20 Dec
2024
20 Dec
'24
4:36 p.m.
The IBM 1403 printer had interchangeable print chains. I know of only
four 1403 printers still working — two at the Computer History Museum
in Mountain View, CA, one at the IBM Technology Center in Böblingen,
Germany, and one near Endicott, NY.
All four have the 48-character "A" or "Business" chain, and CHM has
a
16-character numeric chain that allows the printer to run twice as fast
for numeric-only output. CHM doesn't have an "H" or "Fortran"
chain,
and as far as I know, none of the others do. The difference is that
parentheses are % and "lozenge" — a square with indented edges
— apostrophe is @, and = is # on the "A" chain. IBM also had a 64-
character chain that included box and line drawing graphics. BTW,
nobody seems to know what "lozenge" was meant to represent.
Does anybody know of an existing "H" chain or graphics chain for a
1403?
Van Snyder
20 Dec
20 Dec
5:35 p.m.
On 12/20/24 16:36, Van Snyder via cctalk wrote:
The IBM 1403 printer had interchangeable print chains.
I know of only
four 1403 printers still working — two at the Computer History Museum
in Mountain View, CA, one at the IBM Technology Center in Böblingen,
Germany, and one near Endicott, NY.
All four have the 48-character "A" or "Business" chain, and CHM has
a
16-character numeric chain that allows the printer to run twice as fast
for numeric-only output. CHM doesn't have an "H" or "Fortran"
chain,
and as far as I know, none of the others do. The difference is that
parentheses are % and "lozenge" — a square with indented edges
— apostrophe is @, and = is # on the "A" chain. IBM also had a 64-
character chain that included box and line drawing graphics. BTW,
nobody seems to know what "lozenge" was meant to represent.
I recall only learning once not to leave a cup of coffee atop a running
1403. I think the lozenge-character print train was mostly used on the
BCD machines like the 1401, but I could be mistaken.
I recall one of the old timers at CDC relating that try as they might,
they couldn't get a train printer of their own design without the print
train disintegrating. I believe that CDC quietly purchased a 1403 on
the gray market and took it to pieces. The result was the CDC 512. Up
until this time, the standard high-speed printer at CDC was the 501--a
drum printer.
It was interesting to note that hammer firing differences on a drum
printer resulted in characters being vertically displaced, which was
very annoying to the eye. Whereas the train printers could displace
characters horizontally and was not nearly as unsightly.
--Chuck
7:39 p.m.
On 2024-12-20 21:35, Chuck Guzis via cctalk wrote:
On 12/20/24 16:36, Van Snyder via cctalk wrote:
The chain with box drawing characters mention in the original post where
used to print the ALD. The 1403 had logic that limited the number of
hammers that could fire at once, there was a test routine that would
repeatedly fire the maximum number of hammers it was called the "Chain
Breaker Routine". The only drum printer I ever saw operating I think it
was a Honeywell printer and the person demoing it printed out some
pictures, the printer could fire most if not all hammers at once which
made quite a racket.
Paul.
The IBM 1403 printer had interchangeable print
chains. I know of only
four 1403 printers still working — two at the Computer History Museum
in Mountain View, CA, one at the IBM Technology Center in Böblingen,
Germany, and one near Endicott, NY.
All four have the 48-character "A" or "Business" chain, and CHM has
a
16-character numeric chain that allows the printer to run twice as fast
for numeric-only output. CHM doesn't have an "H" or "Fortran"
chain,
and as far as I know, none of the others do. The difference is that
parentheses are % and "lozenge" — a square with indented edges
— apostrophe is @, and = is # on the "A" chain. IBM also had a 64-
character chain that included box and line drawing graphics. BTW,
nobody seems to know what "lozenge" was meant to represent.
I recall
only learning once not to leave a cup of coffee atop a running
1403. I think the lozenge-character print train was mostly used on the
BCD machines like the 1401, but I could be mistaken.
I recall one of the old timers at CDC relating that try as they might,
they couldn't get a train printer of their own design without the print
train disintegrating. I believe that CDC quietly purchased a 1403 on
the gray market and took it to pieces. The result was the CDC 512. Up
until this time, the standard high-speed printer at CDC was the 501--a
drum printer.
It was interesting to note that hammer firing differences on a drum
printer resulted in characters being vertically displaced, which was
very annoying to the eye. Whereas the train printers could displace
characters horizontally and was not nearly as unsightly.
