Facit 4040 tape reader/punch
Jarkko Teppo
I just dug up a Facit 4040 tape reader/punch with Facit 5148
interface board.
The interface board has two RS-232 (hopefully) connectors,
female DB-25 marked line and aux. It also has a lot of dip switches
for setting the speed and four toggle switches:
punch on/off, line on/off, aux on/off, local loopback on/off.
The front panel has two switches, one marked reader "ext/start" and
punch "code holes/feed holes".
Now does anybody now how I could use this ie. can I connect a
terminal to the "line"-connector, put some tape on the punch
feed and type away ? I've never used these paper tape devices so
any help appreciated.
I think this was previously used in a DG Nova (the place I got
it from had one) and the stickers indicate that the reader device was
TRA and the punch was TPA.
TIA,
--
Jarkko Teppo
jarkko.teppo <at> nfp.fi
Morris Dovey
Jarkko...
Before you give it a try, make certain that the punch mechanism is
well-lubricated. Turn the motor shaft by hand to make certain it's ready
to operate under power. It'd be a shame to break a punch pin or strip a
gear at this point.
Morris Dovey
West Des Moines, Iowa USA
mrd...@iedu.org
Kragen Sitaker
Jarkko Teppo <jarkko...@nfp.fi> wrote:
>Hello all!
>I just dug up a Facit 4040 tape reader/punch with Facit 5148
>interface board.
>The interface board has two RS-232 (hopefully) connectors,
>female DB-25 marked line and aux. It also has a lot of dip switches
>for setting the speed and four toggle switches:
>punch on/off, line on/off, aux on/off, local loopback on/off.
>The front panel has two switches, one marked reader "ext/start" and
>punch "code holes/feed holes".
I seem to remember from my MODEM7 days (circa 1990 :) that my CP/M
machine used to call the serial-port device PUN: and RDR:. So my guess
is that you can indeed connect at least some tape readers and punches
up to serial ports.
http://www.wps.com/IO/codes/index.html has a great deal of historical
information about ASCII, including a significant amount of information
about what ASCII characters mean to paper tape punches.
I am sure many people here can give you much better information.
Good luck! Let us know what happens!
--
<kra...@pobox.com> Kragen Sitaker <http://www.pobox.com/~kragen/>
Mon Sep 27 1999
42 days until the Internet stock bubble bursts on Monday, 1999-11-08.
<URL:http://www.pobox.com/~kragen/bubble.html>
Peter Maydell
>I just dug up a Facit 4040 tape reader/punch with Facit 5148
>interface board.
>The interface board has two RS-232 (hopefully) connectors,
>female DB-25 marked line and aux.
I wouldn't want to guarantee those to be RS232 myself. I used to
have a Facit 4070 punch, which had a female DB25 connector which
was essentially a parallel connection. I drove it from the
parallel port of a Linux box, with the aid of a little interface
I wired up to invert a couple of the signals. [ARD12 will now
point out that strictly this is wrong, since a parallel port's
data strobe is an edge-signal and the punch is level-triggered
and has timing requirements for that pulse which the kernel
doesn't guarantee to satisfy...]
The punch is still used to encode machine-readable data onto
membership cards for the Cambridge University Computer Preservation
Society (http://www.cam.ac.uk/CambUniv/Societies/cucps/)...
Peter Maydell
Jarkko Teppo
Kragen Sitaker <kra...@dnaco.net> wrote in message
news:vtRH3.2242$J66.2...@typ11.nn.bcandid.com...
>
> http://www.wps.com/IO/codes/index.html has a great deal of
>historical
> information about ASCII, including a significant amount of
>information
> about what ASCII characters mean to paper tape punches.
Thanks, that was quite interesting!
>
> I am sure many people here can give you much better information.
>
> Good luck! Let us know what happens!
Well, I can punch manually using the frontpanel keys. "Code holes"
gives me 8 "data" holes and the feeder hole (smaller one) and
"Feed holes" gives me just the feeder hole. So that works. The
reader also works (at least mechanically) the tape moves etc.
The only problem I have is that I get nothing from the RS-232 port
marked "line". Speed is set to 300bps and I tried using a terminal
set to 300bps (7E1, 7N1, 7?1, 8N1) but no go. I tried to read the
all-holes and no-holes and I didn't even get garbage.
So, assume a connection (null modem or straight ?) from a terminal
to the punch. Should the punch start if a type an "a" for example ?
I've never used these, but I'm feeling lucky after I got my 9-track
tape drive to work :-)
ps. I found a web page describing never(?) models of the Facit line
of punches/readers at:
http://www.westnc.com/facpunch.html
> --
> <kra...@pobox.com> Kragen Sitaker
<http://www.pobox.com/~kragen/>
> Mon Sep 27 1999
> 42 days until the Internet stock bubble bursts on Monday, 1999-11
>08.
> <URL:http://www.pobox.com/~kragen/bubble.html>
Jarkko Teppo
Peter Maydell <pm...@watchdragon.demon.co.uk> wrote in message
news:7sorcs$ql2$1...@watchdragon.demon.co.uk...
