It brought a tear to my eye. My dad and I have worked on Americas space programs. My dad worked on Apollo. I have see an Apollo launch. I watched the Moon landing on a crappy black and white TV as a kid at summer camp.
Incredible! Thank you so very much for sharing this! True American engineering, back when they made things to last! These are prototype test boxes, that have suffered some corrosion due to age, but even without anything being replaced 50 years later they still fire right back up without issue. The gentleman that owns this stuff was very wise to realize the value of keeping such things when he convinced NASA too sell him the 2 tons of junk equipment. NASA has a very bad habit (still to this day) of discarding historical items. They left reels of Apollo film & magnetic data tapes sitting in cardboard boxes shoved into storage rooms in FL & TX for decades. Some of it was water damaged from sitting under a leaking roof, others turned to dust due to dryness & some simply thrown in a dumpster to clear out the space. What pisses me off is when people like this gentleman save these items, but them from NASA, store them, care for the & then go through the time & huge expense of restoring these literally priceless historical artifacts & then suddenly NASA wants it back. I have massively mad respect for each of these fine gentleman working on this project. What an amazing thing to be a part of! I'd love to be in that room just for a bit, to be near those devices (it's a space & computer geeks wet dream all in one!). You can tell this is a labor of love for these guys. It's awesome to see different generations working side by side on such a pivotal piece of history.
Encore aujourd'hui en 2024 c'est toujours aussi passionnant ! Ces personnes ont travaillé sur un morceau de l'histoire de la conquête spatiale avec simplicité et compétence, c'est incroyable. En regardant cette seconde partie de la vidéo, et comme ancien informaticien, je ne peux m’empêcher de penser avec admiration et respect à Margaret Hamilton qui a programmé, avec son équipe, certains des modules de ces nouveaux systèmes, c'était une première à l'époque. Un big merci à vous pour cette magnifique vidéo qui traversera, j'en suis certain, les années. Bravo et encore merci à tous, vous êtes de brillants et sympathiques ingénieurs !
This is a bit surreal. From how these units were discovered, and saved. To you guys opening everything up in a hotel room, and sharing it via technology that didn't even exist back when the modules were created.
Man, almost all of this stuff goes over my dumb, little head. But it’s amazing that you’re preserving this important piece of human history! Also, it’s amazing this power supply works perfectly after 50 years...they really knew how to build that stuff back then!
I look forward to subsequent parts of this series. I worked on a large, refrigerator sized computer (digitizer) in the early 1980s. It was a Burroughs machine. It had 753 circuit cards in its three backplanes. The cards contained discrete and DIP devices. All of it was interconnected with 1 to 3 level wire-wrap. I never added up the number of wire-wrap pins, but I would say it would come out to be around 75,000 of them. It was tried and true technology. Very stable and reliable for its day. In the seven years I worked on it we only had one wire-wrap failure. That was related to a wire going around a pin and the edge of the pin cut through the insulation causing an intermittent short.
Seeing the wire wrapped backplane brings back some great memories of my days as an IC design tech. I’ve lost track of how many miles of the stuff I went through back in the 90s, building breadboards for my engineers. All that beautiful silver-plated Kynar, now rotting in a landfill.
I am in absolute historical, nostalgic, nerd-heaven. Thank you for sharing this project with the world (even if it gets fewer views than an out-of-date fashion reviewer exhibiting 2nd hand swimsuits from her local resale shop). Unfortunately this is the world we live in... Thank God there are still some curious individuals with an IQ higher than 70 that share sure h brilliant cobtent with the few who wish to watch it.
This will get better and better with each episode. I can't wait to see the logic come to life, even if it's only partially operational. I'm really interested in seeing what kind of DIY interfaces you might come up with.
The engineering and technology there is absolutely amazing and beautiful. Its absolutely amazing they could get so much power in a small package. Granted this was a very specialized computer designed for one thing unlike many of the time, the design and technology is just literally out of this world for the time!
This is roughly the same computing capability as the IBM 1800, a real-time mainframe of the same timeframe, but instead of taking multiple full sized racks and a lot of power, it fit in a small volume and used very little power.
@@carlclaunch793 I've been aware of this effort for a few years now but I've never actually taken a deep dive into the specifications of how computers of the era were built. How did they achieve that? Was the lower power consumption all thanks to the process used for fabricating the NOR gates it was built out of? All the best to your entire team, I'm looking forward to seeing the AGC put through its paces!
@@sietuuba The low power was the cumulative effect of hundreds of design decisions and component choices. The compact design itself didn't require driving transmission lines between far flung parts of a mainframe computer of the era. Switching power supplies whereas mainframes of the era had big linear supplies. Doing cycle stealing to maintain timers and counters instead of having dedicated circuits for each one. ICs surface mounted on multilayer boards, plus cordwood construction and core rope memory all shrunk volume dramatically.
@@carlclaunch793 I can't imagine that is was super efficient with all the analog power supplies that stepped down all the voltages and the miles of wires connecting the components.
@@FrederickWalser Much more efficient than would appear initially. The power supplies are switching design, not the usual analog supply of the time that burned off excess voltage as heat. Considerably higher efficiency. The majority of the wires you see carry very low currents - signals at .8 to 4 volts to only a few transistor base-emitter junctions. The power feeds used a pseudo bus-bar connecting groups of power pins down at the bottom of the pin, reducing resistance compared to point to point wiring. Many of the connections were traces on the multi-layer boards which held the ICs - short distances and low resistance. Board to board connections passed over gold contacts, once again lowering resistance versus copper.
Being a detail oriented type of techie/geek/nerd, the more detail the more I like it, regardless of the length of the videos. Thanks for another great video, please keep them coming.