--Chuck
21 Dec
21 Dec
5:15 a.m.
On Dec 20, 2024, at 10:39 PM, Paul Berger via cctalk
<cctalk(a)classiccmp.org> wrote:
The chain with box drawing characters mention in the original post where used to print
the ALD. The 1403 had logic that limited the number of hammers that could fire at once,
there was a test routine that would repeatedly fire the maximum number of hammers it was
called the "Chain Breaker Routine". The only drum printer I ever saw operating
I think it was a Honeywell printer and the person demoing it printed out some pictures,
the printer could fire most if not all hammers at once which made quite a racket.
Paul.
As a field engineer for Honeywell, I always dreaded the holidays because so many people
would launch print jobs which used repeated overstrikes to create pictures. Those jobs
sometimes fired the maximum number of hammers at a time to speed up the picture creation
which would sometimes cause multiple hammer actuator fuses to blow. More than once I had
to buy all the 2 amp fuses from multiple Radio Shacks to get the printer operational
again. Those overstrikes also caused the paper to become more saturated with ink which
resulted in more paper/ink residue getting deposited in the print chain, which required
heavier than normal cleanings during the next preventative maintenance window.
Another thing which caused more work for field engineers around the holidays were jobs
sent to the card punches to play Jingle Bells by punching fully laced cards in time to the
music. It was entertaining unless they caused card jams too bad for the operators to be
able to clear by themselves. Fully laced punch cards are too flexible to pass through the
punch path cleanly.
- Rod
5:53 a.m.
Rod Bartlett wrote:
You mean like this?
https://www.ibmjunkman.com/cards/?Holder=6309&Img=1
On Dec 20,
2024, at 10:39 PM, Paul Berger via cctalk
<cctalk(a)classiccmp.org> wrote:
The chain with box drawing characters mention in the original post where used to print
the ALD. The 1403 had logic that limited the number of hammers that could fire at once,
there was a test routine that would repeatedly fire the maximum number of hammers it was
called the "Chain Breaker Routine". The only drum printer I ever saw
operating
I think it was a Honeywell printer and the person demoing it printed out some pictures,
the printer could fire most if not all hammers at once which made quite a racket.
Paul.
As a field engineer for Honeywell, I always dreaded the holidays because
so many people
would launch print jobs which used repeated overstrikes to create pictures. Those jobs
sometimes fired the maximum number of hammers at a time to speed up the picture creation
which would sometimes cause multiple hammer actuator fuses to blow. More than once I
had
to buy all the 2 amp fuses from multiple Radio Shacks to get the printer operational
again. Those overstrikes also caused the paper to become more saturated with ink which
resulted in more paper/ink residue getting deposited in the print chain, which required
heavier than normal cleanings during the next preventative maintenance window.
Another thing which caused more work for field engineers around the holidays were jobs
sent to the card punches to play Jingle Bells by punching fully laced cards in time to
the
music. It was entertaining unless they caused card jams too bad for the operators to be
able to clear by themselves. Fully laced punch cards are too flexible to pass through
the
punch path cleanly.
- Rod
6:49 a.m.
On Dec 21, 2024, at 8:53 AM, Donald Whittemore via
cctalk <cctalk(a)classiccmp.org> wrote:
Rod Bartlett wrote:
As a field engineer for Honeywell, I always dreaded the holidays because so many people
would launch print jobs which used repeated overstrikes to create pictures. Those jobs
sometimes fired the maximum number of hammers at a time to speed up the picture creation
which would sometimes cause multiple hammer actuator fuses to blow. More than once I
had
to buy all the 2 amp fuses from multiple Radio Shacks to get the printer operational
again. Those overstrikes also caused the paper to become more saturated with ink which
resulted in more paper/ink residue getting deposited in the print chain, which required
heavier than normal cleanings during the next preventative maintenance window.
Another thing which caused more work for field engineers around the holidays were jobs
sent to the card punches to play Jingle Bells by punching fully laced cards in time to
the
music. It was entertaining unless they caused card jams too bad for the operators to be
able to clear by themselves. Fully laced punch cards are too flexible to pass through
the
punch path cleanly.
- Rod
You mean like this?
https://www.ibmjunkman.com/cards/?Holder=6309&Img=1
That's it. As far as I know, fully laced cards were only useful for producing a deep
percussive sound while punching the card. Trying to push cards rendered flexible by so
many holes at 300 cards per minute caused some spectacular card jams which occasionally
required partial disassembly to fully clear.
