Some of us beard folks, get tired of dealing with a beard. It gets itchy, all the maintenance, and sometimes you don't want to deal with it, or just want to regrow it. Some just grow one for a bit, and then grow it out and shave, just because...
Just as a precaution hint: I don't know how fast this chain spins, but I've heard from IBM technicians that they once had a case of "chain embedded in wall of datacenter" - don't spin it up to full speed without the case around it I'd say; or at least be well out of the "trajectory" :)
Yeah I've heard the same back then, when a chain breaks it's spectacular. All that metal is there for a reason. It's like a motorcycle chain at full speed.
Plastic clips on a rubber belt. The most used letters would break all the time. This was only 100 lines per minute printer. Slow... Wait until you had a 1000 LPM printer throw a drum belt. That was always fun.
I worked for a mainframe (not IBM) company from 1974 till 2020. We used Odec printers (specifically the one you have) on what we at the time called mini computers. I can reach out to some of the Field Engineers I used to work with to see of they might have some of the documentation for this printer. -Fitz
I'm well into my 60's now and so appreciate those of you who preserve vintage items. I love vintage cars and mid-century modern furniture as well as computers. I harken back to the day when we used to have to reserve time to use a college computer and I first learned Fortran. I was on the cutting edge of college term papers when I first used a Commodore 64 to write my papers - some professors thought that might be cheating! I had a color thermal transfer printers so my papers stood out from the crowd where everyone else was typing - I was using Geos to do word processing and the papers were first rate and highly professional well before others caught on. I so appreciate you and for preserving the giant computers of the era and allowing today's youth to see what came before their bloody Fruit computer.
I am on the edge of 60 myself. we have seen alot of things come and go with each decade. things were made so much better, and built to last. I have a 1950's fridge and stove that are still purring along in 2024. tho I really miss the 70's. as a young teen that was a magical time for me. we were so free and truly cared for each other back then. [even strangers]
Thank you so much! Also, that's awesome that you were using color thermal paper for your work. That's really wild, I don't think I've ever come across a color thermal printer. I love the technology we saw coming out of the 50s, 60s and 70s and it must have been wild to watch computers change so dramatically during that relatively short time. To go from vacuum tubes to the 6502 in just 25 years must have been wild!
Back in the day I did some programming on an IBM 360 mainframe. The printouts came out of an IBM 1403 chain printer that was 1,000LPM. Chain printers are fascinating bits of 70's era technology. The biggest problem you may face with getting that 125LPM printer working properly is adjusting the timing of when the hammers fire that hit the moving chain. This is going to be an interesting journey to watch.
I used to program and sometimes operate IBM 360s and those printers were beasts. The old card readers sounded like a wood chipper. Computer rooms were really loud back then!
Oh man, I can't even imagine how insanely fast that 1,000 LPM printer must have been. The simple 125LPM I have here does essentially two rows per second, but the 1,000 LPM is eight times faster than that!
@@UsagiElectric We had printers that could print as fast as most line printers could do a page eject. Several pages/second. You had to get the paper in just right and there were a bunch of grounding wires to prevent static buildup from the paper shooting by or you would get the biggest paper jam you've ever seen.
@@ChristopherHaileyYou're right about getting the paper in right. But also all the wire mesh that deflected the paper arc into the collection bin. I once saw the operator restart the line printer after adding paper and the deflection cage wasn't locked in place. The stream of paper displaced a ceiling tile and enough paper was on top of the ceiling to break another tile and come down all over the place.
In college, many years ago, we had a DEC printer of similar technology. It had a pod on each side that held the biggest caps I'd ever seen. It gave it a shape that got it named "ED 209" in short order. It had a leathal, sharp metal ribbon that held the embossed letters and spun up like a bandsaw. It wasn't long before we figured out the order of the letters on the band and wrote a program to print a string that would fire all the hammers at once. It was deafening and the lights would flicker as little ED would walk across the room. Highly recommended. Alert the local sub-station first, though.
Those carbon composition resistors tend to drift upward in value with age so when you settle in to start going over the hammer driver boards in detail, you may want to go down the line with an ohmmeter and see how the measured readings compared to the marked value and tolerance.
All electrolytic capacitors leak. As for forming, once you get the voltage applied where you want it, let it sit overnight. That should reduce the leakage to the minimum you're going to get. And you don't need to put a lot of voltage across the capacitor. For a 25Vdc, i would slowly raise it to around 20V and let it sit. 60Vdc, i'd probably go to 40V and let it sit an hour, then 50V, let it sit, then 55V and let it run overnight. Those older electrolytes tend to be very robust. Modern electrolytes don't take sitting around without a charge on them as well, in my experience.
I've gotten lots of old electrolytic caps like these from the local hamfests. Most of them were perfectly fine after reforming overnight. As long as they haven't dried out or been damaged by applying full voltage without current limiting, they will usually be good.
Ahh, the good old days when you would sit in the terminal room, submit your job, and patiently wait for the line printer to print the result. Or to hear the characteristic sound of a full register dump indicating that your program had crashed.
for some of us the printer was 80 miles away, got result next day (why didnt we invest in a monitor ?) only to find one had input the wrong decimal point and the graph and result was nonsense
In my "good old days" you would hand your deck of cards through the window at the computer lab and come back a day later to get your print out. Just to find that a misplaced comma crashed your job.
i worked assembly for Documation in late 70s and built the impact 3800 line printer. That's 3800 lpm. Absolutely insane when they were under test. Yea, they were finicky.
30V is common mechanism power suppply (even for later pin printers), +10 may be for positive RS232 interface (something low, but acceptable), -12V is definitely for RS232 interface (I assume this is serial I/F printer as many of this time), 5V of course is power for legion of TTL chips. Edit: try locate interface chips (MC1488, MC1489) were produced from 1973, most likely are also fitted in this printer. MC1488, driver: pin 1 -VCC, pin 14 +VCC, pin 7 GND. this should help in locating power source from PSU. Datasheets are also available from Texas Intruments site. Good luck.
Yes the 10V rail should be around 12V, because that is the voltage required for true RS232 voltage levels, though it will probably be fine with 10V anyway. The capacitors really should be tested differently, or at least connected to a power supply via a resistor, and left for 24 hours to reform. 50 year old electrolytics are very much at the end of life, though you will find that some are still within spec after a 24 hour period at rated voltage via a 1k resistor, just in case they decide to short out, so the puff of smoke will not be much, though the resistor should be 5W to survive. 24 hours of rated voltage, then leave open circuit, and measure with the multimeter every hour, and before disconnecting check voltage drop across that resistor is under 1V, saying leakage is under 1mA, which is a rough limit for the leakage acceptable. I have tested some NOS electrolytics, and of the dozen I had 6 were physically leaking when unwrapped, so got tossed immediately, 3 were not dropping down in current even after 24 hours of reforming at 5V, and were also tossed, and the remaining 3 had 2 with a rotted vent, so also failed. The sole good one held charge for over 3 months, checked once a week or so. So the can sizes you have are all still available off current suppliers, and you really should replace, just go for ones in the same diameter can, or slightly smaller, and use a sleeve, common on replacements as well, and use similar capacity and voltage, though you can have a slightly higher voltage, and capacitance within 20% of the nominal is still fine, as those large cans have a tolerance on capacitance of +100%/-80% as still being fine. Yes that 60 000 uF one, used likely on the 30V rail, is perfectly fine, even if the capacitance is anything from 12 000uF to 120 000uF, so anything in that range and voltage is suitable as a replacement. Same for the others as well, though going close to the nominal is better. 85C standard electrolytic, computer grade, none of the low ESR units, your rectifier diodes will thank you. Those large cans make a loud bang when they fail, and can take out the diodes and transformer as well. Though with that power supply, and those power resistors, I would turn the board over, wash it with a new toothbrush and some ethanol, and then apply some liquid rosin flux all over the board, and resolder every solder joint associated with the power diodes, the power transistors and the power resistors, and clean the board again with alcohol to remove the rosin. Will likely fix any dry joints there, and also solve future issues with the power supply. Clean that edge connector, deoxit on both sides, and plug in and out around 5 times, clean again, and apply a last coat. Same for the TTL boards and drivers, and same for any IC in a socket, clean the legs, then plug in again, pull out, clean pin and socket, and put in again. will solve any poor contact issues before you have them, leaving 1000 less problems to solve. Before powering up the printer, get some 24V turn signal lamps, and some 12V ones, and apply the 24V lamps to the 30V supply and the 12V supplies, and the 12V one to the 5V supply, to give the PSU a small load. Off the 30V rail you will draw roughly 2A, the 12V rails around 1A, and the 5v rail same, well below what they likely are rated for, at least for a 5 minute test. Then use your multimeter to measure the voltage out, should be within 10% of what you measured no load. Then take your scope and look for ripple at 60 and 120 Hz, on the output, with AC coupling. Should be under 100mV, and on the capacitors should be under 1V peak to peak of ripple, and at 120Hz, if there is a 60 Hz component you have a diode that is open, or a bad joint, and the 2 peaks for the 120Hz should be roughly the same height, if not you have a failing diode, either nearly open circuit or leaking badly, thermal camera will tell you if a separate diode that it is running at toasty level, or a bridge rectifier block running at 50C above ambient, which under low load should not be the case.
