1 May
2020
1 May
'20
8:31 a.m.
At 07:18 AM 5/1/2020, Hugh Pyle via cctalk wrote:
I've cut Mylar tape with a Glowforge laser. It
cuts very nicely but the
alignment is a major hassle, plus you can only cut ~15" of tape which
doesn't go very far. Not worth the effort. If you were to build a custom
linear drive it might work. But also very slow.
Hmm. You could have N fixed lasers at the spots of potential holes,
and then a mechanism to move the whole assembly of them in the shape of a
single hole, drawing them all at once.
You could have one laser on that moved precisely along the hole row,
and use the same sort of mechanical motion to draw a hole.
How much laser do you need to cut paper, how much to cut mylar?
Were there any paper tape devices that did not use the sprocket holes
to move the tape?
- John
1 May
1 May
8:39 a.m.
On Fri, May 1, 2020 at 4:32 PM John Foust via cctalk
<cctalk at classiccmp.org> wrote:
Were there any paper tape devices that did not use the
sprocket holes
to move the tape?
Yes, both punches and readers.
The Facit 4070 (and the N4000) punch uses a capstan and pinch roller
to move the tape. It's driven by a stepper motor. Of course it punches
the sprocket holes.
As I said the other day several high-quality optical tape readers
(Trend, HP2748) feed the tape with a capstan and pinch roller. They do
take a timing signal from the sprocket holes of course.
-tony
- John
8:48 a.m.
Wow, what a response! Really appreciate the docs and first-hand experience,
this is super helpful.
I'm also floored by the complexity of that Roytron punch! Looks like it
contains around one hundred separate parts. I'm convinced the punch parts
will have to be precision metal so while that's not quite "DIY" it might
still be a reasonable bit to CNC, mill or water-jet cut.
It also looks like that punch has an escapement mechanism of some sort?
Seems like a simple way to keep regular spacing, but if course more parts
to buy from McMaster Carr or whatever.
Question: did all tapes have indexing holes separate from the drive
sprocket holes? Also is there a source for tape with sprocket holes anymore?
I also like the idea of using a LASER cutter and a rotating platform like
they use to engrave drinking glasses. Add a light vacuum nozzle under the
tape past the cutting surface to suck up the chads, and perhaps make the
cutting surface wheel with of rows of thin plates as the cutting surface,
similar to the thin honeycomb cutting beds of typical LASER cutters. I
imagine it'd take an awfully long time to use a traditional core XY cutter
versus a galvo-mirror cutter.
I have some experience cutting planar items in LASER cutters but little
experience beyond basic principles regarding machining/modeling/rotating
LASER stuff.
On Fri, May 1, 2020, 11:32 AM John Foust via cctalk <cctalk at classiccmp.org>
wrote:
At 07:18 AM 5/1/2020, Hugh Pyle via cctalk wrote:
I've cut Mylar tape with a Glowforge laser.
It cuts very nicely but the
alignment is a major hassle, plus you can only cut ~15" of tape which
doesn't go very far. Not worth the effort. If you were to build a custom
linear drive it might work. But also very slow.
Hmm. You could have N fixed lasers at the spots of potential holes,
and then a mechanism to move the whole assembly of them in the shape of a
single hole, drawing them all at once.
You could have one laser on that moved precisely along the hole row,
and use the same sort of mechanical motion to draw a hole.
How much laser do you need to cut paper, how much to cut mylar?
Were there any paper tape devices that did not use the sprocket holes
to move the tape?
- John
9:05 a.m.
On Fri, May 1, 2020 at 4:48 PM Anders Nelson via cctalk
<cctalk at classiccmp.org> wrote:
Wow, what a response! Really appreciate the docs and first-hand experience,
this is super helpful.
I feel very greedy now. Looking arounds I have (not all working, but
all could be got to work, it's things like drive belts I need to get )
3 Facit 4070s and an N4000
A DEC PC04
Two Data Dynamics units (1130 and 1183) which are 8-level punches
3 Teletype BRPEs
A Trend Paper Tape Station (GNT 34 punch and Trend HSR500 reader in a
rack chassis)
A couple of Teletype Model 33 ASR's
A Friden Flexowriter with punch/reader
The punch from a Flexowriter on a rack chassis with its own motor.
I think there's a couple of bare-chassis solenoid-operated punches
here too but I would have to hunt for them.
And that's the 8 level ones. For 5 level (only) I have a Creed 444
teleprinter and a Booth-Willmott Model 44 Keypunch (totally mechanical
between the keyboard and the 5 punch pins so not much use for computer
interfacing).