As an 40y-old fanboy of the AGC I'm full of admiration of these guys, doing it in a HOTEL ROOM, where others would be incapable doing it in a fully equipped laboratory. But the AGC is in awesome condition. Absolute unique specimen you got there. Just wow. I'm following Mike Stewart's work for some time too, he's THE expert and I'm full of respect for his efforts, I'd cook a meal for him every day, just call a day before you visit northern Germany, Mike ;) No, seriously, as a collector of old tech I'm convinced it's important to preserve it for the coming generations. Ask my daughter how often I told her how to use an old HP calculator in postfix notation (and why one should do this) or when I hang another picture of the Curta calculator in the hall... Just saying, there's a community around the world that's with you. All the time. Go on with your good work!
Lots of engineering gets done in hotel rooms. BTDT, getting something ready for an urgent customer request a day before it was going to be installed. It’s a somewhat distraction-free setting.
if this is how computers were physically laid out and cased and labeled, I understand why the computer in 2001: A Space Odyssey looked like it did: it's like these gate modules, but made in a "futuristic" material ie plexiglas
2001 is interesting for how hard it is to project what tech will look like in the future, even though there where experts hired to do just that and it was taken seriously. The most telling for me is the shuttle flight going to the space station fitted out inside like it was an airliner doing a something mundane and the space station was on a scale and homogeneous design that is still science fiction in 2018. Then there is the manned mission to Jupiter that is just laughable. Considering the movie was released in the first year of the Apollo space program and how rapidly that culminated in numerous manned missions to the moon the vision into the future seem very realistic if not conservative for the investment in space exploration of that day.
Definitely inspired from large 1960's computers. Actually I think it is more inspired from something like the PDP-1: ruclips.net/video/1EWQYAfuMYw/видео.html .
Considering how long Voyager 2's been working for, its not a doubt how well things were engineered for space, im almost positive this computer will work without too much hassle with the proper love and care it seems to be getting
What an amazing piece of engineering. It’s all too easy to say this computer is so slow by today’s standards but I bet it would be more resilient than any general purpose computer in the environment it was designed for!
Yes, you can't put your standard home computer into space. You need to design for the high acceleration and the vibration that the rocket or spacecraft will experience. Oh and a software crash is totally unacceptable.
@@gorillaau Think about Apollo 12 and how the IBM IU just kept chugging away after the booster was hit by lightning 2x!!! A modern computer is toasted if the power flickers to hard!!! If your computer locks up during launch your gonna have a bad day!!!
The question is not "was it slower then todays computers?" - Yes, of course it was. The real question is "Was it to slow for the tasks it had to do?" - And to this question the sound answer is: No, it wasn't. Although not by a large margin, it still was fast enough to do that.
For now, this all series could be resumed by "Trying to find something bronken on a 50yo space computer, damit it was working all along!". I will forever be amaze by Nasa engineer and reliability
I wouldn't say is just Nasa, I worked in several systems in the US Navy as an Interior Communications Tech and they're just as rugged as these ones. I realized most people haven't seen military grade systems, but believe me they're just as astonishing. Apollo systems are obviously on a top tier due to the nature of what it was supposed to do.
Every time someone says "spacecraft," I get a chill. I'm so used to that word only being used in the context of science fiction that hearing it here, in regards to an actual vehicle designed to fly beyond the atmosphere of Earth, just...I don't know how to express it. Induces a feeling of awe, I guess. Amazing.
I have to say I'm actually quite frustrated that I found these videos now and not in 3 months when the project was complete. I NEED MORE. Thank you for all your hard work. We appreciate the history being honored. THIS. IS. AWESOME!!!!
It's fascinating to look at the size of some of the components and the technology applied within the AGC; only then to realize that we're actually at the end of 2018 and this was built 50 'kin years ago! 50 years...bloody hell, man! The point-to-point wiring has me drooling and procrastinating in equal amounts - so without to do; I'll sit here in my wet patch and watch the next installment(s) :)
By observing how complicated it is to dive in and check these modules one can imagine how complicated and goal-driven the design process was in the first place. Everything in the Apollo program, not "just" the AGC, was state of the art and beyond at that time. The engines, the two types of spacecraft, the suits, the docking, the rovers, even the parachutes, everything. This is truly one if not THE most ambitious single effort human kind has made to reach one goal. And so often there were the right people in the right place at the right time. E.g., Apollo 12 with the lightning strikes and the "SCE to AUX" story. Or Apollo 11 with Armstrong at the controls in the final descent with no fuel left. Or Apollo 13. or or or. Each time I dig into something related to Apollo I think "This is unbelievable".
Watched part 1 last night, just watched part 2 and I’m totally blown away by how amazing this is. I have no interest in this stuff but I’m glued to my screen! Amazing stuff guys
Great video. Quite the leap in technology over the decades from the time of the TTY from you last video. Amazing to ponder. If there was ever any doubt, this confirms I don’t have the “right stuff.” If someone asked me to do a half-million mile round trip on that thing to the moon and back, I think I’d have to pass, regardless of the quality..
Fascinating and not long enough. This is so cool. Awesome to have the technology explained also. Thank you for posting this. The comments are really useful as well. Wish I was there though.
Mind blown. Not only because the computer but the insane accurate knowledge this guys have of this system. Probably even better than a Nasa Technician at the moment of repair or upgrade.
I need to say it out loud--Curious Marc RULES. Great channel!! I am always blown away by the crazy mechanical and digital computer restorations you take on--fearlessly--and always enjoy seeing the "Marc Army", Carl, Master Ken, etc. practicing their unbelievable tech kung fu. Glad to see you got this restoration gig, you deserve it, and it's all done with your usual dedication, determination, humility and humor. It's already fully ok to be a nerd (Bill Gates gave us that?) but seeing this sort of thing is icing on the cake. Keep on keeping on!