@@kromaine13 No difference in the power supply though, and the 12V rails likely are also used in the hammer drive section to provide bias voltages for the driver transistors, so they turn on and off fast to drive the hammers.
@@kromaine13If there's no RS-232, I wonder what it uses the negative rail for? Centronics is 0 to +5v, iirc, and so are all the TTL, of course. The hammer drivers are probably a simple open collector driver, so no need for a negative rail there, that I can see. Maybe some early NMOS RAM used for a line buffer?
Thanks for the printer tour and welcome to the new year. When you think about it, printing a line of 132 characters every 480 mS pretty much maxes out the capability of 7400 series TTL on boards of the size in this printer. Gate delays are on the order of 10 to 20 nS, thus the maximum rate for a system clock is probably between 5 and 10 MHz. The chain/band probably has the character set repeated around six times. You’ve got around 2 mS or so for each hammer to fire over the correct character before the rack moves in to the position to print the adjoining position. That makes this machine pretty much state of the art for its time.
@@highpath4776 The chain or band speed is constant. I do not know this printer, so I don’t know whether it is chain or band. The two types are materially the same in operation. The difference is the band has embossed characters and a chain type has interlocking cast type slugs. Both move the type at constant velocity and fire the hammer when the hammer is over the desired character. Repeated character set allow lower velocity. The advantage of band printers is that end users can change the band when it wears or to change character sets. This printer economizes by having each hammer service two adjacent positions by jogging the rack. Printers like the IBM 1403 had a full 132 hammers and could print 600 LPM. The type chain of a 1403 moves at scary velocity, hence the other comment about a chain stuck in a wall is believable. That is pretty remarkable, considering the 1403 dates to 1959 and was part of the transistor based 1401 system, no ICs. The 1403 saw duty on other systems in laters years up through at least the 1970s and it would not surprised me if some 1403s are still in operation.
I used to have to fix line printers. I can't remember half of it 😅. I do remember a lot of issues were simply caused by the mechanism getting dust from the paper in it.
@@highpath4776 you would think that would be true but I found it in the weirdest places. One of the worst dot matrix (not the same I know) that I ever worked on was in a jewelry store. This thing had a noise shroud and was in a drop feed cabinet. It needed to be completely disassembled because hair caused just enough of a clearance issue that the paper would jam nearly every time. I think it matters how the paper is stored and who changes it.
@@lindoran I used kitchen gloves, one size small. Yup, normal gloves have no chance of survival. Ah, the stuff i used to find inside printers... :)) Worst job i had to pull was chewing gum inside a big Canon printer. Had to clean it, because it was an expensive unit, not like a cheapo these days where you can write it off. That was fun.
18:18 That looks suspiciously like the form control tape - the ICL 1932 drum printers that I used in the 70s used them so that form feed and other form controls worked for whatever stationery was required (remember FORTRAN4 H1 and H2 print controls? the 8 tracks of that loop were what it referenced and H1 was by convention the top of form). We had something like 20 loops for the various stationery types that we used and they were made of a mylar reinforced paper tape for durability.
Not this unit, but have some experience with another manufacturer's 'band line' type printer. The 'belt' as you call it on mine was literally a 'band' of metal with small metal fingers cut in one side with each letter so that the hammers 'flexed' the fingers against the ribbon/ paper. Anytime we changed the band out for another one with different sets of characters, we also changed that paper tape that 'told' the printer what band was installed. I saw you have two paper tapes, so.... make of that what you will.
@tradde11You know, now that you mentioned this, it DOES ring a bell. Something about header space and vertical 'tab'?!? Maybe it was when we switched from normal 'green bar' paper to shorter 'letter' sized tractor feed. Sorry, it's been a long time... :) P.S. Yeah, i said 'paper tape' meaning that type of punched tape, but you're right it wasn't physically 'paper'.
@tradde11Yeah, the one we had, if you had a mixed case band (both upper and lower case) it would do 300 lpm. But if you had an upper-case only, it had all the letters on the band twice, directly opposite and it would print 600 lpm. Paper practically SHOT through that thing. lol
electrolytics dont have totally zero current even when fully charged, its normal .. especially with very large value ones ...charge to full working voltage via current limiting, hold for half a hour or so, discharge via a bulb or power resistor load, recharge, then discharge again, then test for capacity and esr if you can ...
The memory on that is maxed out. 256KiB is the most you can address with an 18-bit address space. LSI was actually invented in the late 1960s at Rockwell. They developed a four-chip solution which replaced over two dozen chips for a portable calculator produced by Sharp which was introduced in 1969.
Right. By the start of 1975 LSI was widely available and we were into the second generation of 8-bit microprocessors, with both the 8080 and the 6800 having come out the previous year, the Altair on the cover of _Popular Electronics,_ and the 6502 less than a year away.
When I was a senior in high school in 1994 I got a job at a local advertising company and did nightly tape backups of their archaic system. The printer used was over twice the size of that one and I had to print the payroll checks and lots of daily updates on it.
When I was working in a Lab at my university almost 20 years ago there was a 1mb hard drive being used as an occasional doorstop. The thing was probably the size of a concrete block and weighed 25 lbs or more. Had to be from the 80s. I always wanted to build a wiring harness and see what was on it - even now 15 years later I still sometimes wonder what was on it. Missed opportunities. Strange things we choose to remember
I worked with similar high capacity printers in a large air conditioned ICL Series 39 mainframe computer room for a while in the early 1990s, in fact they were even bigger and were very old even then, but many such commercial printers were built to last and with preventative maintenance could survive many years of heavy punishment.
The fun of the Data-100 is the complete Hammer Bank assembly shifts to the left then prints the odd positions then shifts to the right to print all the even positions on the line being printed, the paper moves up one line. The fun is the timing of the print-belt spinning across the front of the paper and the Hammer Bank shifting Left - Right - Left .... 250 times per minute for a line speed of 125 Line Pwr Minutes ( 125 LPM ]. FYI - The power OFF position of the Hammer Bank is in the center, not Left or Right. The motor & Cam Link drives the Hammer Bank when the printer is printing, it loud...!
When I saw the address of the manufacturer of this printer, I had to Google it. As a strange matter of fact, I've actually been inside that exact building at one point! Though it was many years after this printer would have been made and the building had been repurposed by a different owner by that point. It's definitely a small world!
The location of the tape loops suggests to me that it was used to control the platen for preprinted custom forms, the default tape loop would just skip over the perforations in the fan fold paper,
Its cool to see alot of older retro tech that normally isnt on other channels. Alot of this stuff takes up a ton of space and would be hard for alot of people to fit into their collection.
When it comes time to clean up the character chain, brakeclean in a container that won't melt from it or high-proof alcohol are going to be your friends with a toothbrush to really get into the nooks and crannies of every character, follow up with a brass wire wheel on a cordless drill to get them to shine... Adding WD-40 to the ribbon, wherver it goes, should add enough detergents to the ribbon again to produce ink, and be mindful of what lubricants are used where, some lubricants won't like plastic while others aren't made for high speed applications.
This is fascinating. I worked on drum based printers in the late 70s (adjusting hammer flight time, etc.), and my first printer at home (in 1979, no less) was an ICL Termiprinter, which was slightly similar in operation to this chain printer. In the Termiprinter, the typeface pieces were at the end of steel fingers extending from the side of a belt (kind of equivalent to the chain here), a bit like a giant label embossing device. It was only 80 columns with 80 hammers, and also quite slow, but it still weighed a ton. The ingenuity in such machines never ceases to amaze me.