I'm also floored by the complexity of that Roytron punch! Looks like it
contains around one hundred separate parts. I'm convinced the punch parts
will have to be precision metal so while that's not quite "DIY" it might
still be a reasonable bit to CNC, mill or water-jet cut.
It also looks like that punch has an escapement mechanism of some sort?
Seems like a simple way to keep regular spacing, but if course more parts
to buy from McMaster Carr or whatever.
Some punches fed the tape with a sprocket wheel, one way to drive that
is a suitably toothed ratchet wheel and drive pawl that moves it one
tooth per punch cycle. As I said in another message, the Facit 4070
uses an accurately-machined (right diameter) capstan and pinch roller,
the capstan moves the tape one chracter spacing for a certain number
of steps (it _may_ be just one step) of the motor. I have another
punch unit here which has the crank-and-interposer mechanism to punch
the holes but has as sprocket with its own stepper motor to feed the
tape. Different manufacturers did it in different ways.
Question: did all tapes have indexing holes separate from the drive
Eh? The sprocket holes _are_ the indexing holes, surely?
sprocket holes? Also is there a source for tape with
sprocket holes anymore?
Every punch I've seen (and own) can punch the sprocket holes on
totally blank tape. Some insist on doing it (there's no way to punch a
character without a sprocket hole), for others the sprocket punch is
just another pin operated by the same type of mechanism as the data
punches and you have to fire the solenoid at the right time if you
want a sprocket hole.
-tony
10:20 a.m.
Tony, maybe your collection can help me answer a puzzle: which side is
"top"? By my reading, for 8-level tape,
- ANSI and other US standards have three data bits / index / then five data
bits
- ECMA has five/index/three... :)
https://twitter.com/33asr/status/1138758004747177984
On Fri, May 1, 2020 at 12:05 PM Tony Duell via cctalk <cctalk at classiccmp.org>
wrote:
On Fri, May 1, 2020 at 4:48 PM Anders Nelson via
cctalk
<cctalk at classiccmp.org> wrote:
Wow, what a response! Really appreciate the docs and first-hand
experience,
this is super helpful.
I feel very greedy now. Looking arounds I have (not all working, but
all could be got to work, it's things like drive belts I need to get )
3 Facit 4070s and an N4000
A DEC PC04
Two Data Dynamics units (1130 and 1183) which are 8-level punches
3 Teletype BRPEs
A Trend Paper Tape Station (GNT 34 punch and Trend HSR500 reader in a
rack chassis)
A couple of Teletype Model 33 ASR's
A Friden Flexowriter with punch/reader
The punch from a Flexowriter on a rack chassis with its own motor.
I think there's a couple of bare-chassis solenoid-operated punches
here too but I would have to hunt for them.
And that's the 8 level ones. For 5 level (only) I have a Creed 444
teleprinter and a Booth-Willmott Model 44 Keypunch (totally mechanical
between the keyboard and the 5 punch pins so not much use for computer
interfacing).
I'm also floored by the complexity of that Roytron punch! Looks like it
contains around one hundred separate parts. I'm convinced the punch parts
will have to be precision metal so while that's not quite "DIY" it might
still be a reasonable bit to CNC, mill or water-jet cut.
It also looks like that punch has an escapement mechanism of some sort?
Seems like a simple way to keep regular spacing, but if course more parts
to buy from McMaster Carr or whatever.
Some punches fed the tape with a sprocket wheel, one way to drive that
is a suitably toothed ratchet wheel and drive pawl that moves it one
tooth per punch cycle. As I said in another message, the Facit 4070
uses an accurately-machined (right diameter) capstan and pinch roller,
the capstan moves the tape one chracter spacing for a certain number
of steps (it _may_ be just one step) of the motor. I have another
punch unit here which has the crank-and-interposer mechanism to punch
the holes but has as sprocket with its own stepper motor to feed the
tape. Different manufacturers did it in different ways.
Question: did all tapes have indexing holes separate from the drive
Eh? The sprocket holes _are_ the indexing holes, surely?
sprocket holes? Also is there a source for tape
with sprocket holes
anymore?
Every punch I've seen (and own) can punch the sprocket holes on
totally blank tape. Some insist on doing it (there's no way to punch a
character without a sprocket hole), for others the sprocket punch is
just another pin operated by the same type of mechanism as the data
punches and you have to fire the solenoid at the right time if you
want a sprocket hole.
-tony
10:32 a.m.
