documentation, including schematics and timing diagrams) may be the
only recourse.

Should you or any of the IM Force be caught or killed, The Secretary will
disavow any knowledge of your actions.  This device will self-destruct in
five seconds.  Good luck, Mr. Phelps.

FFFFFFSHSHSHSHSSSSSSSSKIKIKIKIK!!!


On Fri, Jul 5, 2019 at 12:39 PM Adam Thornton via cctech <
cctech at classiccmp.org> wrote:

> I have an Apple /// that I've had for many years; it's never worked.
>
> When you power it up, you get a checkerboard screen, where half the squares
> are solid white, and the other half have a little mosaic pattern in them.
>
> Looks like this:
> https://share.icloud.com/photos/0NHNkEG9ssPsi65ojivBteKaQ
>
> Does this failure mode ring any bells?  Obviously the video signal is being
> generated well enough to sync a composite output.  Any idea whether I
> should start by replacing the CPU or the ROMs?
>
> Adam
>


More information about the cctalk mailing list