Oh wow. I wish you were local to me. Since the 80s, I've serviced a lot of printers, ranging from impact beasts like that one (although not that specific make) through to modern lasers & inkjets.Getting yours up & running would be a fun project.
For the mechanical side of the printer, in case you are wondering: the blue tape with the perforations that you removed at 17:58, is a "carriage tape". It allows a program to skip to a certain vertical position on the page. For example if the program wants to jump to the bottom of the page to print a footer, it tells the printer to skip lines till it encounters a perforation in the column (in IBM terms, that would be called a "channel") that corresponds to the "end of the page" (if I remember correctly, in the IBM days that would have been channel 12, by convention). When I started working in the early eighties, we had an IBM 3203 printer that printed 600LPM (there existed also a 1200LPM model, but we didn't have that one). It was a beast. If you opened the cover, even if it was not printing, the sound was like a steam locomotive entering the train station. Very impressive.
"We're throwing science at the wall here and seeing what sticks." Excellent work on that power supply! Waiting for more on the printer - The perforated tape reminds me of CuriousMarc's IBM printer. I think it's mechanically coupled with the character being printed, the combination may be an ASCII code. Then it's just rotating the shaft until a desired character is found... 8:258:38 hope there was no asbestos in it, haha! 15:03 [FANSPIN INTENSIFIES]
The paper tape loop is vertical forms [paper] movement. One paper tape channel is for TOF [top of forms] hole & one channel for Vertical Tab. The paper tape loop is 66 positions long (or the number of lines per page). Being able for the programmer to tell the printer to move the paper to TOF or do a Vertical-Tap was easier than keeping track of line counts.
@@kromaine13 While easier than counting lines, the real reason is the printer could "tab" to the spot much faster than "stepping" to that spot via CR/LF. It's like the difference between the tab key and the space bar on a typewriter. Those engineers used a lot of tricks to get more speed out of their machines.
ahhh band printers.. sound like a band saw and a crate of wood peckers.. had one that a friend of mine used to print books with in the early 80's interfaced with a trs 80 along with a stack of floppy drives.. madness
As kids, that paper made the best glider airplanes. Our dad would take us to work with him on the weekends. We would throw them down the massive atrium entry ways from the high balconies, and they would go forever!
I was kinda amused by the way he removed the paper. I wonder if he was concerned bout the clamps. IIRC, just slide the little middle bit down and it will come open.
The ole System 3 we had at SYSTIME had a 1403 attached. The 1403 is the same printer shown in Dr Strangelove where a pocket radio was kept. The weird thing about the print chain we used is the fact that the numbers were all turned sideways in order to print serial labels. Also fixed the program which required supervision and fiddling. Programmer made a math error. Ran perfect thereafter. Nice timing belts, btw. EDIT: enjoy purchasing new ink rollers. Heh. 132 wide, huh. Ouch
Used to work on Dataprinter corp. line printers. 600 &1200 lpm. The 1200 had a vacuum cleaner built into the cabinet. Major dust creator. Tricky timing circuits.
Word of warning -- make sure the form (or carriage) control tape is in place before pressing top of form. If the form control tape is missing or not reading properly, the paper will fly out of the printer in an attempt to find "top of form". After several seconds of flying paper the printer will fault and stop advancing the paper. Ask me how I know?.... As a junior developer, I was responsible for - among other things - creating the carriage-control tapes for the various tax forms used by several financial companies. So I would create the tapes and then test the forms -- most print jobs would have about 10 or so "line up" forms at the front of the print job so the operators could align the output. Once aligned they would allow the print to resume. But occasionally, the tape would be wrong or break or the form control reader would clog .. and you got to watch a "forms run-away" ... I was working with 3 CDC 512 printers - 1403 equivalent -- and these would regularly spit the paper to the ceiling. Additionally, there is a switch that will sense if the paper has run out -- and will stop the printer from advancing -- if this gets jammed open (or misaligned)-- you may believe other issue are causing the problem -- check this first -- it' s sorta the same thing is asking "is there gas in the tank?" -- again - ask me how I know. Column 1 of a report was associated with the carriage control -- so a 1 in column 1 of the report would advance to the punched hole in the carriage control tape at column 1 typically called the "top of form". A 2 in column 1 would advance the form to hole in the carriage control tape in column 2 -- and so on. Hence the reason for 133 character report that only printed 132 characters wide ... In really complex forms there could be several different "vertical tabs" positions. Also the length of the control tape has to match that of the form - exactly!. If you got the length wrong -- the form would misalign on every "top of form" of "form-feed". -- again - ask me how I know. -- stories of a junior developer learning the trade...
The "Odious Odec": watch those edge connectors, they tend to be fragile. Expect to find a buffer containing a line of text, with (if memory serves) a pattern clocked over it synced to the chain speed: when they matched a hammer fired.
This printer reminds me of the 6500 IBM Printer series... it adopt basically the same configuration for the hammers and the characters belt. I was, in my youth, a good technichan on those printers and the adjacent series 6400 ( this one had a single line of heads in a vibrating bar that compose the characters dot by dot in lines) Both are printronics rebranded if i remember well an probably i still have the field manuals in my garage, after 40 years i have to do a bit of scavenging to find out but i can searche them. Mechanically are quite serviceables but the eletronic was a pain.
Once upon a very long time ago, when dinosaurs roamed the data-centres and I was but a pimply faced youth (PFY) We got a very 2nd hand drum printer a DEC branded one I think... The ribbon was a massive roll that ran over the face of the drum (think the old cloth hand towel things you used to see) My job was to clean this beast and the drum... Armed with a large bottle of IPA a few skewers and a tooth brush, the job was to "unclog" the type faces on the drum. One well spent afternoon and doing a great impression of a smurf once I was done.... the printer worked a treat and the output was mostly legible :)
To reform capacitors, just set the current limit to "a few mA" (I just found out today my powersupply has a 5mA offset! When OFF, you get 5mA. That's a suprise! When set to 30mA you get 36! and the display shows 30!). Just set the voltage at the max voltage and turn it on. It should be ok after just one cycle. If you want to check for leakage, unplug the powersupply from the capacitor and watch the multimeter. Measure the time it takes for a certain voltage drop, say 5V. You then get I = C dv/dt, you know/measured everything on the right side! . Oh. In the video, you start saying there remains some current while in fact my gut feeling is saying: you're deciding that too fast. I'd find it totally normal if it stays at 1mA for say 10 to 60 seconds before dropping to zero. I "salvaged" a PDP 8 in the late eighties. I scrapped it in 1997. :-( I still have some of the capcitors (160000 uF, 25V). And its two fans. The fans still get regular use. The capacitors are marked that I think they are early seventies.
The Fairchild part in the video at 5:34 looks like it is marked “96602PC.” There is a 96S02PC, which is a dual retriggerable one-shot. The RC network next to the chip would make sense. It may be socketed because it was hand selected for a fast enough part. The empty socket next door may be for test probing - hope so. Of course, the manufacturer’s service book would reveal all. Hopefully someone out there has a copy. The service book is probably essential for avoiding disaster working on a machine with parts moving at the speed used in this printer.
I would check the rripple voltage of the caps before progressing to power logic boards. As a general rule for me, I replace old caps like that on equipment I am going to keep. It might get expensive, but worth it.
"Ok, so I bought myself a printer. Now I just need a front loader to move it around" Not new to the channel or for that matter to the Centurion, but I think even if you made a video working on a dishwasher I'd appreciate it. Your enthusiasm is contagious! Great content over here! 👍
Alright, you've hit on an area of computer repair that is a personal nightmare for me. When the printer was a much more commonly serviced device, and they were much larger and more mechanical beasts, I'd break into a sweat every time I got a service call for one. I'm not at all mechanically inclined, although I do wish I were. But I know this is going to be a fascinating trip, and I wish you the best of luck!
I ended up,with one of those in my living room back in the 80’s. I had to leave it behind when I was forced to move out. Hope the new owners had a use for it.
The fan needs a rebuild. That's a looong long time to spool up to top speed. Take it apart and scrub the bearings clean. Depending on how loose things are use thicker or thinner oil (thicker oil also accounts for the fan getting hot during work, so think about that aspect as well).
I did output collection and sorting from a band printer in the late eighties. Never saw a chain printer running. But if it is 100 lpm (not cpm) I'd imagine it's near the same. It was amazing watching that band printer run.... chomp chomp chomp grrrr grrrrrr chomp chomp chomp and ten 13" fanfold pages had printed.
