3 Sep
2014
3 Sep
'14
7:27 a.m.
I'm in the process of reproducing front panels for PDP-8/E and/or M machines, as well
as switch levers for these machines and others that use the same shape but different
colors (e.g. 8/L, 11/20).
For the front panels, I'm working with a local (Chicago) silk-screen house to create
the reproductions. At this point (still exploring final pricing), it looks like unit cost
will be about $200 for a run of 5. This will most likely be a CNC cut acrylic panel (laser
probably won't work because of the angled relief for the key switch) with silk-screen
printing front and back. The panel shape is the same for E and M but the screens are quite
different. I've also got a couple different M variations. Note that each panel
requires five or six screens, including black masking on the back and transparent red or
"frosted" masks for the indicator lights in addition to the three colors on the
front.
I'm not sure about possibly adding a transparent overlay (non-original) over the front
silk screening to protect the final product. Please let me know your interest and thoughts
about a protective overlay.
For the switch levers, "phooky" from the NYC Resistors has done a 3D rendering -
http://www.thingiverse.com/thing:360853 - and printed prototype levers and mounting
frames. Note that there are three renderings at the thingverse site - original lever,
lever with improved pivot (metal pin) and bracket for attaching lever to available slide
switch. Phooky has also printed up samples of these objects. The next step is to work on
injection molding the parts. The Resistors have an injection molding setup and phooky is
planning to create a mold using CNC equipment and his 3D renderings. There are several
reasons for moving from 3D printing to injection molding for these parts. The first two,
finish and mechanical strength, are slowly being addressed by improved printers but the
third reason, color matching, is the critical difference.
The front panel project is in progress and hopefully panels will be available in a few
months.
The switch project is very much exploratory with no promises or dates at this time.
Comments please!
Jack
3 Sep
3 Sep
9:35 a.m.
Very cool project, I've got several 8's that I can provide photos of if
that would help (8/e , 8/f, 8/m). I've been looking at a similar project
with my 3D printer but rather than re-create the PDP-8 switch handle
directly, build a switch handle that would work with modern (and
inexpensive) tactile buttons. My goal is to have a way of creating an
arbitrary front panel.
From my experience you will want to injection mold
these switches as the
variability between printers would probably render parts of
one switch on
one printer from not working with parts printed on a different printer.
Such is the newness of these things.
--Chuck
On Wed, Sep 3, 2014 at 8:27 AM, Jack Rubin <j at ckrubin.us> wrote:
I'm in the process of reproducing front panels
for PDP-8/E and/or M
machines, as well as switch levers for these machines and others that use
the same shape but different colors (e.g. 8/L, 11/20).
For the front panels, I'm working with a local (Chicago) silk-screen house
to create the reproductions. At this point (still exploring final pricing),
it looks like unit cost will be about $200 for a run of 5. This will most
likely be a CNC cut acrylic panel (laser probably won't work because of the
angled relief for the key switch) with silk-screen printing front and back.
The panel shape is the same for E and M but the screens are quite
different. I've also got a couple different M variations. Note that each
panel requires five or six screens, including black masking on the back and
transparent red or "frosted" masks for the indicator lights in addition to
the three colors on the front.
I'm not sure about possibly adding a transparent overlay (non-original)
over the front silk screening to protect the final product. Please let me
know your interest and thoughts about a protective overlay.
For the switch levers, "phooky" from the NYC Resistors has done a 3D
rendering - http://www.thingiverse.com/thing:360853 - and printed
prototype levers and mounting frames. Note that there are three renderings
at the thingverse site - original lever, lever with improved pivot (metal
pin) and bracket for attaching lever to available slide switch. Phooky has
also printed up samples of these objects. The next step is to work on
injection molding the parts. The Resistors have an injection molding setup
and phooky is planning to create a mold using CNC equipment and his 3D
renderings. There are several reasons for moving from 3D printing to
injection molding for these parts. The first two, finish and mechanical
strength, are slowly being addressed by improved printers but the third
reason, color matching, is the critical difference.
The front panel project is in progress and hopefully panels will be
available in a few months.
The switch project is very much exploratory with no promises or dates at
this time.
Comments please!
Jack
1:34 p.m.
On Wed, Sep 3, 2014 at 11:27 AM, Jack Rubin <j at ckrubin.us> wrote:
I'm in the process of reproducing... switch
levers (e.g. 8/L, 11/20).