I'm so sorry i needed days to come down to watch your video...... Seriously, you are to me one of the holy people of 'archeology programmer' sort of people, generations to come will reference and learn from you.. THX for your awesome work and parts from the alto, the teletype and now the AGC. ;) Hacky, from Undeloh, Lower Saxony, Germany, Europe, Sol 3 (Earth), Sol System...... ;)
This is fabulous and essential work you are doing: there needs to be greater awareness of these computers because of the legacy they continue to leave.
It does kind of crack me up that every device used to test, measure, and document this restoration, is more powerful than the computer you are actually restoring. Probably even including the digital multimeter!
The new stuff is much more powerful, but also more fragile, both physically and operationally. I'll take the AGC over the newer stuff any day if I have to fly to the moon.
@@phil4826 What you say is not really true. You are comparing something from the past that was built specifically for reliability in an extreme environment to something from today that was built to be as cheap as possible and still do the job in a home environment. A well made modern cpu is actually an extremely reliable and durable component especially compared to this. It's everything attached to it, powering it, supporting it that is made to be cheap and not up to the task. And there was plenty of cheap, fragile crap being built for home use back in the 50s and 60s too. I wouldn't want to fly to the moon if my life depended on the reliability of the cheapest radio found at Sears in the 60s. My cheap modern cell phone is both more powerful and more reliable than that. Fortunately, neither our brave astronauts of the past nor us today have to do so. The stuff built for Apollo was not standard off the shelf electronics. Every piece was built to a higher standard and tested thoroughly. If those power supplies had off the shelf 60s capacitors in it, that thing would have smoked the minute they put power to it. And our astronauts of today don't go up with an off the shelf Dell laptop. They have systems designed, built and tested for that specific job and environment with a backup and another backup. But if you just have to have a super reliable, nearly indestructible calculator with a lifetime warranty on it to help do your taxes, I would be happy to build you one. Just send me $100,000.00 US dollars and you will have it in a years time.
@@wingracer1614 Yup. Anyone who's been inside an "All American Five" radio can attest that consumer electronics have always been designed to be "good enough".
@@wingracer1614 he said hed take the old stuff over the new when going to the moon. you agreed but called him wrong. he wasnt talking about consumer electronics in the 60s
Link to the backplane viewer: apolloguidance.computer/2003100_071/pins I'm not a web developer, so please excuse the poor resizing capability and lack of mobile support!
Thank you so much for posting these videos, absolutely fascinating to see the technology used and realize the best minds at the time were working on these, and you can see it in the quality of the work! However, from a poor broadcast "engineer" who wishes he could understand the math behind this all, could you add a drop shadow to your CG text? Just a simple black border or drop shadow would do wonders against the gleaming bright metal work of the AGC. It's sometimes hard to read the text comments with white on white background etc. . Thanks a ton! Keep posting updates!
This will sound like a dumb comment but this reminds me of the computers some crazed lunatics have built in Minecraft. The similarity being that the "electronics" in Minecraft were never intended to build such a thing, it was only meant as a little curiosity for making simple switches yet when there is a possibility some people will always push things way beyond their limit. When I see these nests of intricate wires I don't just see the same kind of structure in MC computers but the same mindset.
Saying this is 16-bit was like people saying the Nintendo 64 was 64-bit, "Sort of kind of not really" for pretty much similar reasons interestingly enough.
@11:35 so they had found out that using PCBs and soldered components in the Polaris missile was unreliable, so they didn't take any chances and went with wire wrap and welded components for the manned missions. That made me realize that the technology used in our cars today IS soldered components and PBCs... I guess (hope) there has been huge progress in the reliability of that technology since then ? lol ! Anyway, very enjoyable video as always. Thanks for sharing, Marc. BTW, 18 minutes is NOT long-ish at all with this kind of fascinating content (for the nerd in me anyway). It seemed like 5 minutes !
AlainHubert - Yes, it’s a lot more reliable now than it was in the ‘50s & ‘60s; however a welded connection is still better than a soldered one. Also, FWIW, the wire-wrapping process effectively creates a cold-weld at the corners of the pins where the sharp edge cuts thru the kynar insulation and digs into the wire - this is due to the tremendous tension that the wrapping turns are under. This tension also helps with vibration resistance. The reason soldering is used instead of welding is a) it’s good enough most of the time; and b) it’s easier to do quickly in large volumes.
AlainHubert Of course there has been development. However, cars don't usually experience 5-10g, possible extreme heat ot cold (-150c-500c) :D They serve their purpose. I've seen a lot of failed stuff like failed joints on automotive stuff, but usually after more than a decade.
@@rkan2 You forget the 5G 50mm vibration in all 3 axes the missile boards had to endure during testing and use. Missile boards are still fully conformal coated in a flexible coat, then potted in a hard coat after testing, then potted into modules as final step because of the vibration. Soft coat absorbs stress and the hard potting speads the stress evenly. only has to work flawlessly once, but has to endure a lot of testing before that, and a lot of abuse.
Thoroughly enjoying these videos. I recognize that there were many, many hours of study and preparation to manage the real thing as deftly as you are doing. Can't wait to see the next one.
Having spent close to two decades working on computers with 3 level wire-wrap backplanes, I can state that wire wrap backplanes are not necessarily 100% reliable. Those G-D machines could pull the wire tight around the pins when cornering a long wire, and the Kynar insulation would eventually cold flow and short to the cornering pin. Depending on what was on that pin, this could cause all manner of havoc, and could be darn hard to locate. It looks like the wiring in that backplane area is very open and I didn't immediately see any tight cornering, but it is certainly something I'd carefully look for.