Hope those old Sangamo capacitors are still in good shape - that was a company based in Illinois and which also made electric meters (as in on the back of the house) in addition to other products...
I worked as a computer operator in the '70s and worked with a lot of line printers and had to deal with a lot of issues but I have to admit I never had to deal with dirt daubers!
That green and white wide paper brings back memories of huge printouts that were in huge binders. We used to have to dig through them to find what we needed. No internet in those days.
Kind of curious that the latest model computer (the Centurion) has less maximum memory than the models about five years older. But again those machines were programmed tightly. They would have scoffed at our modern bloatware as their systems always got right down to brass tacks.
After all the grief that the Centurion has given you, I can't wait for part 2 to hopefully be "I greased and oiled everything, plugged it in, turned it on, plugged it in to the Centurion, and printed stuff!" You definitely deserve a project to go easy for once!
In the early eighties I used to work in an office where the had a Data100 (I noticed that brand name on the back of the centurion printer...) drum printer. When that was doing a test run, man that was hammering. I also have some chain printers in bad shape, not by centurion. Waiting to be restored for 40 years now I guess:). Anyway, if you look for documentation, 'DATA100' is a good link.
It's always a good day when old IC or transitior computers like this are shown off. The Centurion may as well be a character alongside Usagi the rabbit, they're synonymous with your channel at this point lol.
Worked for Data 100 from 76 to 85. Used this on a lot of RJE's. This was the smallest line printer we used. Good printer but they wore out quickly. Too many moving parts...
@anthonydenn4345 Yes, it's messy but not from oil. Between the paper dust and printer ink ribbon dust, it could be tracked all over the office. Almost lived with a vacuum cleaner nearby just to open it up.
The desk system at college, you had a connector wire to the drive issue with no write happening, was the cable fixed and the software replaced and now working as a CPU 6 machine
I know that somehow that printer is going to find itself incredibly useful. That Centurion alrwdy seems to see its fair share of use as a diagnostics workstation.
Back in the 80's I worked as a computer repair technician. Computers and drives I had no problems with but printers, boy. Almost every one I worked on I ended up with a small bag of parts left over after putting them back together but they were working so were the parts really necessary or was my boss messing with me by slipping them in when I wasn't looking. We'll never know.
That is really neat stuff, again! When I was a tech, I loved working on the old TTL logic boards. Working on those boards really makes you think hard about how things work because it's not a matter of just pulling a single chip to fix the problem. I learned more about circuits working on those than I did when I took my engineering classes in school. That power supply looked a bit scary to work on. I was never keen on working on power supplies after getting a good zap from a CRT circuit. I got teased a lot by my coworkers because of it.
These printers are SO COOL. A copier tech friend of mine told me about a call that he got on one of these. They had it serviced the day before, came in the next day and the paper was running backwards. Turns out that the paper feed motor was 3-phase. Two of the leads had gotten swapped, reversing the motor. Ooops!
Praying you did not get a Bang‼. I try to put on glasses when I fire unknown hardware. Thankfully she powered up quietly. Great old stuff. This is the first chain LP I have seen where ours were drum.
My company still prints thousands of pages monthly on green bar paper. Our daily use printer is from this millennium and has Ethernet but it's interesting to see how far back this format goes
Slowly catching up on your videos really great channel, I enjoy watching you fixing those old treasures! I think you should get an LCR meter to check the caps properly. I think the DE-5000 is quite popular. It can not only check the capacity but also the ESR value of the cap.
Hah I pulled a hefty cap out of an early 80s Xerox machine as a kid in the 90s - the cap may even be older than me. It's in my kid's custom power wheels acting as a large power smoother.
Maybe it's the Beautiful Flowing Hair. (Maybe it's Maybelline) But I Swear until you put your Hand on the Centurion Terminal, It Looked like you Were Superimposed in front of a Green Screen LOL
A very nice. Learned something new I think. Paper tape reader for programming. Not surprised the psu came up. I don't have any experience with chain printers. But, the old dot matrix printers I did just needed a good cleaning and lube. One tip is to use paint marks or what not for alignment. Also, unless you find a service manual. I would try not to take it 100% apart unless needed. Luckily you can move most parts of the printer by hand. The only down side of this is you pointed out. Printers are a whole new experience. Also good chance those logic boards will probably just come up to. In fact the only thing I'm worried about is any hardened lube. Turn it on and it throws a part.
Reminds me of the HP2619 printers I worked on back in the early 80s. Hated them. Getting the print chain and hammer bank set for even non smeared printing was almost impossible. And with a swinging hammer bank -GOOD LUCK!. (Reminisce of the 2613) Watch the clearance or you will tear out sections of the print chain. Like the fact they put fuses on each driver at least if a hammer shorts you should only blow the fuse. That little paper tape on the side controls the paper feed. One punch was end of page another top of form. The others could be used to advance the form to specific lines on the page via control in the programs.
I live in NY I had no clue what a dirt dauber was till vice grip garage then that summer i had 3 of them building nest and now i see them all the time LOL
So many comments already! Hope you see this one - a few possibilities for further research. 05:55 That sticker bears Data 100 Corporation markings. Data 100 was a manufacturer of IBM-compatible peripherals (spun off from CDC in 1968, bought out by Northern Telecom in 1978, although the brand name stayed alive until the early '80s). As far as I know, Data 100 was the original manufacturer only of the RJE and direct-data-entry terminals. The company rebranded and resold card readers, paper tape readers, and drum printers from Dataproducts; card punches from Groupe Bull; chain train printers from CDC; and tape and disk drives from MPI. But I'm not familiar with any CDC train printer that was that slow. The ones I worked on were 600, 900, or 1200 lines/minute. It looks as if the printer is trying to be a clone of the IBM 5203. That's in the right period, similar form factor, similar speed, and since the Data 100 stuff was all supposed to be IBM compatible, I bet the beast was made to interface to an IBM 2821 controller. It's clearly NOT a 5203, though. It has two print positions served by each hammer. The 5203 had one hammer for every four print positions. The manuals for the IBM gear are still around, which would at least tell you what sort of thing that Data 100 was trying to clone.
where beard
Asking the important questions right here!
@UsagiElectric were beard :C
@@UsagiElectric You look a lot younger without it. :)
Some of us beard folks, get tired of dealing with a beard. It gets itchy, all the maintenance, and sometimes you don't want to deal with it, or just want to regrow it. Some just grow one for a bit, and then grow it out and shave, just because...
January must be a "no beard" month
Just as a precaution hint: I don't know how fast this chain spins, but I've heard from IBM technicians that they once had a case of "chain embedded in wall of datacenter" - don't spin it up to full speed without the case around it I'd say; or at least be well out of the "trajectory" :)
Nah that's a wives tale. That would not be possible.
Yeah I've heard the same back then, when a chain breaks it's spectacular. All that metal is there for a reason. It's like a motorcycle chain at full speed.
Plastic clips on a rubber belt. The most used letters would break all the time. This was only 100 lines per minute printer. Slow... Wait until you had a 1000 LPM printer throw a drum belt. That was always fun.
I worked for a mainframe (not IBM) company from 1974 till 2020. We used Odec printers (specifically the one you have) on what we at the time called mini computers. I can reach out to some of the Field Engineers I used to work with to see of they might have some of the documentation for this printer. -Fitz
I'm well into my 60's now and so appreciate those of you who preserve vintage items. I love vintage cars and mid-century modern furniture as well as computers. I harken back to the day when we used to have to reserve time to use a college computer and I first learned Fortran. I was on the cutting edge of college term papers when I first used a Commodore 64 to write my papers - some professors thought that might be cheating! I had a color thermal transfer printers so my papers stood out from the crowd where everyone else was typing - I was using Geos to do word processing and the papers were first rate and highly professional well before others caught on. I so appreciate you and for preserving the giant computers of the era and allowing today's youth to see what came before their bloody Fruit computer.
As of Jan 1, SEVENTIES! Woohoo.
I am on the edge of 60 myself. we have seen alot of things come and go with each decade. things were made so much better, and built to last. I have a 1950's fridge and stove that are still purring along in 2024. tho I really miss the 70's. as a young teen that was a magical time for me. we were so free and truly cared for each other back then. [even strangers]
For a second I thought "are Rasberry Pis really that widely used?" before I realized you were referring to Apple.