9:42 a.m.
On 2024-12-21 09:49, Rod Bartlett via cctalk wrote:
You're lucky, Rod. When I was a trainee FE at a Univac site, we had a
smart instructor that accidentally managed to block the reset signal of
the hammers on a 1004 drum printer. All 132 columns struck the drum,
gouged the surface, and blew hundreds of power transistors! Univac had
to fly parts (and an FE) from all around the world to get us going.
The sad part is that the 1004 was used for billing blue chip companies
that used our store-and-forward message switcher (using teletype
machines)! After that episode, they bought another 1004 for the backup
system!
Oh! The memories!
cheers,
Nigel
--
Nigel Johnson, MSc., MIEEE, MCSE VE3ID/G4AJQ/VA3MCU
Amateur Radio, the origin of the open-source concept!
Skype: TILBURY2591
On Dec 21,
2024, at 8:53 AM, Donald Whittemore via cctalk<cctalk(a)classiccmp.org> wrote:
Rod Bartlett wrote:
That's it. As far as I know, fully laced cards were only useful for producing
a deep percussive sound while punching the card. Trying to push cards rendered flexible
by so many holes at 300 cards per minute caused some spectacular card jams which
occasionally required partial disassembly to fully clear.
As a field engineer for Honeywell, I always
dreaded the holidays because so many people
would launch print jobs which used repeated overstrikes to create pictures. Those jobs
sometimes fired the maximum number of hammers at a time to speed up the picture
creation
which would sometimes cause multiple hammer actuator fuses to blow. More than once I
had
to buy all the 2 amp fuses from multiple Radio Shacks to get the printer operational
again. Those overstrikes also caused the paper to become more saturated with ink which
resulted in more paper/ink residue getting deposited in the print chain, which required
heavier than normal cleanings during the next preventative maintenance window.
Another thing which caused more work for field engineers around the holidays were jobs
sent to the card punches to play Jingle Bells by punching fully laced cards in time to
the
music. It was entertaining unless they caused card jams too bad for the operators to
be
able to clear by themselves. Fully laced punch cards are too flexible to pass through
the
punch path cleanly.
- Rod
You mean like this?
https://www.ibmjunkman.com/cards/?Holder=6309&Img=1
9:26 a.m.
On 12/21/24 07:53, Donald Whittemore via cctalk wrote:
Rod Bartlett wrote:
You mean like this?
https://www.ibmjunkman.com/cards/?Holder=6309&Img=1
I made these by hand on a 514AD duplicating card punch. I
would make cards on a regular keypunch with 4 rows of holes
(all the same character 80 times), such that when flipped
and run through the duplicator 3 times onto the same output
deck, it would punch all holes. Occasionally the 514 would
jam, but it was not that hard to open up and clean out.
Jon
On Dec
20, 2024, at 10:39 PM, Paul Berger via cctalk
<cctalk(a)classiccmp.org> wrote:
The chain with box drawing characters mention in the original post where used to
print
the ALD. The 1403 had logic that limited the number of hammers that could fire at
once,
there was a test routine that would repeatedly fire the maximum number of hammers it
was
called the "Chain Breaker Routine". The only drum printer I ever saw
operating
I think it was a Honeywell printer and the person demoing it printed out some pictures,
the printer could fire most if not all hammers at once which made quite a racket.
Paul.
As a field engineer for Honeywell, I always dreaded the holidays because so many
people
would launch print jobs which used repeated overstrikes to create pictures. Those jobs
sometimes fired the maximum number of hammers at a time to speed up the picture
creation
which would sometimes cause multiple hammer actuator fuses to blow. More than once I
had
to buy all the 2 amp fuses from multiple Radio Shacks to get the printer operational
again. Those overstrikes also caused the paper to become more saturated with ink which
resulted in more paper/ink residue getting deposited in the print chain, which required
heavier than normal cleanings during the next preventative maintenance window.
Another thing which caused more work for field engineers around the holidays were jobs
sent to the card punches to play Jingle Bells by punching fully laced cards in time to
the
music. It was entertaining unless they caused card jams too bad for the operators to
be
able to clear by themselves. Fully laced punch cards are too flexible to pass through
the
punch path cleanly.
- Rod
9:19 a.m.