On Fri, May 1, 2020 at 6:20 PM Hugh Pyle <hpyle at cabezal.com> wrote:
Tony, maybe your collection can help me answer a puzzle: which side is "top"?
By my reading, for 8-level tape,
- ANSI and other US standards have three data bits / index / then five data bits
- ECMA has five/index/three... :)
https://twitter.com/33asr/status/1138758004747177984
As far as I know the 5-hole side is always the most significant 5
bits, the 3 hole side the least significant 3 bits. In other words the
holes go :
76543S210
Where S is the sprocket hole and a digit is the bit number (bit 0 is
the least significant bit).
As to which way use the tape, with the tape feeding towards you, the
sprocket hole is normally offset towards the right, that is the bits
would read with the MSB on the left. I am not sure if that is
universal though
-tony
10:47 a.m.
On May 1, 2020, at 1:32 PM, Tony Duell via cctalk
<cctalk at classiccmp.org> wrote:
On Fri, May 1, 2020 at 6:20 PM Hugh Pyle <hpyle at cabezal.com> wrote:
I suppose things get really confusing when dealing with 6 row tape
(https://www.smecc.org/teletypes_in_typesetting.htm)
paul
Tony, maybe your collection can help me answer a puzzle: which side is "top"?
By my reading, for 8-level tape,
- ANSI and other US standards have three data bits / index / then five data bits
- ECMA has five/index/three... :)
https://twitter.com/33asr/status/1138758004747177984
As far as I know the 5-hole side is always the most significant 5
bits, the 3 hole side the least significant 3 bits. In other words the
holes go :
76543S210
Where S is the sprocket hole and a digit is the bit number (bit 0 is
the least significant bit).
As to which way use the tape, with the tape feeding towards you, the
sprocket hole is normally offset towards the right, that is the bits
would read with the MSB on the left. I am not sure if that is
universal though
-tony 
2 May
2 May
4:14 a.m.
On Fri, May 1, 2020 at 7:32 PM Tony Duell via cctalk
<cctalk at classiccmp.org> wrote:
As far as I know the 5-hole side is always the most significant 5
bits, the 3 hole side the least significant 3 bits. In other words the
holes go :
76543S210
Where S is the sprocket hole and a digit is the bit number (bit 0 is
the least significant bit).
As to which way use the tape, with the tape feeding towards you, the
sprocket hole is normally offset towards the right, that is the bits
would read with the MSB on the left. I am not sure if that is
universal though
FWIW, the ECMA standard is here https://www.polyomino.org.uk/computer/ECMA-10/
(found via Wikipedia)
--
Regards,
Torfinn Ingolfsen
7:02 a.m.
If you go with punch pins, standard drill rod is not hard to get. It often comes in
materials that can be easily heat hardened.
Dwight
________________________________
From: cctalk <cctalk-bounces at classiccmp.org> on behalf of Torfinn Ingolfsen via
cctalk <cctalk at classiccmp.org>
Sent: Saturday, May 2, 2020 4:14 AM
To: General Discussion: On-Topic and Off-Topic Posts <cctalk at classiccmp.org>
Subject: Re: DIY Paper Tape Punch - Mechanism diagram?
On Fri, May 1, 2020 at 7:32 PM Tony Duell via cctalk
<cctalk at classiccmp.org> wrote:
As far as I know the 5-hole side is always the most significant 5
bits, the 3 hole side the least significant 3 bits. In other words the
holes go :
76543S210
Where S is the sprocket hole and a digit is the bit number (bit 0 is
the least significant bit).
As to which way use the tape, with the tape feeding towards you, the
sprocket hole is normally offset towards the right, that is the bits
would read with the MSB on the left. I am not sure if that is
universal though
FWIW, the ECMA standard is here https://www.polyomino.org.uk/computer/ECMA-10/
(found via Wikipedia)
--
Regards,
Torfinn Ingolfsen
1 May
1 May
9:50 a.m.
On May 1, 2020, at 11:31 AM, John Foust via cctalk
<cctalk at classiccmp.org> wrote:
Were there any paper tape devices that did not use the sprocket holes
to move the tape?
A lot of high speed optical readers use capstans and pinch rollers. But they still need
the sprocket holes to be the source of the clock signal for latching the data.
paul
1694
days inactive
1695
days old
9 comments
7 participants
participants (7)
-
anders.k.nelson＠gmail.com -
ard.p850ug1＠gmail.com -
dkelvey＠hotmail.com -
hpyle＠cabezal.com -
jfoust＠threedee.com -
paulkoning＠comcast.net -
tingox＠gmail.com