@@pancreasenthusiast I also prefer Apple Pie to Raspberry Pie (couldn't help my self) :P
Thank you so much!
Also, that's awesome that you were using color thermal paper for your work. That's really wild, I don't think I've ever come across a color thermal printer.
I love the technology we saw coming out of the 50s, 60s and 70s and it must have been wild to watch computers change so dramatically during that relatively short time. To go from vacuum tubes to the 6502 in just 25 years must have been wild!
Back in the day I did some programming on an IBM 360 mainframe. The printouts came out of an IBM 1403 chain printer that was 1,000LPM. Chain printers are fascinating bits of 70's era technology. The biggest problem you may face with getting that 125LPM printer working properly is adjusting the timing of when the hammers fire that hit the moving chain. This is going to be an interesting journey to watch.
I used to program and sometimes operate IBM 360s and those printers were beasts. The old card readers sounded like a wood chipper. Computer rooms were really loud back then!
Oh man, I can't even imagine how insanely fast that 1,000 LPM printer must have been. The simple 125LPM I have here does essentially two rows per second, but the 1,000 LPM is eight times faster than that!
@@UsagiElectric We had printers that could print as fast as most line printers could do a page eject. Several pages/second. You had to get the paper in just right and there were a bunch of grounding wires to prevent static buildup from the paper shooting by or you would get the biggest paper jam you've ever seen.
@@ChristopherHaileyYou're right about getting the paper in right. But also all the wire mesh that deflected the paper arc into the collection bin.
I once saw the operator restart the line printer after adding paper and the deflection cage wasn't locked in place. The stream of paper displaced a ceiling tile and enough paper was on top of the ceiling to break another tile and come down all over the place.
@@mikebarushok5361 Wow, never saw one go that amok, but i've seen them shoot paper straight out across the room. That must have been a sight!
In college, many years ago, we had a DEC printer of similar technology. It had a pod on each side that held the biggest caps I'd ever seen. It gave it a shape that got it named "ED 209" in short order. It had a leathal, sharp metal ribbon that held the embossed letters and spun up like a bandsaw.
It wasn't long before we figured out the order of the letters on the band and wrote a program to print a string that would fire all the hammers at once. It was deafening and the lights would flicker as little ED would walk across the room. Highly recommended. Alert the local sub-station first, though.
Those carbon composition resistors tend to drift upward in value with age so when you settle in to start going over the hammer driver boards in detail, you may want to go down the line with an ohmmeter and see how the measured readings compared to the marked value and tolerance.
All electrolytic capacitors leak. As for forming, once you get the voltage applied where you want it, let it sit overnight. That should reduce the leakage to the minimum you're going to get. And you don't need to put a lot of voltage across the capacitor. For a 25Vdc, i would slowly raise it to around 20V and let it sit. 60Vdc, i'd probably go to 40V and let it sit an hour, then 50V, let it sit, then 55V and let it run overnight.
Those older electrolytes tend to be very robust. Modern electrolytes don't take sitting around without a charge on them as well, in my experience.
I've gotten lots of old electrolytic caps like these from the local hamfests. Most of them were perfectly fine after reforming overnight.
As long as they haven't dried out or been damaged by applying full voltage without current limiting, they will usually be good.
Ahh, the good old days when you would sit in the terminal room, submit your job, and patiently wait for the line printer to print the result. Or to hear the characteristic sound of a full register dump indicating that your program had crashed.
for some of us the printer was 80 miles away, got result next day (why didnt we invest in a monitor ?) only to find one had input the wrong decimal point and the graph and result was nonsense
@@highpath4776 - for me it was the other way around. The Unisys mainframe was 25 miles away, but the printer was local.
In my "good old days" you would hand your deck of cards through the window at the computer lab and come back a day later to get your print out. Just to find that a misplaced comma crashed your job.
@@KameraShy Been there done that, basically just once as I'm only 59 years old but yeah, just managed to get that experience.
so we finaly get to see that printer
i worked assembly for Documation in late 70s and built the impact 3800 line printer. That's 3800 lpm. Absolutely insane when they were under test. Yea, they were finicky.
30V is common mechanism power suppply (even for later pin printers), +10 may be for positive RS232 interface (something low, but acceptable), -12V is definitely for RS232 interface (I assume this is serial I/F printer as many of this time), 5V of course is power for legion of TTL chips.
Edit: try locate interface chips (MC1488, MC1489) were produced from 1973, most likely are also fitted in this printer. MC1488, driver: pin 1 -VCC, pin 14 +VCC, pin 7 GND. this should help in locating power source from PSU. Datasheets are also available from Texas Intruments site. Good luck.
Yes the 10V rail should be around 12V, because that is the voltage required for true RS232 voltage levels, though it will probably be fine with 10V anyway. The capacitors really should be tested differently, or at least connected to a power supply via a resistor, and left for 24 hours to reform. 50 year old electrolytics are very much at the end of life, though you will find that some are still within spec after a 24 hour period at rated voltage via a 1k resistor, just in case they decide to short out, so the puff of smoke will not be much, though the resistor should be 5W to survive. 24 hours of rated voltage, then leave open circuit, and measure with the multimeter every hour, and before disconnecting check voltage drop across that resistor is under 1V, saying leakage is under 1mA, which is a rough limit for the leakage acceptable.
I have tested some NOS electrolytics, and of the dozen I had 6 were physically leaking when unwrapped, so got tossed immediately, 3 were not dropping down in current even after 24 hours of reforming at 5V, and were also tossed, and the remaining 3 had 2 with a rotted vent, so also failed. The sole good one held charge for over 3 months, checked once a week or so.
So the can sizes you have are all still available off current suppliers, and you really should replace, just go for ones in the same diameter can, or slightly smaller, and use a sleeve, common on replacements as well, and use similar capacity and voltage, though you can have a slightly higher voltage, and capacitance within 20% of the nominal is still fine, as those large cans have a tolerance on capacitance of +100%/-80% as still being fine. Yes that 60 000 uF one, used likely on the 30V rail, is perfectly fine, even if the capacitance is anything from 12 000uF to 120 000uF, so anything in that range and voltage is suitable as a replacement. Same for the others as well, though going close to the nominal is better. 85C standard electrolytic, computer grade, none of the low ESR units, your rectifier diodes will thank you. Those large cans make a loud bang when they fail, and can take out the diodes and transformer as well.
Though with that power supply, and those power resistors, I would turn the board over, wash it with a new toothbrush and some ethanol, and then apply some liquid rosin flux all over the board, and resolder every solder joint associated with the power diodes, the power transistors and the power resistors, and clean the board again with alcohol to remove the rosin. Will likely fix any dry joints there, and also solve future issues with the power supply. Clean that edge connector, deoxit on both sides, and plug in and out around 5 times, clean again, and apply a last coat. Same for the TTL boards and drivers, and same for any IC in a socket, clean the legs, then plug in again, pull out, clean pin and socket, and put in again. will solve any poor contact issues before you have them, leaving 1000 less problems to solve.
Before powering up the printer, get some 24V turn signal lamps, and some 12V ones, and apply the 24V lamps to the 30V supply and the 12V supplies, and the 12V one to the 5V supply, to give the PSU a small load. Off the 30V rail you will draw roughly 2A, the 12V rails around 1A, and the 5v rail same, well below what they likely are rated for, at least for a 5 minute test.
Then use your multimeter to measure the voltage out, should be within 10% of what you measured no load. Then take your scope and look for ripple at 60 and 120 Hz, on the output, with AC coupling. Should be under 100mV, and on the capacitors should be under 1V peak to peak of ripple, and at 120Hz, if there is a 60 Hz component you have a diode that is open, or a bad joint, and the 2 peaks for the 120Hz should be roughly the same height, if not you have a failing diode, either nearly open circuit or leaking badly, thermal camera will tell you if a separate diode that it is running at toasty level, or a bridge rectifier block running at 50C above ambient, which under low load should not be the case.
This printer is a Centronics parallel 7 bit wide interface. No RS-232 interface in the Data-100 model.
@@kromaine13Thanks for clarify.
@@kromaine13 No difference in the power supply though, and the 12V rails likely are also used in the hammer drive section to provide bias voltages for the driver transistors, so they turn on and off fast to drive the hammers.
@@kromaine13If there's no RS-232, I wonder what it uses the negative rail for? Centronics is 0 to +5v, iirc, and so are all the TTL, of course. The hammer drivers are probably a simple open collector driver, so no need for a negative rail there, that I can see. Maybe some early NMOS RAM used for a line buffer?