On 12/20/24 21:39, Paul Berger via cctalk wrote:
The chain with box drawing characters mention in the
original post where used to print the ALD. The 1403 had
logic that limited the number of hammers that could fire
at once, there was a test routine that would repeatedly
fire the maximum number of hammers it was called the
"Chain Breaker Routine". The only drum printer I ever saw
operating I think it was a Honeywell printer and the
person demoing it printed out some pictures, the printer
could fire most if not all hammers at once which made
quite a racket.
Paul.
I had a BIG Honeywell drum printer on my S-100 Z80 system.
It was from a system used to print from tapes, offline. It
had some odd characters for making boxes on printed forms.
There were shift registers that controlled the hammer
transistors in pairs. Even column characters on the drum
lined up with the hammers alternately with the odd columns.
The same driver transistor was used for adjacent hammers,
but the hammers were actually fired by big SCRs that
alternately powered two power rails. The SCRs were
commutated by big Germanium transistors. While getting this
beast running, I managed to disconnect an important signal
cable and it caused all hammers to fire every character
time, and the hammers pounded the thin metal shield in front
of them. The noise could probably have been heard a block
away! It also blew out the commutating transistors. I was
able to replace them with some TO-3 transistors out of my
junk box, and it worked fine for the rest of the time I had
it. I did use overprints to create some of the extended
ASCII characters, so my Pascal listings could only be read
by me. Such things as curly braces were overprints of ( and
<, for instance. If you printed a full line of * or _ it
made a loud "bump" noise!
Jon
10:29 a.m.
On 12/21/24 09:19, Jon Elson via cctalk wrote:
I had a BIG Honeywell drum printer on my S-100 Z80
system. It was from
a system used to print from tapes, offline.
Just wondering if anyone but yours truly used the Teletype Dataspeed 40
band printer on their PCs. 150 LPM if memory serves. I still have the
schematics for the thing. I ditched mine along with my daisywheel
printer when laser printers became the state of the art.
--Chuck
9:07 a.m.
On 12/20/24 18:36, Van Snyder via cctalk wrote:
The IBM 1403 printer had interchangeable print chains.
I know of only
four 1403 printers still working — two at the Computer History Museum
in Mountain View, CA, one at the IBM Technology Center in Böblingen,
Germany, and one near Endicott, NY.
All four have the 48-character "A" or "Business" chain, and CHM has
a
16-character numeric chain that allows the printer to run twice as fast
for numeric-only output. CHM doesn't have an "H" or "Fortran"
chain,
and as far as I know, none of the others do. The difference is that
parentheses are % and "lozenge" — a square with indented edges
— apostrophe is @, and = is # on the "A" chain. IBM also had a 64-
character chain that included box and line drawing graphics. BTW,
nobody seems to know what "lozenge" was meant to represent.
Does anybody know of an existing "H" chain or graphics chain for a
1403?
Van Snyder
My understanding is that the 1403 N1 used a "train" and the
1403 N3 used a chain. I have seen a 1403 N1 train up close,
and there was no chain that connected the type blocks, 2
characters to a block.
Jon
9:41 a.m.
my recollection from being a computer operator at the time was the the earlier 1403s did
NOT have an interchangeable chain.
Also curious if any 1404 printers still exist. They were wider and you could shift the
print mechanism to the side where there was a mechanisms to print on tabulating cards, one
or two at a time.
<pre>--Carey</pre>
On 12/21/2024 11:07 AM CST Jon Elson via cctalk
<cctalk(a)classiccmp.org> wrote:
On 12/20/24 18:36, Van Snyder via cctalk wrote:
> The IBM 1403 printer had interchangeable print chains. I know of only
> four 1403 printers still working — two at the Computer History Museum
> in Mountain View, CA, one at the IBM Technology Center in Böblingen,
> Germany, and one near Endicott, NY.
>
> All four have the 48-character "A" or "Business" chain, and CHM
has a
> 16-character numeric chain that allows the printer to run twice as fast
> for numeric-only output. CHM doesn't have an "H" or
"Fortran" chain,
> and as far as I know, none of the others do. The difference is that
> parentheses are % and "lozenge" — a square with indented edges
> — apostrophe is @, and = is # on the "A" chain. IBM also had a 64-
> character chain that included box and line drawing graphics. BTW,
> nobody seems to know what "lozenge" was meant to represent.
>
> Does anybody know of an existing "H" chain or graphics chain for a
> 1403?
>
> Van Snyder
12:30 p.m.
On Dec 21, 2024, at 12:41 PM, CAREY SCHUG via cctalk
<cctalk(a)classiccmp.org> wrote:
my recollection from being a computer operator at the time was the the earlier 1403s did
NOT have an interchangeable chain.