The computer design review brings up the cold-flow problem, and it was taken into deep consideration. Here's an except from the review book: The simple solution to this problem would be to obtain a wire with a greater "cold flow" resistance. However, these tougher wires present problems to the wire-wrap machine. They do not strip easily and have a mechanical memory which makes them spring up from the wire wrapping surface. It is, therefore, necessary to make a compromise between the toughness of the insulation and the ability of the wire-wrap machine to perform the wire-wrap operation successfully. After more than a year of search for an acceptable wire, Raytheon has finally located a vendor who can produce a wire which is both tough and "wrappable."
@@mikestewart8928 Simple solution I saw was to leave some slack in the wire every time, so that it was not pulled flat on pins. Then they took the runs and laced them together with lacing twine in a loose loom. Lot of manual labour extra, but as they only made around 50 units not going to add too much cost wise.
IBM had a control unit, the 2821, that suffered from lots of cold flow problems which required rewiring the backplanes for all the wires that had been pulled tight around pins. Most other machines they built with wire-wrap had used procedural steps during wrap to avoid this problem
It could be a real problem. I worked on a machine at Burroughs that used wirewrap coax for all of the signal connections in the processor in the first prototype machine. The cold flow problems with that were so bad that we went to wirewrap twisted pair for the second prototype and the real machines.
On the other hand: Every single piece of equipment used went through a long series of testing before the launch. Especially with the electronics NASA tried to limit down the possibilities of random errors to as much as possible. Eg. there was the concern that the chips might fail. Thus random samples of chips were taken and drowned into a liquid (I am not 100% sure but I think I remember it was Freon). The chips were weighted before and after that procedure. If the weight differed it meant that the casing was not 100% tight. If just 1 chip failed that test, the whole batch was thrown away.
I had no idea there was 2 dozen people watching and clapping at the power supply. Fitting that many people in the corner of a hotel room out of sight is a huge feat on it's own.
The raw scans of Don Eyles's AGC Handbook can be found here: archive.org/details/agc_handbook_jp2 They have been cataloged and indexed, along with Eldon's schematics, here: www.ibiblio.org/apollo/ElectroMechanical.html
They could’ve used my iPhone back in the day. My dad was management in TRW and helped design the redundant guidance systems in Apollo. They used it on ONE flight.
While going thru the "AGC Handbook" noticed that the Signal Wiring diagrams were missing for modules 13 and 23 and a few other miscellaneous pages. Was the document you received incomplete? Thx for updating that backplane link. I probably missed it but is there a wiring diagram that covers the backplane?
Yeah, there were sadly a few missing pages in the Handbook. The good news is that we've since recovered the missing pages, as well as many many more revisions of the schematics, from the National Archives (in higher quality too!) The AGC schematics are split between boxes 459 and 460: virtualagc.github.io/virtualagc/AgcDrawingIndexBox459.html virtualagc.github.io/virtualagc/AgcDrawingIndexBox460.html Search for the drawing numbers on those pages and you should be able to find anything missing from the Handbook. There sadly wasn't a wiring diagram that covers the backplane... and (amazingly) not even one that covers the internal inter-tray connectors. All of that information was stored in a physical book called the "AGC Wirelist", of which there were very few copies. Instead of digitizing those onto microfilm, they instead scanned little notices that said to get a hard copy from MIT (here's an example: archive.org/stream/apertureCardBox459NARASW_images/apertureCardBox459NARASW#page/n1691/mode/1up ) It's not clear at the moment if MIT still has any copies of the wirelists, but hopefully they do, and finding out for certain has been on my list of things to do.
Is there a complete set of schematics available anywhere. I am in the process of building a hardware gate accurate AGC but I only have schematics A01 - A29. By the way I tried the backplane tool but could not get it to work. It just showed a fixed image.
The raw scans of Don Eyles's AGC Handbook can be found here: archive.org/details/agc_handbook_jp2 They have been cataloged and indexed, along with Eldon's schematics, here: www.ibiblio.org/apollo/ElectroMechanical.html
They would be happy to if you would actually buy it but would you pay triple the price if the only feature it had over the cheaper versions was a better power supply? You'd be better off buying the cheap one and when it dies, buy another one.
I'm an embedded software engineer, and have been doing flight software for various spacecraft from various newspace companies for the past ~8 years. My AGC hobby project started as an on-and-off thing of a lot of thinking and reading about five years ago, and a bit over three years ago I started in on an effort to simulate and then build a gate-accurate replica from Eldon's schematics. Roughly a year after that the project began to grow rapidly in scope, after I got in touch with Don Eyles who has, over the past couple of years, contributed for scanning over half of the program listings we have today, as well as all of the Tray B schematics for the AGC itself. And here we are today... :) It sure does feel like a lifetime, though, haha.
Fantastic stuff. Any plans once it's up and tested to hook it up to emulators of the peripherals, load up the software, and run the whole shebang in a closed loop? Maybe even do whole simulated missions complete with failure scenarios? I'd tune into that show.
Yes, we hope to do something like that. It depends on how well everything works and how ambitious we are. The peripherals do I/O with pulses at weird levels, so we'd need to make some emulation interfaces. One problem with doing a simulated mission is that it would happen in real-time, so it would be very slow to perform unless we can jump to the parts where something happens.
I sat through a lecture a few years ago about PWM. It was at a Seattle Makers thingy. Lots of remote control "robots" on display. Did you know PWM was invented "around the turn of the millennium?" Hilarious.
They used spot welding; basically a quick zap on each joint. It wasn't like a guy with an arc welder joining steel beams together. Here's a documentary on the assembly of the AGC showing the welding taking place: ruclips.net/video/ndvmFlg1WmE/видео.html
Did you have to solder wires to accomplish the strapping of the power supplies? Can't see what they are soldered to. Fascinating technology! Seems like this computer connected directly to lots of external things. I had thought it would be used more for manual calculations, but seems like it was doing real time control based on sensor inputs. Thanks for making these videos.