Thanks for the printer tour and welcome to the new year. When you think about it, printing a line of 132 characters every 480 mS pretty much maxes out the capability of 7400 series TTL on boards of the size in this printer. Gate delays are on the order of 10 to 20 nS, thus the maximum rate for a system clock is probably between 5 and 10 MHz. The chain/band probably has the character set repeated around six times. You’ve got around 2 mS or so for each hammer to fire over the correct character before the rack moves in to the position to print the adjoining position. That makes this machine pretty much state of the art for its time.
are not some common characters repeated more to minimise chain movement ?
@@highpath4776 The chain or band speed is constant. I do not know this printer, so I don’t know whether it is chain or band. The two types are materially the same in operation. The difference is the band has embossed characters and a chain type has interlocking cast type slugs. Both move the type at constant velocity and fire the hammer when the hammer is over the desired character. Repeated character set allow lower velocity. The advantage of band printers is that end users can change the band when it wears or to change character sets. This printer economizes by having each hammer service two adjacent positions by jogging the rack.
Printers like the IBM 1403 had a full 132 hammers and could print 600 LPM. The type chain of a 1403 moves at scary velocity, hence the other comment about a chain stuck in a wall is believable. That is pretty remarkable, considering the 1403 dates to 1959 and was part of the transistor based 1401 system, no ICs. The 1403 saw duty on other systems in laters years up through at least the 1970s and it would not surprised me if some 1403s are still in operation.
@@wtmayhew There is a working 1403 at the Computer History Museum. There are a number of youtube videos of it running.
I used to have to fix line printers. I can't remember half of it 😅. I do remember a lot of issues were simply caused by the mechanism getting dust from the paper in it.
I bet. Pulp + moisture (or oil from the running gear) would for sure gum up those things. It does even today on cardboard or thicker paper printers.
@@aserta yes that and actually (maybe TMI) a lot of human hair. I used to wear gloves which would be shredded by the time I was done.
@@lindoran presumably those with (Acoquisic) covers tended to have less hair in them.
@@highpath4776 you would think that would be true but I found it in the weirdest places. One of the worst dot matrix (not the same I know) that I ever worked on was in a jewelry store. This thing had a noise shroud and was in a drop feed cabinet. It needed to be completely disassembled because hair caused just enough of a clearance issue that the paper would jam nearly every time. I think it matters how the paper is stored and who changes it.
@@lindoran I used kitchen gloves, one size small. Yup, normal gloves have no chance of survival.
Ah, the stuff i used to find inside printers... :)) Worst job i had to pull was chewing gum inside a big Canon printer. Had to clean it, because it was an expensive unit, not like a cheapo these days where you can write it off. That was fun.
18:18 That looks suspiciously like the form control tape - the ICL 1932 drum printers that I used in the 70s used them so that form feed and other form controls worked for whatever stationery was required (remember FORTRAN4 H1 and H2 print controls? the 8 tracks of that loop were what it referenced and H1 was by convention the top of form). We had something like 20 loops for the various stationery types that we used and they were made of a mylar reinforced paper tape for durability.
Not this unit, but have some experience with another manufacturer's 'band line' type printer. The 'belt' as you call it on mine was literally a 'band' of metal with small metal fingers cut in one side with each letter so that the hammers 'flexed' the fingers against the ribbon/ paper. Anytime we changed the band out for another one with different sets of characters, we also changed that paper tape that 'told' the printer what band was installed. I saw you have two paper tapes, so.... make of that what you will.
@tradde11You know, now that you mentioned this, it DOES ring a bell. Something about header space and vertical 'tab'?!? Maybe it was when we switched from normal 'green bar' paper to shorter 'letter' sized tractor feed. Sorry, it's been a long time... :)
P.S. Yeah, i said 'paper tape' meaning that type of punched tape, but you're right it wasn't physically 'paper'.
@@mikefochtman7164 I knew it as "Vertical Forms Unit". The printers I worked on had an electronic version of that or EVFU.
@tradde11Yeah, the one we had, if you had a mixed case band (both upper and lower case) it would do 300 lpm. But if you had an upper-case only, it had all the letters on the band twice, directly opposite and it would print 600 lpm. Paper practically SHOT through that thing. lol
electrolytics dont have totally zero current even when fully charged, its normal .. especially with very large value ones ...charge to full working voltage via current limiting, hold for half a hour or so, discharge via a bulb or power resistor load, recharge, then discharge again, then test for capacity and esr if you can ...
The memory on that is maxed out. 256KiB is the most you can address with an 18-bit address space.
LSI was actually invented in the late 1960s at Rockwell. They developed a four-chip solution which replaced over two dozen chips for a portable calculator produced by Sharp which was introduced in 1969.
Right. By the start of 1975 LSI was widely available and we were into the second generation of 8-bit microprocessors, with both the 8080 and the 6800 having come out the previous year, the Altair on the cover of _Popular Electronics,_ and the 6502 less than a year away.
When I was a senior in high school in 1994 I got a job at a local advertising company and did nightly tape backups of their archaic system. The printer used was over twice the size of that one and I had to print the payroll checks and lots of daily updates on it.
When I was working in a Lab at my university almost 20 years ago there was a 1mb hard drive being used as an occasional doorstop. The thing was probably the size of a concrete block and weighed 25 lbs or more. Had to be from the 80s. I always wanted to build a wiring harness and see what was on it - even now 15 years later I still sometimes wonder what was on it. Missed opportunities. Strange things we choose to remember
@@FreejackVesa you probably could have bought a decent car for what that originally cost!
I worked with similar high capacity printers in a large air conditioned ICL Series 39 mainframe computer room for a while in the early 1990s, in fact they were even bigger and were very old even then, but many such commercial printers were built to last and with preventative maintenance could survive many years of heavy punishment.
It’s satisfying to watch you clean and test circuits.
I spotted that ‘-‘ sign! It’s easy when you’re in the peanut gallery. 😅
The fun of the Data-100 is the complete Hammer Bank assembly shifts to the left then prints the odd positions then shifts to the right to print all the even positions on the line being printed, the paper moves up one line. The fun is the timing of the print-belt spinning across the front of the paper and the Hammer Bank shifting Left - Right - Left .... 250 times per minute for a line speed of 125 Line Pwr Minutes ( 125 LPM ]. FYI - The power OFF position of the Hammer Bank is in the center, not Left or Right. The motor & Cam Link drives the Hammer Bank when the printer is printing, it loud...!
The foam on the hood that helps with the soundproofing can depreciate to dust
I think it has
When I saw the address of the manufacturer of this printer, I had to Google it. As a strange matter of fact, I've actually been inside that exact building at one point! Though it was many years after this printer would have been made and the building had been repurposed by a different owner by that point. It's definitely a small world!
4:20 Instead of pulling up the paper, should have flipped open the side gates to remove it.
The location of the tape loops suggests to me that it was used to control the platen for preprinted custom forms, the default tape loop would just skip over the perforations in the fan fold paper,
Its cool to see alot of older retro tech that normally isnt on other channels. Alot of this stuff takes up a ton of space and would be hard for alot of people to fit into their collection.
When it comes time to clean up the character chain, brakeclean in a container that won't melt from it or high-proof alcohol are going to be your friends with a toothbrush to really get into the nooks and crannies of every character, follow up with a brass wire wheel on a cordless drill to get them to shine... Adding WD-40 to the ribbon, wherver it goes, should add enough detergents to the ribbon again to produce ink, and be mindful of what lubricants are used where, some lubricants won't like plastic while others aren't made for high speed applications.
I would be scared to connect that old power supply to the 5V rail, lot of magic smoke under pressure in those chips. Does it have a crowbar???
This is fascinating. I worked on drum based printers in the late 70s (adjusting hammer flight time, etc.), and my first printer at home (in 1979, no less) was an ICL Termiprinter, which was slightly similar in operation to this chain printer. In the Termiprinter, the typeface pieces were at the end of steel fingers extending from the side of a belt (kind of equivalent to the chain here), a bit like a giant label embossing device. It was only 80 columns with 80 hammers, and also quite slow, but it still weighed a ton. The ingenuity in such machines never ceases to amaze me.
Oh wow. I wish you were local to me. Since the 80s, I've serviced a lot of printers, ranging from impact beasts like that one (although not that specific make) through to modern lasers & inkjets.Getting yours up & running would be a fun project.