For the switch levers, "phooky" from the NYC Resistors has done a 3D rendering
- http://www.thingiverse.com/thing:360853
Yes! We have needed this for years! (I'm not a 3D design guy so this
has been a little beyond my skill level or I would have done it a long
time ago).
Phooky has also printed up samples of these objects.
How well did they work? (I realize that FDM-printed parts wouldn't
"look as nice" - I'm curious about close enough to work without
breaking, as a stopgap)
The next step is to work on injection molding the
parts. The Resistors have an injection molding setup and phooky is planning to create a
mold using CNC equipment and his 3D renderings.
Very nice!
There are several reasons for moving from 3D printing
to injection molding for these parts. The first two, finish and mechanical strength, are
slowly being addressed by improved printers but the third reason, color matching, is the
critical difference.
Yes. Finish is entirely different for FDM vs injection molding. One
_could_ pay for a custom run of filament to match, but the resulting
part would still be striated.
I haven't fiddled with injection molding, but I look forward to the reports.
-ethan
P.S. - if this is a success, perhaps there's enough interest for
11-70/35/45/etc triangular toggles. I need a bunch for an 11/70 front
panel I got from Jay that he harvested all the toggles from before
giving me the carcass.
2:20 p.m.
On 03/09/2014 16:27, Jack Rubin wrote:
I'm in the process of reproducing front panels for
PDP-8/E and/or M
[...]
For the front panels, I'm working with a local
(Chicago) silk-screen
house to create the reproductions.
[...]
The panel shape is the same
for E and M but the screens are quite different. I've also got a
couple different M variations. Note that each panel requires five or
six screens, including black masking on the back and transparent red
or "frosted" masks for the indicator lights in addition to the three
colors on the front.
There are also (at least) two variants of the 8/E panel screening.
The early ones used a rotary switch with 36-degree spacing, so the
highest position (STATE) and the lowest (BUS) are 180 degrees apart, and
the lines for those are vertical. On this panel, there are no vertical
white lines between the octal groups, and if you look closely you'll see
that there is no gap between the EMA section and the rest of the MEMORY
ADDRESS legend. The darker orange colouring is also slightly brighter
(or less dull) than on later panels. The early panels often had an
octal listing of the RIM loader on the right, above the
HALT/SING_STEP/DEP switches, but that's a glued-on plaque, not part of
the silkscreen.
The later panels were redesigned to accommodate a more conventional
rotary switch with 30-degree spacing (so like a 12-position switch
instead of a 10-position one) and the lines from the highest and lowest
positions are not vertical; they're 15 degrees to the left. There are
five vertical white lines between the MEMORY ADDRESS legend and the 0-11
positions below, separating the two rows of lamps into four groups of
three (octal numbers). There's a small gap between the EMA section and
the rest of the MEMORY ADDRESS legend, and the darker orange colour is a
little duller (might just be batch variation, though). I don't recall
ever seeing a RIM Loader listing on one of these, but I might not have
paid enough attention to every one I've seen so don't quote me on that
part :-)
None of the 8/E panels I've seen have a red wash for the lamp positions,
just a little frosting.
I've seen a reference to the rotary switch change and the angles in one
of the DEC manuals or maybe an ECO, but can't see it just at the moment.
I've a picture showing both panels (er, machines!) side by side at
https://www.flickr.com/photos/pnt103/9365750952/
--
Pete Peter Turnbull
Network Manager
University of York
2:30 p.m.
On 03/09/2014 16:27, Jack Rubin wrote:
I'm in the process of reproducing front panels for
PDP-8/E and/or M
machines, as well as switch levers for these machines and others that
use the same shape but different colors (e.g. 8/L, 11/20).
The switch project is very much exploratory with no
promises or dates
at this time.
One of my friends might be interested in a few switch levers to repair
one or two existing panels. AFAIR the panels and slide switches
themselves are intact.
--
Pete Peter Turnbull
Network Manager
University of York
3:19 p.m.
On Sep 6, 2014, at 10:51 AM, "Brian L. Stuart" <blstuart at bellsouth.net>
wrote:
I second that, I need a load address and deposit switch for my 11/20.
The switch
project is very much exploratory with no promises
or dates at this time.
I'm definitely interested in the switch project. I've got an 8/M
that I'm about to start working on restoring, and it's missing one
of the paddles.
BLS
3591
days inactive
3594
days old
6 comments
6 participants
participants (6)
-
blstuart＠bellsouth.net -
chuck.mcmanis＠gmail.com -
ethan.dicks＠gmail.com -
j＠ckrubin.us -
pete＠dunnington.plus.com -
roeapeterson＠gmail.com