Some of the pins on the bottom of the module accomplish the configuration. ON the backplane where the power supply is plugged, wires are routed to the pins in the pattern needed to select either 4V or 14V
Yes, the guidance computer is doing real-time control. It receives data from the gyroscopes, accelerometers and radar to determine the spacecraft's position. Meanwhile, it controls those devices as well as the engine and RCS jets to keep the spacecraft on course. There's a diagram of its inputs and outputs here: archive.org/details/acelectroniclmma00acel/page/n165
I'm slightly confused. If this came out of the first man-rated lunar module, then why was it populated with a non flight-rated unit, and why is it being repeated that it was just a development unit? Unfortunately Google isn't much help right now, googling anything about "first" and "apollo" or "lunar module" just yields results about Ryan Gosling.
Grumman produced a set of test versions of the LM as well as the machines destined for the specific missions such as Apollo 10, 11, etc. These earlier machines were called Lunar module Test Articles and our AGC was in LTA-8. LTA-8 was used at the Manned Spacecraft Center (now Johnson center) in the thermal/vacuum chambers. These chambers were the only ones that NASA owned which were large enough to hold an LM, CM, or CM+SM. They would subject the spacecraft to temperatures ranging from hundreds below zero to hundreds above, while in a vacuum, to verify that the spacecraft interiors would remain habitable. Astronauts sat in the spacecraft during the tests, simulating a mission. LTA-8 was basically the same design as the production LMs used for manned flights, but was subjected to abuse on the ground to ensure that the production ones were safe for flight. heroicrelics.org/space-ctr/lta-8/index.html Our AGC was the 14th of 15 prototypes before the first flight models were produced, although it has the same circuits as the production models. It was installed in LTA-8 and used for the various tests. There was some overlap in production, but the major difference in the AGC we restored is the use of aluminum for the chassis instead of the lighter magnesium chassis in the flown computers.
The first man-rated lunar module (LTA-8) was used for testing on Earth. Astronauts practiced in this lunar module inside a giant vacuum chamber for over 48 hours, using the module for life support. (This is why a man-rated module was needed.) Since the module was never flown, this AGC didn't need to be rated for flight. This is also why LTA-8 is on display in Houston rather than being in space. The first video discussed this a bit more and details on how the lunar module was tested are here: spacecenter.org/attractions/starship-gallery/lunar-module-lta-8/
They man-rated it on earth, in a giant vacuum chamber in Houston. It not intended to fly. Good info on how they did it in this link: www.nasa.gov/feature/50-years-ago-two-critical-apollo-tests-in-houston .
About the power supplies - I heard the word oscillator. If this is a DC to DC power supply from1970 would you need an oscillator. Then I see on the 'scope that there is a square wave. There was also mention of the ripple (shown on the scope) being caused by the caps charging and discharging. All this points to this being an early switching power supply. Cool. True?
I'm curious what were the attempted PCB's manufactured from? I know that wood fiber PCBs which were used in the 1960s for consumer and some professional products were notorious for humidity problems.
@Jonas The Movie I would think Mica would be too fragile. I'm referencing thin multi-layer mica washers and power transistors insulators (TO-3) where slight flexing of the material can cause the layers to start separating.
Thanks And keep them long. The longer the better. Would be nice if less edited and 1 hr each. Love every video you do. Much love for large super computers
It brought a tear to my eye. My dad and I have worked on Americas space programs. My dad worked on Apollo. I have see an Apollo launch. I watched the Moon landing on a crappy black and white TV as a kid at summer camp.
Incredible! Thank you so very much for sharing this! True American engineering, back when they made things to last! These are prototype test boxes, that have suffered some corrosion due to age, but even without anything being replaced 50 years later they still fire right back up without issue. The gentleman that owns this stuff was very wise to realize the value of keeping such things when he convinced NASA too sell him the 2 tons of junk equipment. NASA has a very bad habit (still to this day) of discarding historical items. They left reels of Apollo film & magnetic data tapes sitting in cardboard boxes shoved into storage rooms in FL & TX for decades. Some of it was water damaged from sitting under a leaking roof, others turned to dust due to dryness & some simply thrown in a dumpster to clear out the space. What pisses me off is when people like this gentleman save these items, but them from NASA, store them, care for the & then go through the time & huge expense of restoring these literally priceless historical artifacts & then suddenly NASA wants it back. I have massively mad respect for each of these fine gentleman working on this project. What an amazing thing to be a part of! I'd love to be in that room just for a bit, to be near those devices (it's a space & computer geeks wet dream all in one!). You can tell this is a labor of love for these guys. It's awesome to see different generations working side by side on such a pivotal piece of history.
Encore aujourd'hui en 2024 c'est toujours aussi passionnant ! Ces personnes ont travaillé sur un morceau de l'histoire de la conquête spatiale avec simplicité et compétence, c'est incroyable. En regardant cette seconde partie de la vidéo, et comme ancien informaticien, je ne peux m’empêcher de penser avec admiration et respect à Margaret Hamilton qui a programmé, avec son équipe, certains des modules de ces nouveaux systèmes, c'était une première à l'époque. Un big merci à vous pour cette magnifique vidéo qui traversera, j'en suis certain, les années. Bravo et encore merci à tous, vous êtes de brillants et sympathiques ingénieurs !
This is a bit surreal. From how these units were discovered, and saved. To you guys opening everything up in a hotel room, and sharing it via technology that didn't even exist back when the modules were created.
Man, almost all of this stuff goes over my dumb, little head. But it’s amazing that you’re preserving this important piece of human history! Also, it’s amazing this power supply works perfectly after 50 years...they really knew how to build that stuff back then!