For the mechanical side of the printer, in case you are wondering: the blue tape with the perforations that you removed at 17:58, is a "carriage tape". It allows a program to skip to a certain vertical position on the page. For example if the program wants to jump to the bottom of the page to print a footer, it tells the printer to skip lines till it encounters a perforation in the column (in IBM terms, that would be called a "channel") that corresponds to the "end of the page" (if I remember correctly, in the IBM days that would have been channel 12, by convention). When I started working in the early eighties, we had an IBM 3203 printer that printed 600LPM (there existed also a 1200LPM model, but we didn't have that one). It was a beast. If you opened the cover, even if it was not printing, the sound was like a steam locomotive entering the train station. Very impressive.
This video reminds me of how much power we used to use for things... It's amazing how efficient things are today.
You should get an ESR meter. Using a power supply isn't a good choice for testing a CAP.
"We're throwing science at the wall here and seeing what sticks."
Excellent work on that power supply! Waiting for more on the printer -
The perforated tape reminds me of CuriousMarc's IBM printer. I think it's mechanically coupled with the character being printed, the combination may be an ASCII code. Then it's just rotating the shaft until a desired character is found...
8:25 8:38 hope there was no asbestos in it, haha!
15:03 [FANSPIN INTENSIFIES]
The paper tape loop is vertical forms [paper] movement. One paper tape channel is for TOF [top of forms] hole & one channel for Vertical Tab. The paper tape loop is 66 positions long (or the number of lines per page). Being able for the programmer to tell the printer to move the paper to TOF or do a Vertical-Tap was easier than keeping track of line counts.
@@kromaine13 thanks for the info! Very interesting, and appreciated :)
@@kromaine13 While easier than counting lines, the real reason is the printer could "tab" to the spot much faster than "stepping" to that spot via CR/LF. It's like the difference between the tab key and the space bar on a typewriter. Those engineers used a lot of tricks to get more speed out of their machines.
ahhh band printers.. sound like a band saw and a crate of wood peckers.. had one that a friend of mine used to print books with in the early 80's interfaced with a trs 80 along with a stack of floppy drives.. madness
"I cannot move it around by myself" over a b-roll clip of him moving it around by himself.
As kids, that paper made the best glider airplanes. Our dad would take us to work with him on the weekends. We would throw them down the massive atrium entry ways from the high balconies, and they would go forever!
Thank you for the cap reforming. That gives me a lot more confidence in reforming the caps for my old organ amplifier.
Eek 4:21 don't pull the paper. Just open the lips at the side on the feeder
I was kinda amused by the way he removed the paper. I wonder if he was concerned bout the clamps. IIRC, just slide the little middle bit down and it will come open.
The ole System 3 we had at SYSTIME had a 1403 attached. The 1403 is the same printer shown in Dr Strangelove where a pocket radio was kept. The weird thing about the print chain we used is the fact that the numbers were all turned sideways in order to print serial labels. Also fixed the program which required supervision and fiddling. Programmer made a math error. Ran perfect thereafter. Nice timing belts, btw. EDIT: enjoy purchasing new ink rollers. Heh. 132 wide, huh. Ouch
Used to work on Dataprinter corp. line printers. 600 &1200 lpm. The 1200 had a vacuum cleaner built into the cabinet.
Major dust creator. Tricky timing circuits.
Word of warning -- make sure the form (or carriage) control tape is in place before pressing top of form. If the form control tape is missing or not reading properly, the paper will fly out of the printer in an attempt to find "top of form". After several seconds of flying paper the printer will fault and stop advancing the paper. Ask me how I know?....
As a junior developer, I was responsible for - among other things - creating the carriage-control tapes for the various tax forms used by several financial companies. So I would create the tapes and then test the forms -- most print jobs would have about 10 or so "line up" forms at the front of the print job so the operators could align the output. Once aligned they would allow the print to resume. But occasionally, the tape would be wrong or break or the form control reader would clog .. and you got to watch a "forms run-away" ... I was working with 3 CDC 512 printers - 1403 equivalent -- and these would regularly spit the paper to the ceiling.
Additionally, there is a switch that will sense if the paper has run out -- and will stop the printer from advancing -- if this gets jammed open (or misaligned)-- you may believe other issue are causing the problem -- check this first -- it' s sorta the same thing is asking "is there gas in the tank?" -- again - ask me how I know.
Column 1 of a report was associated with the carriage control -- so a 1 in column 1 of the report would advance to the punched hole in the carriage control tape at column 1 typically called the "top of form". A 2 in column 1 would advance the form to hole in the carriage control tape in column 2 -- and so on. Hence the reason for 133 character report that only printed 132 characters wide ... In really complex forms there could be several different "vertical tabs" positions. Also the length of the control tape has to match that of the form - exactly!. If you got the length wrong -- the form would misalign on every "top of form" of "form-feed". -- again - ask me how I know.
-- stories of a junior developer learning the trade...
0:36 lol at the compression artifacts on the computer grills.
One small steep to the completion of the printer. Look a fun journey to the end. I wish you good luck David.
These monsters make one hell of a noise when printing. Get some ear plugs. LOL.
Happy new year.
True! He's not joking.
When I was in college, the data center had a "page" printer. Paper came out of it in an arc...
The "Odious Odec": watch those edge connectors, they tend to be fragile. Expect to find a buffer containing a line of text, with (if memory serves) a pattern clocked over it synced to the chain speed: when they matched a hammer fired.
So when you get this thing working, you can honestly claim that it's hammer time.
This printer reminds me of the 6500 IBM Printer series... it adopt basically the same configuration for the hammers and the characters belt.
I was, in my youth, a good technichan on those printers and the adjacent series 6400 ( this one had a single line of heads in a vibrating bar that compose the characters dot by dot in lines)
Both are printronics rebranded if i remember well an probably i still have the field manuals in my garage, after 40 years i have to do a bit of scavenging to find out but i can searche them.
Mechanically are quite serviceables but the eletronic was a pain.
Once upon a very long time ago, when dinosaurs roamed the data-centres and I was but a pimply faced youth (PFY)
We got a very 2nd hand drum printer a DEC branded one I think... The ribbon was a massive roll that ran over the face of the drum (think the old cloth hand towel things you used to see)
My job was to clean this beast and the drum... Armed with a large bottle of IPA a few skewers and a tooth brush, the job was to "unclog" the type faces on the drum.
One well spent afternoon and doing a great impression of a smurf once I was done.... the printer worked a treat and the output was mostly legible :)
To reform capacitors, just set the current limit to "a few mA" (I just found out today my powersupply has a 5mA offset! When OFF, you get 5mA. That's a suprise! When set to 30mA you get 36! and the display shows 30!). Just set the voltage at the max voltage and turn it on. It should be ok after just one cycle. If you want to check for leakage, unplug the powersupply from the capacitor and watch the multimeter. Measure the time it takes for a certain voltage drop, say 5V. You then get I = C dv/dt, you know/measured everything on the right side! . Oh. In the video, you start saying there remains some current while in fact my gut feeling is saying: you're deciding that too fast. I'd find it totally normal if it stays at 1mA for say 10 to 60 seconds before dropping to zero.
I "salvaged" a PDP 8 in the late eighties. I scrapped it in 1997. :-(
I still have some of the capcitors (160000 uF, 25V). And its two fans. The fans still get regular use. The capacitors are marked that I think they are early seventies.
I can't wait to hear the sound that thing makes going full bore printing.
The Fairchild part in the video at 5:34 looks like it is marked “96602PC.” There is a 96S02PC, which is a dual retriggerable one-shot. The RC network next to the chip would make sense. It may be socketed because it was hand selected for a fast enough part. The empty socket next door may be for test probing - hope so. Of course, the manufacturer’s service book would reveal all. Hopefully someone out there has a copy. The service book is probably essential for avoiding disaster working on a machine with parts moving at the speed used in this printer.
I would check the rripple voltage of the caps before progressing to power logic boards. As a general rule for me, I replace old caps like that on equipment I am going to keep. It might get expensive, but worth it.
"Ok, so I bought myself a printer. Now I just need a front loader to move it around"
Not new to the channel or for that matter to the Centurion, but I think even if you made a video working on a dishwasher I'd appreciate it. Your enthusiasm is contagious! Great content over here! 👍
Alright, you've hit on an area of computer repair that is a personal nightmare for me. When the printer was a much more commonly serviced device, and they were much larger and more mechanical beasts, I'd break into a sweat every time I got a service call for one. I'm not at all mechanically inclined, although I do wish I were. But I know this is going to be a fascinating trip, and I wish you the best of luck!
