On 2021-07-30 11:34 p.m., Chuck Guzis via cctalk wrote:
On 7/30/21 6:22 PM, Paul Berger via cctalk wrote:
The MT/ST did pretty good for being a electro-mechanical device,all the
logic was relays in it. I seem to recall many years ago one of the old
OP guys telling me that it write in stripes across the tape. It would
have to be some very simple format because it would be hard to have the
thing sync on headers with only a little relay logic.
Like a motion picture film projector. Brings the tape to a complete
stop for each character and then scans across it with a single head;
going in, the character is read, going out, the character just read is
checked. I assume (but am not sure) that if the check fails, a retry is
attempted. The head moves at 45 ips and records bits at a 45 degree
angle relative to the tape axis. This is so the tape can be scanned
without moving the head for a mark in the control track (reading
parallel to the axis of tape movement) or reading characters with the
tape stopped (reading orthogonal to the tape movement. Obviously,
precise tape positioning is important (even at 20 cpi), hence the
sprocket feed.
In off-list conversations with others, I keep trying to impress on the
younger folks that this is basically an electro-mechanical device with
heavy emphasis on mechanics. After all, the people who serviced these
things were typewriter repair people. I doubt that the innards of the
MT/ST were much more complex than those of the Selectric itself. (One of
these days, I'll get up the nerve to replace the motor drive belt in my
Correcting Selectric III).
But when you've grown up with microprocessors, I guess it can be hard to
envision a world with only rudimentary electronics.
--Chuck
I changed lots of motor belts when I first started as a CE, but not on
OP selectrics but on selectric terminals. The OP selectric used a
relatively weak motor that would stall if the mechanism jammed, but the
terminals used a stronger capacitor start motor that when the mechanism
jammed it would either break the belt or tear the teeth off it at the
motor pulley. Changing the belts on the terminals was even more fun
because of all the added contact blocks to make it function as a terminal.
The ones I spent the most time one the first few years where banking
terminals the control for them was electronic, low density probably on
the scale of early 360 computers, and was very solid so all of the time
was spent fixing mechanical issues with the selectric. It was not helped
by these terminals having a small core buffer so messages would be
buffered by the control unit and then printed on the selectric as fast
as the mechanism would go which contributed to wear. The selectrics on
these terminals where just standard OP selectrics with solenoids
(magnets) and contacts hung on them to make them work as an I/O. I
suspect the MT/ST had a similar typer on them and I feel for the guys
who had to maintain them because I am sure MT/ST customers when a lot
more picky about print quality that terminal customers, and then there
was the composer version of the MT/ST, I am told that composer customer
where super picky about print quality.
It seems to me that the mag card machines had a version of selectric I/O
that was designed as an I/O unit with key parts of the mechanism beefed
up to improve reliability. Some of the selectric terminals I worked on
had a similar mechanism that separated the keyboard from the printer, it
was still a standard selectric keyboard but the printer did not take a
cycle when a key was pressed. The open transfer contacts where
replaced by reed switches and magnets so they did not get fouled by oil
and grease like the open contacts.
Since there was still a few 360s around when I started I also got to see
the inside of a 1052 a few times, they are a really stripped down
keyboardless selectric. They used a function cam to space and since
they did not have a tab rack they would space a lot which would cause
the space cam to wear, I remember one that was so worn that when it
cycled it wobbled very noticeably, the customer would not let us replace
it as this was the console for the 360 and they did not want it
unavailable for the time it would take to replace it. Some customers
apparently would have a spare 1052 onsite. The keyboard on the 1052 is
the keyboard from a keypunch machine.
So yes even though I started in 1979 there was still a requirement for
having good mechanical skill as well as knowing electronics to fix DP
equipment, even for the guys looking after mainframes and associated
I/O. Card equipment was still common and the most common printer was
1403 with it hydraulic paper feed. 3890 cheque sorters arrived around
the same time and the mechanics of them was pretty maintenance intensive.
Paul.
