Gregory's ServerThe Shuttle printer uses a spinning metal drum with raised characters. 80 hammers, one for each column, fire at the exact moments to hit the ribbon and paper against a character on the drum as it spins by. 2/12
|
32 comments
 The teleprinter design thrown together in just 7 months after a delay in the TDRS satellites meant that the fancier digital printer wouldn't work for the first few flights. Although the Interim Teleprinter was expected to be used for a short time, it remained in operation for over 50 flights, acting as a backup printer. 4/12 The teleprinter was based on a military communications terminal, with many modifications. The keyboard was removed and boards were added to interface with the Shuttle's audio system. The system still contained a word processor, unusable without the keyboard. 5/12 This view inside the teleprinter shows the three custom Shuttle boards (left), the power supply (blue), and the four logic boards (right). I reverse-engineered the three custom Shuttle boards to determine the signal format (FSK with 3600 Hz and 7200 Hz). The Shuttle teleprinter kept the four logic boards from the military teleprinter: the CPU board, print control board, communication board, and memory board. These implemented a 6800-based microprocessor. 8/12  The teleprinter was mounted in a storage locker in the Shuttle middeck, one level below the flight deck. 9/12 We managed to get the printer operational. This wasn't easy because the rubber rollers had turned to liquid, gumming up the mechanism. CuriousMarc carefully disassembled the printer, cleaned all the parts, and realigned the hammers. 10/12 For details (including schematics), see my blog post https://www.righto.com/2024/08/space-shuttle-interim-teleprinter.html Credits: printer restoration done with CuriousMarc, @tubetime, and Mike Stewart. Printer provided by Marcel. Shuttle photos from catalog.archives.gov. Locker photo from DMolybdenum, drawing from the military teleprinter manual. @kenshirriff @tubetime that is incredibly cool. Itβs a shame the YouTube video wasnβt longer showing the reverse engineering effort undertaken. I love all your work as itβs quite extraordinary what was built so quickly with such engineering skills. @plambrechtsen @tubetime CuriousMarc will make a video on the restoration at some point. @kenshirriff @tubetime your AGC rebuild series is still one of my favourite on YouTube. @kenshirriff @kenshirriff @tubetime Wonderful restoration! Tangentially related, I read that the optional copier for the DEC VT52 used electrolysis to print onto damp paper using a helical electrode on a spinning drum. I found drawings and docs, but as far as I can tell, there are no working units that have been demonstrated on video anywhere. Would be really cool to see that in action! Very few terminals seem to have βadd waterβ as a regular maintenance item π   @glowl What could go wrong with bees on a Space Shuttle? Appropriately, the experiment was sponsored by HONEYwell. Photo from https://airandspace.si.edu/collection-objects/bee-enclosure-module-sts-41-c/nasm_A19880031000 @kenshirriff i remembered there being something real about that, i assume the "bees" in the picture i initially answered to is an acronym for something else?  @kenshirriff these old designs are amazing, real time data processing with a hundred hardware actuators all on a ~1 MHz 8-bit CPU, 4K ROM and 4K RAM. These days you can buy watches with several GHz class CPUs and a couple of gigabytes of RAM. I love auto-correlation functions / functionality and used that for certain tasks in the past, too.  @kenshirriff Interesting that they deliberately went for the delay-line approach with harmonic FSK tones. All other contemporary systems (e.g. Bell 202 and V.23) used non-harmonic tones (in order to avoid bit errors due to nonlinear distortion on the line).  @kenshirriff Worked on that comm terminal as engineer for US Army Signal Corps in 1969. Was used in Vietnam War. May still have the manual I proofread way back when. Interesting it found another life π You're confusing some of us now because C.Itoh made actual "shuttle (line-)printers" Their "shuttle printer" was a matrix printer, but instead of a head which traveled the full length of the line, they had a head with 132 pins which just vibrated the width of one single character each. Huge beasts, very fast (300 and 600 LPM as I recall) and you could hear the hum from the vibrating head two floors down when it ran...  @kenshirriff With the characters skewed to reduce the chances of wanting to hit all 80 hammers at once (eg if printing a row of * or -). But didn't it still need big capacitors just in case?  @kenshirriff holy crap, we flew a drum printer into orbit. I bet the noise was fun. Thank you for these threads, they're always fascinating.  @oddhack The military teleprinter supported both ASCII and Baudot, but the Shuttle teleprinter was wired for ASCII. @kenshirriff @kenshirriff Oh, memories of the 1980s line printer we had at the university! (which had 132 columns) |
Here's a closeup of the hammers in action as the Shuttle teleprinter prints a line. 3/12
https://youtu.be/1SjtmePBZjo