I look forward to subsequent parts of this series. I worked on a large, refrigerator sized computer (digitizer) in the early 1980s. It was a Burroughs machine. It had 753 circuit cards in its three backplanes. The cards contained discrete and DIP devices. All of it was interconnected with 1 to 3 level wire-wrap. I never added up the number of wire-wrap pins, but I would say it would come out to be around 75,000 of them. It was tried and true technology. Very stable and reliable for its day. In the seven years I worked on it we only had one wire-wrap failure. That was related to a wire going around a pin and the edge of the pin cut through the insulation causing an intermittent short.
Seeing the wire wrapped backplane brings back some great memories of my days as an IC design tech. I’ve lost track of how many miles of the stuff I went through back in the 90s, building breadboards for my engineers. All that beautiful silver-plated Kynar, now rotting in a landfill.
I am in absolute historical, nostalgic, nerd-heaven. Thank you for sharing this project with the world (even if it gets fewer views than an out-of-date fashion reviewer exhibiting 2nd hand swimsuits from her local resale shop). Unfortunately this is the world we live in... Thank God there are still some curious individuals with an IQ higher than 70 that share sure h brilliant cobtent with the few who wish to watch it.
50 years on and still making history! Thanks for letting us see how things are progressing.
This will get better and better with each episode. I can't wait to see the logic come to life, even if it's only partially operational. I'm really interested in seeing what kind of DIY interfaces you might come up with.
Hey Maxx, hope you're doing fine. Miss your videos, dude.
My father worked on the tracking systems for Apollo recovery, and was down on Cape Kennedy. This truly warms my heart.
The engineering and technology there is absolutely amazing and beautiful. Its absolutely amazing they could get so much power in a small package. Granted this was a very specialized computer designed for one thing unlike many of the time, the design and technology is just literally out of this world for the time!
This is roughly the same computing capability as the IBM 1800, a real-time mainframe of the same timeframe, but instead of taking multiple full sized racks and a lot of power, it fit in a small volume and used very little power.
@@carlclaunch793 I've been aware of this effort for a few years now but I've never actually taken a deep dive into the specifications of how computers of the era were built. How did they achieve that? Was the lower power consumption all thanks to the process used for fabricating the NOR gates it was built out of?
All the best to your entire team, I'm looking forward to seeing the AGC put through its paces!
@@sietuuba The low power was the cumulative effect of hundreds of design decisions and component choices. The compact design itself didn't require driving transmission lines between far flung parts of a mainframe computer of the era. Switching power supplies whereas mainframes of the era had big linear supplies. Doing cycle stealing to maintain timers and counters instead of having dedicated circuits for each one. ICs surface mounted on multilayer boards, plus cordwood construction and core rope memory all shrunk volume dramatically.
@@carlclaunch793 I can't imagine that is was super efficient with all the analog power supplies that stepped down all the voltages and the miles of wires connecting the components.
@@FrederickWalser Much more efficient than would appear initially.
The power supplies are switching design, not the usual analog supply of the time that burned off excess voltage as heat. Considerably higher efficiency.
The majority of the wires you see carry very low currents - signals at .8 to 4 volts to only a few transistor base-emitter junctions.
The power feeds used a pseudo bus-bar connecting groups of power pins down at the bottom of the pin, reducing resistance compared to point to point wiring.
Many of the connections were traces on the multi-layer boards which held the ICs - short distances and low resistance. Board to board connections passed over gold contacts, once again lowering resistance versus copper.
Being a detail oriented type of techie/geek/nerd, the more detail the more I like it, regardless of the length of the videos. Thanks for another great video, please keep them coming.
As an 40y-old fanboy of the AGC I'm full of admiration of these guys, doing it in a HOTEL ROOM, where others would be incapable doing it in a fully equipped laboratory. But the AGC is in awesome condition. Absolute unique specimen you got there. Just wow. I'm following Mike Stewart's work for some time too, he's THE expert and I'm full of respect for his efforts, I'd cook a meal for him every day, just call a day before you visit northern Germany, Mike ;) No, seriously, as a collector of old tech I'm convinced it's important to preserve it for the coming generations. Ask my daughter how often I told her how to use an old HP calculator in postfix notation (and why one should do this) or when I hang another picture of the Curta calculator in the hall... Just saying, there's a community around the world that's with you. All the time. Go on with your good work!
Thanks so much for the kind words! It means a lot. :)
Lots of engineering gets done in hotel rooms. BTDT, getting something ready for an urgent customer request a day before it was going to be installed. It’s a somewhat distraction-free setting.
if this is how computers were physically laid out and cased and labeled, I understand why the computer in 2001: A Space Odyssey looked like it did: it's like these gate modules, but made in a "futuristic" material ie plexiglas
I always thought that of Star Trek TNG and "isolinear chips". Reminds me of the flip chip modules used to construct older PDPs.
2001 is interesting for how hard it is to project what tech will look like in the future, even though there where experts hired to do just that and it was taken seriously.
The most telling for me is the shuttle flight going to the space station fitted out inside like it was an airliner doing a something mundane and the space station was on a scale and homogeneous design that is still science fiction in 2018. Then there is the manned mission to Jupiter that is just laughable.
Considering the movie was released in the first year of the Apollo space program and how rapidly that culminated in numerous manned missions to the moon the vision into the future
seem very realistic if not conservative for the investment in space exploration of that day.
Definitely inspired from large 1960's computers. Actually I think it is more inspired from something like the PDP-1: ruclips.net/video/1EWQYAfuMYw/видео.html .
I know this is years old but I had the exact same thought.