Interesting. I was an occasional operator on a 360 model 44 in 1974, and it had an
interchangeable chain. That installation had several, including an upper/lower case one
(TN perhaps) which I used to print one of my college papers (formatted with RUNOFF on a
PDP-11 system, carried to the 360 on tape and printed there). The printer also had a film
ribbon, basically a 17 inch wide version of the ribbons used on Selectric typewriters for
quality work.
Also curious if any 1404 printers still exist. They
were wider and you could shift the print mechanism to the side where there was a
mechanisms to print on tabulating cards, one or two at a time.
Speaking of wider: I remember the curious line printers on the Electrologica computer at
TU Eindhoven, where I did my first programming. Those had 144 columns, though I think the
paper was the usual 17 inches wide. I don't have any of those printouts anymore,
unfortunately.
paul
1:46 p.m.
Remember, it is called the 1403 because it was originally built for the 1401 computer.
IIRC there were several iterations of speed before the 360 came out, and the N1 with the
power lid and sound baffling.
from AI:
No, not all IBM 1403 printers had interchangeable type chains:
Note that while the IBM 1403 Models 2 and 7 do offer a feature known as the
Interchangeable Train Cartridge Adapter, which allows the operator to easily remove and
insert a chain cartridge with a different font or character set arrangement, this feature
does not have a separate machine type.
Yes, I had forgotten the name, we had a TN chain we used for one or two jobs.
The IBM 1443 printer could also do 144 columns.
With the 1404 printer, one could install blocks to print on the shortie 51 column cards.
We hated doing that, it did not fit well.
<pre>--Carey</pre>
On 12/21/2024 2:30 PM CST Paul Koning
<paulkoning(a)comcast.net> wrote:
On Dec 21, 2024, at 12:41 PM, CAREY SCHUG via
cctalk <cctalk(a)classiccmp.org> wrote:
my recollection from being a computer operator at the time was the the earlier 1403s did
NOT have an interchangeable chain.
Interesting. I was an occasional operator on a 360 model 44 in 1974, and it had an
interchangeable chain. That installation had several, including an upper/lower case one
(TN perhaps) which I used to print one of my college papers (formatted with RUNOFF on a
PDP-11 system, carried to the 360 on tape and printed there). The printer also had a film
ribbon, basically a 17 inch wide version of the ribbons used on Selectric typewriters for
quality work.
Also curious if any 1404 printers still exist.
They were wider and you could shift the print mechanism to the side where there was a
mechanisms to print on tabulating cards, one or two at a time.
Speaking of wider: I remember the curious line printers on the Electrologica computer at
TU Eindhoven, where I did my first programming. Those had 144 columns, though I think the
paper was the usual 17 inches wide. I don't have any of those printouts anymore,
unfortunately.
paul
1:22 p.m.
On Sat, 2024-12-21 at 11:07 -0600, Jon Elson via cctalk wrote:
On 12/20/24 18:36, Van Snyder via cctalk wrote:
The 1403s at the Computer History Museum definitely have chains, not
trains.
The IBM 1403 printer had interchangeable print
chains. I know of
only
four 1403 printers still working — two at the Computer History
Museum
in Mountain View, CA, one at the IBM Technology Center in
Böblingen,
Germany, and one near Endicott, NY.
All four have the 48-character "A" or "Business" chain, and CHM has
a
16-character numeric chain that allows the printer to run twice as
fast
for numeric-only output. CHM doesn't have an "H" or "Fortran"
chain,
and as far as I know, none of the others do. The difference is that
parentheses are % and "lozenge" — a square with indented edges
— apostrophe is @, and = is # on the "A" chain. IBM also had a 64-
character chain that included box and line drawing graphics. BTW,
nobody seems to know what "lozenge" was meant to represent.
Does anybody know of an existing "H" chain or graphics chain for a
1403?
Van Snyder
My understanding is that the 1403 N1 used a "train" and the
1403 N3 used a chain. I have seen a 1403 N1 train up close,
and there was no chain that connected the type blocks, 2
characters to a block.
Jon
1
days inactive
1
days old
14 comments
9 participants
participants (9)
-
CAREY SCHUG -
Chuck Guzis -
Donald Whittemore -
Jon Elson -
Nigel Johnson Ham -
Paul Berger -
Paul Koning -
Rod Bartlett -
Van Snyder