23 Jul
2016
23 Jul
'16
1:55 p.m.
From: Corey Cohen
It was not someone at the PCB manufacturer. They would
not have had
access to the prom software.
So, do you have a theory about where this came from? (There is absolutely zero
snark here, this is a serious question. It's quite a puzzle, and an
interesting one.)
Maybe a collaboration between two people, one at Apple, one at the PCB house?
'Make two extra boards, and I'll trade you the PROMs for one of them.'
Can't
do it with just a person at the PCB house - as you point out, need the ROMs.
But you'd think that if someone at Apple just pulled a board, that would be
noticed (board count wrong).
Noel
23 Jul
23 Jul
2:37 p.m.
My guess is that is was a test board for Apple. There are some weird mods to the ram
timing with a variable cap and to the negative supply that looks like they were
experiments to figure out the tolerances of the chips. The board was wave soldered. You
can't fake that on an Apple-1 because of what happens to the back of the board by the
regulators. I had conversations about this board with Woz and Daniel Kottke who along
with Steve Jobs were the only ones who could have had access to the roms and would have
known what the board was. The PCB house workers wouldn't have cared or known what to
do with it. This is before anyone even knew the name Apple.
Other than a single replacement IC. All the chips and soldered components are correct for
something put together before the byte shop order just different parts than the rest of
the boards. All the pre NTI boards are very consistent in parts just the edge connectors
were installed backwards on some. The NTI varied on the smaller electrolytic caps.
There is only one other known but lost to time Apple-1 with the same decoupling caps (they
are different than the NTI, though similar). That board was the preproduction board used
for the Apple-1 printed Advertisement. That board also used the same RAM chips, which
Apple did not use in the end when they shipped the Apple-1, they used the cheaper plastic
ones.
So lots of evidence this was not something where someone grabbed a PCB for an unknown
computer company risking their job and built their own.
Cheers,
Corey
corey cohen
u??o? ???o?
On Jul 23, 2016, at 8:55 AM, Noel Chiappa <jnc at mercury.lcs.mit.edu> wrote:
From: Corey
Cohen
It was not someone at the PCB manufacturer. They
would not have had
access to the prom software.
So, do you have a theory about where this came from? (There is absolutely zero
snark here, this is a serious question. It's quite a puzzle, and an
interesting one.)
Maybe a collaboration between two people, one at Apple, one at the PCB house?
'Make two extra boards, and I'll trade you the PROMs for one of them.'
Can't
do it with just a person at the PCB house - as you point out, need the ROMs.
But you'd think that if someone at Apple just pulled a board, that would be
noticed (board count wrong).
Noel 
3:20 p.m.
On Sat, 23 Jul 2016, Corey Cohen wrote:
My guess is that is was a test board for Apple. There
are some weird
mods to the ram timing with a variable cap and to the negative supply
that looks like they were experiments to figure out the tolerances of
the chips. The board was wave soldered. You can't fake that on an
Apple-1 because of what happens to the back of the board by the
regulators. [...]
If you mean the crinkle tin plate under the solder mask, that doesn't
happen due to wave soldering. The heavy tin plate was applied that way in
a separate process before the solder mask was applied to the board. It
used to be common to do that to all sorts of boards in the 1970s-1980s.
With modern boards, is much more common now to just leave exposed
copper/gaps in the solder mask and allow those areas to take up solder
from the wave soldering (or reflow) processes.
4:30 p.m.
On Jul 23, 2016, at 10:20 AM, Tothwolf <tothwolf at
concentric.net> wrote:
The crinkle is exaggerated by the heat from the wave soldering. Your right doesn't
happen in modern boards because of how the tin was applied.
Mike Newton tried to reproduce the technique in China for his replicas. After many
failed/peeling attempts he came close but even when wave soldered, it doesn't match
the 1970's effect exactly, though it does crinkle.
The other thing with wave soldered boards is how the Vias get filled in. They sorta
pucker, it's hard to explain but easy to show. You can fill the Vias by hand, but
they don't look the same as a wave soldered board. In experiments I have tried a lot
of different ways to replicate it. You can always tell the difference if you look under a
loupe.
Cheers,
Corey
corey cohen
u??o? ???o?
On Sat, 23 Jul 2016, Corey Cohen wrote:
My guess is that is was a test board for Apple. There are some weird mods to the ram
timing with a variable cap and to the negative supply that looks like they were
experiments to figure out the tolerances of the chips. The board was wave soldered. You
can't fake that on an Apple-1 because of what happens to the back of the board by the
regulators. [...]
If you mean the crinkle tin plate under the solder mask, that doesn't happen due to
wave soldering. The heavy tin plate was applied that way in a separate process before the
solder mask was applied to the board. It used to be common to do that to all sorts of
boards in the 1970s-1980s. With modern boards, is much more common now to just leave
exposed copper/gaps in the solder mask and allow those areas to take up solder from the
wave soldering (or reflow) processes.
3098
days inactive
3098
days old
3 comments
3 participants
participants (3)
-
applecorey＠optonline.net -
jnc＠mercury.lcs.mit.edu -
tothwolf＠concentric.net