Considering how long Voyager 2's been working for, its not a doubt how well things were engineered for space, im almost positive this computer will work without too much hassle with the proper love and care it seems to be getting
I never in my life thought I would ever get to see such an amazing thing. Thank you so much!
I am currently taking a break from study for my Computer Engineering final exams. This is great motivation.
What an amazing piece of engineering. It’s all too easy to say this computer is so slow by today’s standards but I bet it would be more resilient than any general purpose computer in the environment it was designed for!
Yes, you can't put your standard home computer into space. You need to design for the high acceleration and the vibration that the rocket or spacecraft will experience. Oh and a software crash is totally unacceptable.
@@gorillaau Think about Apollo 12 and how the IBM IU just kept chugging away after the booster was hit by lightning 2x!!! A modern computer is toasted if the power flickers to hard!!!
If your computer locks up during launch your gonna have a bad day!!!
@@chrisjohnson4666oh totally. Five nines (99.999%) is not good enough. To paraphrase a petrol engine slogan, It has to work first time, every time.
The question is not "was it slower then todays computers?" - Yes, of course it was.
The real question is "Was it to slow for the tasks it had to do?" - And to this question the sound answer is: No, it wasn't. Although not by a large margin, it still was fast enough to do that.
@@kallewirsch2263 unless you wanted to could the 1201 and 1202 error codes as being due to insufficient available processing grunt.
Luv that you are showing this to the public great work everyone
That is freaking awesome. The quality put into this is amazing. I’m blown away. Keep them updates coming
For now, this all series could be resumed by "Trying to find something bronken on a 50yo space computer, damit it was working all along!". I will forever be amaze by Nasa engineer and reliability
Yes, this is no ordinary hardware. But you'll see we have at least one big bad fault.
I wouldn't say is just Nasa, I worked in several systems in the US Navy as an Interior Communications Tech and they're just as rugged as these ones. I realized most people haven't seen military grade systems, but believe me they're just as astonishing. Apollo systems are obviously on a top tier due to the nature of what it was supposed to do.
Every time someone says "spacecraft," I get a chill. I'm so used to that word only being used in the context of science fiction that hearing it here, in regards to an actual vehicle designed to fly beyond the atmosphere of Earth, just...I don't know how to express it. Induces a feeling of awe, I guess. Amazing.
I also had a bit of an "ah-ha!" moment with the digital to analog converters, since that's basically the same thing as in my HiFi rig!
I have to say I'm actually quite frustrated that I found these videos now and not in 3 months when the project was complete. I NEED MORE. Thank you for all your hard work. We appreciate the history being honored. THIS. IS. AWESOME!!!!
This project is perfect for long-format videos
I am so pleased you are getting this icon of history going!
It's fascinating to look at the size of some of the components and the technology applied within the AGC; only then to realize that we're actually at the end of 2018 and this was built 50 'kin years ago! 50 years...bloody hell, man!
The point-to-point wiring has me drooling and procrastinating in equal amounts - so without to do; I'll sit here in my wet patch and watch the next installment(s) :)
By observing how complicated it is to dive in and check these modules one can imagine how complicated and goal-driven the design process was in the first place. Everything in the Apollo program, not "just" the AGC, was state of the art and beyond at that time. The engines, the two types of spacecraft, the suits, the docking, the rovers, even the parachutes, everything. This is truly one if not THE most ambitious single effort human kind has made to reach one goal. And so often there were the right people in the right place at the right time. E.g., Apollo 12 with the lightning strikes and the "SCE to AUX" story. Or Apollo 11 with Armstrong at the controls in the final descent with no fuel left. Or Apollo 13. or or or. Each time I dig into something related to Apollo I think "This is unbelievable".
Watched part 1 last night, just watched part 2 and I’m totally blown away by how amazing this is.
I have no interest in this stuff but I’m glued to my screen!
Amazing stuff guys
Very interesting! The AGC is a masterpiece in design and construction.
Might be nice to look at components operating with a thermal camera, you could see quickly is anything was on the way to failure
Great video. Quite the leap in technology over the decades from the time of the TTY from you last video. Amazing to ponder. If there was ever any doubt, this confirms I don’t have the “right stuff.” If someone asked me to do a half-million mile round trip on that thing to the moon and back, I think I’d have to pass, regardless of the quality..
And to think Teletypes were still in widespread use at the time to relay the landing news - and probably to communicate with the recovery navy ships.
I worked on a mainframe computer line that into the 1970s was still using a KSR 33 as it's main console interface.
Fascinating and not long enough. This is so cool. Awesome to have the technology explained also. Thank you for posting this. The comments are really useful as well. Wish I was there though.
Mind blown. Not only because the computer but the insane accurate knowledge this guys have of this system. Probably even better than a Nasa Technician at the moment of repair or upgrade.
Thank you so much for sharing these videos! I'll be glued in front of youtube every time you post an update! This is my favorite computer of all time.
I need to say it out loud--Curious Marc RULES. Great channel!! I am always blown away by the crazy mechanical and digital computer restorations you take on--fearlessly--and always enjoy seeing the "Marc Army", Carl, Master Ken, etc. practicing their unbelievable tech kung fu. Glad to see you got this restoration gig, you deserve it, and it's all done with your usual dedication, determination, humility and humor. It's already fully ok to be a nerd (Bill Gates gave us that?) but seeing this sort of thing is icing on the cake. Keep on keeping on!
I'm so sorry i needed days to come down to watch your video......
Seriously, you are to me one of the holy people of 'archeology programmer' sort of people, generations to come will reference and learn from you..
THX for your awesome work and parts from the alto, the teletype and now the AGC. ;)
Hacky, from Undeloh, Lower Saxony, Germany, Europe, Sol 3 (Earth), Sol System...... ;)
Very cool! I work on embedded hardware and software, and the AGC is a remarkable machine. Can't wait to see more!