> I wouldn't want to guarantee those to be RS232 myself. I used to
> have a Facit 4070 punch, which had a female DB25 connector which
> was essentially a parallel connection. I drove it from the
> parallel port of a Linux box, with the aid of a little interface
> I wired up to invert a couple of the signals. [ARD12 will now
> point out that strictly this is wrong, since a parallel port's
> data strobe is an edge-signal and the punch is level-triggered
> and has timing requirements for that pulse which the kernel
> doesn't guarantee to satisfy...]
Hmm.. I don't know. The interface board has dip switches for
setting the speed for the "aux" and "line" connectors, labeled
(hazy memory) 75,110, 300, 600 (probably something else too, there
are at least 5*8 dip switches on the if-board). If I remember
correctly there was a speed setting of up to 4800 for the reader. I
*really* have to dig up an oscilloscope or something... or maybe
somebody has some documentation ?
>
> The punch is still used to encode machine-readable data onto
> membership cards for the Cambridge University Computer Preservation
Very nice! So do you just blast plain ascii to the punch or ..?
> Society (http://www.cam.ac.uk/CambUniv/Societies/cucps/)...
>
> Peter Maydell
--
Jarkko Teppo
jarkko...@nfp.fi
Tony Duell
: Jarkko Teppo <jarkko...@nfp.fi> wrote:
: >I just dug up a Facit 4040 tape reader/punch with Facit 5148
: >interface board.
: >The interface board has two RS-232 (hopefully) connectors,
: >female DB-25 marked line and aux.
: I wouldn't want to guarantee those to be RS232 myself. I used to
: have a Facit 4070 punch, which had a female DB25 connector which
: was essentially a parallel connection. I drove it from the
That depends on the exact options in the 4070. The 4070 has an internal
edge connector that's electrically between the external DB25-S and the
main logic board (which, indeed has a parellel interface that's almost,
but not quite, centronics).
The 'standard' model has a 'jumper board' in that connector that simply
wires the DB25 to the logic board. But there were other interfaces -- I
have the RS232 card for the 4070 here (==UART + a number of TTL chips)
and I've heard rumours of there being an IEC625 (==IEEE-488/GPIB on DB25
connectors) card as well. Maybe others.
: parallel port of a Linux box, with the aid of a little interface
: I wired up to invert a couple of the signals. [ARD12 will now
: point out that strictly this is wrong, since a parallel port's
: data strobe is an edge-signal and the punch is level-triggered
: and has timing requirements for that pulse which the kernel
: doesn't guarantee to satisfy...]
Firstly I'm not longer ARD12 (I assume I'm who you're refering to).
Secondly, if you hold Punch Instruction (==strobe) asserted for too long
then it will punch multiple characters. If you don't hold it asserted for
long enough, then it won't punch at all. It's not clear to me that the
standard linux lp driver is going to ensure that Punch Instruction is
held asserted for the right length of time under all conditions (other
interrupts occuring, for example), on all machines.
Since its easy to either write a special driver that does this (as I did
as part of a badly-behaved MS-DOS program) or wire up a couple of TTL
chips to interlock Punch Instruction with Punch Ready (as I did when
linking a 4070 to my Intel MCS8i), then I think it should be done
properly :-)
I think this differentiates a hardware hacker from a software hacker. I'm
the former, and I try to ensure that timing conditions are _always_ met
(and not just met most of the time). But I will write quick-and-dirty
code, like statically allocating memory (hopefully 'enough for any
reasonable use' :-)). Software hackers seem to do the reverse :-) :-) :-)
-tony
Peter Maydell
>Peter Maydell <pm...@watchdragon.demon.co.uk> wrote in message
>news:7sorcs$ql2$1...@watchdragon.demon.co.uk...
>Hmm.. I don't know. The interface board has dip switches for
>setting the speed for the "aux" and "line" connectors, labeled
>(hazy memory) 75,110, 300, 600
Sounds serial to me, then.
>> The punch is still used to encode machine-readable data onto
>> membership cards for the Cambridge University Computer Preservation
>> Society
>
>Very nice! So do you just blast plain ascii to the punch or ..?
With a trivial interfacing circuit, it looks more-or-less like a
rather slow printer. I wrote a perl script which runs tunelp to
set the width of the strobe pulse appropriately and then opens
/dev/lp1 and writes data out to the punch, with a big delay between
each character to allow for the fact it's punching card and not
paper tape... (some parts of the membership card are ASCII and
some are 8-bit binary data).
Oh, and there's a trivial mod to the punch to increase the punch
pin drive for punching card.
As I say, there are technically timing problems here which a less
quick-n-dirty hack would avoid, but for just punching membership
cards it suffices.
I've just put the pages I wrote about this up at:
http://www.chiark.greenend.org.uk/~pmaydell/misc/cardpunch/punch.htm
I meant to get them put up on the society website but never got
around to it (hence the somewhat inaccurate footer...)
Peter Maydell