I ended up,with one of those in my living room back in the 80’s. I had to leave it behind when I was forced to move out. Hope the new owners had a use for it.
The fan needs a rebuild. That's a looong long time to spool up to top speed. Take it apart and scrub the bearings clean. Depending on how loose things are use thicker or thinner oil (thicker oil also accounts for the fan getting hot during work, so think about that aspect as well).
yep , it is dusty and any lube has dried out
I have to admit... I was expecting ancient blue smoke to escape as the angry pixies got woken up. Great job!
I did output collection and sorting from a band printer in the late eighties. Never saw a chain printer running. But if it is 100 lpm (not cpm) I'd imagine it's near the same. It was amazing watching that band printer run.... chomp chomp chomp grrrr grrrrrr chomp chomp chomp and ten 13" fanfold pages had printed.
Hope those old Sangamo capacitors are still in good shape - that was a company based in Illinois and which also made electric meters (as in on the back of the house) in addition to other products...
I worked as a computer operator in the '70s and worked with a lot of line printers and had to deal with a lot of issues but I have to admit I never had to deal with dirt daubers!
That green and white wide paper brings back memories of huge printouts that were in huge binders. We used to have to dig through them to find what we needed. No internet in those days.
Kind of curious that the latest model computer (the Centurion) has less maximum memory than the models about five years older. But again those machines were programmed tightly. They would have scoffed at our modern bloatware as their systems always got right down to brass tacks.
5:30 aww that's cool
After all the grief that the Centurion has given you, I can't wait for part 2 to hopefully be "I greased and oiled everything, plugged it in, turned it on, plugged it in to the Centurion, and printed stuff!"
You definitely deserve a project to go easy for once!
I wish you and Mr. Carson's Lab could do a collaboration.
That thing will sound wicked when it goes full blast
In the early eighties I used to work in an office where the had a Data100 (I noticed that brand name on the back of the centurion printer...) drum printer. When that was doing a test run, man that was hammering. I also have some chain printers in bad shape, not by centurion. Waiting to be restored for 40 years now I guess:). Anyway, if you look for documentation, 'DATA100' is a good link.
It's always a good day when old IC or transitior computers like this are shown off. The Centurion may as well be a character alongside Usagi the rabbit, they're synonymous with your channel at this point lol.
Worked for Data 100 from 76 to 85. Used this on a lot of RJE's. This was the smallest line printer we used. Good printer but they wore out quickly. Too many moving parts...
I'm guessing they need to be oiled on a regular basis for longevity. Must get messy when they need cleaning.
@anthonydenn4345 Yes, it's messy but not from oil. Between the paper dust and printer ink ribbon dust, it could be tracked all over the office. Almost lived with a vacuum cleaner nearby just to open it up.
Agree no oil used in the Data-100...! The paper does make a lot of dust & lots of little round paper dots...?
Some really weird compression artifacts on that tilt up shot where you reintroduce the Centurion.
The desk system at college, you had a connector wire to the drive issue with no write happening, was the cable fixed and the software replaced and now working as a CPU 6 machine
I know that somehow that printer is going to find itself incredibly useful. That Centurion alrwdy seems to see its fair share of use as a diagnostics workstation.
This is also Clean Shaven David. My Rabbit Electronics shirt was a hit with my Japanese friends at our New Year’s Eve party too.
Those fuse board are impressive! There was so much that can go wrong!
I love restoration videos.
Back in the 80's I worked as a computer repair technician. Computers and drives I had no problems with but printers, boy. Almost every one I worked on I ended up with a small bag of parts left over after putting them back together but they were working so were the parts really necessary or was my boss messing with me by slipping them in when I wasn't looking. We'll never know.
That is really neat stuff, again! When I was a tech, I loved working on the old TTL logic boards. Working on those boards really makes you think hard about how things work because it's not a matter of just pulling a single chip to fix the problem. I learned more about circuits working on those than I did when I took my engineering classes in school.
That power supply looked a bit scary to work on. I was never keen on working on power supplies after getting a good zap from a CRT circuit. I got teased a lot by my coworkers because of it.
I always called 'em "mud daubers" but I like the alliteration in "dirt daubers"
These printers are SO COOL. A copier tech friend of mine told me about a call that he got on one of these. They had it serviced the day before, came in the next day and the paper was running backwards. Turns out that the paper feed motor was 3-phase. Two of the leads had gotten swapped, reversing the motor. Ooops!
Praying you did not get a Bang‼. I try to put on glasses when I fire unknown hardware. Thankfully she powered up quietly. Great old stuff. This is the first chain LP I have seen where ours were drum.
My company still prints thousands of pages monthly on green bar paper. Our daily use printer is from this millennium and has Ethernet but it's interesting to see how far back this format goes
Slowly catching up on your videos really great channel, I enjoy watching you fixing those old treasures! I think you should get an LCR meter to check the caps properly. I think the DE-5000 is quite popular. It can not only check the capacity but also the ESR value of the cap.
Epic! I was only in the room with running (ibm) chain line printers a few times, a little terrifying, but loved the wide output (and scratch paper)
Hah I pulled a hefty cap out of an early 80s Xerox machine as a kid in the 90s - the cap may even be older than me. It's in my kid's custom power wheels acting as a large power smoother.
I had no idea how young you are. This channel pops up in my feeds from time to time but I never watched a video until now.
Happy New Year, Happy New Project and Bunny!
HELLORLD! gonna look pretty sweet on that green bar paper
Ah printers. The most pain-in-the rear device.
0:36 I thought I was tripping there for a second
Maybe it's the Beautiful Flowing Hair. (Maybe it's Maybelline)
But I Swear until you put your Hand on the Centurion Terminal, It Looked like you Were Superimposed in front of a Green Screen LOL
A very nice. Learned something new I think. Paper tape reader for programming. Not surprised the psu came up. I don't have any experience with chain printers. But, the old dot matrix printers I did just needed a good cleaning and lube. One tip is to use paint marks or what not for alignment. Also, unless you find a service manual. I would try not to take it 100% apart unless needed. Luckily you can move most parts of the printer by hand. The only down side of this is you pointed out. Printers are a whole new experience. Also good chance those logic boards will probably just come up to. In fact the only thing I'm worried about is any hardened lube. Turn it on and it throws a part.
0:36 looks like video compression algorithm went crazy 😄
Reminds me of the HP2619 printers I worked on back in the early 80s. Hated them. Getting the print chain and hammer bank set for even non smeared printing was almost impossible. And with a swinging hammer bank -GOOD LUCK!. (Reminisce of the 2613)
Watch the clearance or you will tear out sections of the print chain.
Like the fact they put fuses on each driver at least if a hammer shorts you should only blow the fuse.
That little paper tape on the side controls the paper feed. One punch was end of page another top of form.
The others could be used to advance the form to specific lines on the page via control in the programs.
I live in NY I had no clue what a dirt dauber was till vice grip garage then that summer i had 3 of them building nest and now i see them all the time LOL
Anoother great video! thank you! we have the same passion and the same hair color :-) .. cheers from Italy
So many comments already! Hope you see this one - a few possibilities for further research.
05:55 That sticker bears Data 100 Corporation markings. Data 100 was a manufacturer of IBM-compatible peripherals (spun off from CDC in 1968, bought out by Northern Telecom in 1978, although the brand name stayed alive until the early '80s).
As far as I know, Data 100 was the original manufacturer only of the RJE and direct-data-entry terminals. The company rebranded and resold card readers, paper tape readers, and drum printers from Dataproducts; card punches from Groupe Bull; chain train printers from CDC; and tape and disk drives from MPI. But I'm not familiar with any CDC train printer that was that slow. The ones I worked on were 600, 900, or 1200 lines/minute.
It looks as if the printer is trying to be a clone of the IBM 5203. That's in the right period, similar form factor, similar speed, and since the Data 100 stuff was all supposed to be IBM compatible, I bet the beast was made to interface to an IBM 2821 controller. It's clearly NOT a 5203, though. It has two print positions served by each hammer. The 5203 had one hammer for every four print positions.
The manuals for the IBM gear are still around, which would at least tell you what sort of thing that Data 100 was trying to clone.
Absolute legend.