This is fabulous and essential work you are doing: there needs to be greater awareness of these computers because of the legacy they continue to leave.
Yes! Part 2 is here!
That backplane is beautiful and the way the power supplys were constructed is very nice.
It does kind of crack me up that every device used to test, measure, and document this restoration, is more powerful than the computer you are actually restoring. Probably even including the digital multimeter!
The new stuff is much more powerful, but also more fragile, both physically and operationally. I'll take the AGC over the newer stuff any day if I have to fly to the moon.
@@phil4826 What you say is not really true. You are comparing something from the past that was built specifically for reliability in an extreme environment to something from today that was built to be as cheap as possible and still do the job in a home environment. A well made modern cpu is actually an extremely reliable and durable component especially compared to this. It's everything attached to it, powering it, supporting it that is made to be cheap and not up to the task.
And there was plenty of cheap, fragile crap being built for home use back in the 50s and 60s too. I wouldn't want to fly to the moon if my life depended on the reliability of the cheapest radio found at Sears in the 60s. My cheap modern cell phone is both more powerful and more reliable than that.
Fortunately, neither our brave astronauts of the past nor us today have to do so. The stuff built for Apollo was not standard off the shelf electronics. Every piece was built to a higher standard and tested thoroughly. If those power supplies had off the shelf 60s capacitors in it, that thing would have smoked the minute they put power to it. And our astronauts of today don't go up with an off the shelf Dell laptop. They have systems designed, built and tested for that specific job and environment with a backup and another backup.
But if you just have to have a super reliable, nearly indestructible calculator with a lifetime warranty on it to help do your taxes, I would be happy to build you one. Just send me $100,000.00 US dollars and you will have it in a years time.
@@wingracer1614 Yup. Anyone who's been inside an "All American Five" radio can attest that consumer electronics have always been designed to be "good enough".
@@wingracer1614 he said hed take the old stuff over the new when going to the moon. you agreed but called him wrong. he wasnt talking about consumer electronics in the 60s
wingracer 16 ya u didn’t read his message properly bro
Never clicked so fast on a new video
This is so cool testing old tech and experiencing the out put from the device that you are play with...
Link to the backplane viewer: apolloguidance.computer/2003100_071/pins
I'm not a web developer, so please excuse the poor resizing capability and lack of mobile support!
Thank you so much for posting these videos, absolutely fascinating to see the technology used and realize the best minds at the time were working on these, and you can see it in the quality of the work! However, from a poor broadcast "engineer" who wishes he could understand the math behind this all, could you add a drop shadow to your CG text? Just a simple black border or drop shadow would do wonders against the gleaming bright metal work of the AGC. It's sometimes hard to read the text comments with white on white background etc. . Thanks a ton! Keep posting updates!
This will sound like a dumb comment but this reminds me of the computers some crazed lunatics have built in Minecraft. The similarity being that the "electronics" in Minecraft were never intended to build such a thing, it was only meant as a little curiosity for making simple switches yet when there is a possibility some people will always push things way beyond their limit. When I see these nests of intricate wires I don't just see the same kind of structure in MC computers but the same mindset.
Reverse engineering of the backplane! Amazing! Just out of curiosity! Something to put on your resume for applying at the CIA or any Chinese company.
An absolute marvel of human ingenuity!
Wow, wow, wow. Amazing work to everyone. Thank you for sharing!
Amazing work guys, Im super excited to see this up and running
Epic. Beautiful engineering. Fantastic video!
This is fascinating. My dad was a Saturn/Apollo physicist. Thanks for sharing.
Oh man, I can't wait for the next! Incredible.
Awesome just finished part 1 and 2 was just posted!
Saying this is 16-bit was like people saying the Nintendo 64 was 64-bit, "Sort of kind of not really" for pretty much similar reasons interestingly enough.
@11:35 so they had found out that using PCBs and soldered components in the Polaris missile was unreliable, so they didn't take any chances and went with wire wrap and welded components for the manned missions. That made me realize that the technology used in our cars today IS soldered components and PBCs... I guess (hope) there has been huge progress in the reliability of that technology since then ? lol !
Anyway, very enjoyable video as always. Thanks for sharing, Marc. BTW, 18 minutes is NOT long-ish at all with this kind of fascinating content (for the nerd in me anyway). It seemed like 5 minutes !
AlainHubert - Yes, it’s a lot more reliable now than it was in the ‘50s & ‘60s; however a welded connection is still better than a soldered one. Also, FWIW, the wire-wrapping process effectively creates a cold-weld at the corners of the pins where the sharp edge cuts thru the kynar insulation and digs into the wire - this is due to the tremendous tension that the wrapping turns are under. This tension also helps with vibration resistance.
The reason soldering is used instead of welding is a) it’s good enough most of the time; and b) it’s easier to do quickly in large volumes.
AlainHubert Of course there has been development. However, cars don't usually experience 5-10g, possible extreme heat ot cold (-150c-500c) :D They serve their purpose. I've seen a lot of failed stuff like failed joints on automotive stuff, but usually after more than a decade.
@@rkan2 You forget the 5G 50mm vibration in all 3 axes the missile boards had to endure during testing and use. Missile boards are still fully conformal coated in a flexible coat, then potted in a hard coat after testing, then potted into modules as final step because of the vibration. Soft coat absorbs stress and the hard potting speads the stress evenly. only has to work flawlessly once, but has to endure a lot of testing before that, and a lot of abuse.
This is awesome. Looking forward to future episodes
What a beautiful thing to witness!
Wow. I've got a lump in my throat. This is beautiful.
Thoroughly enjoying these videos. I recognize that there were many, many hours of study and preparation to manage the real thing as deftly as you are doing. Can't wait to see the next one.
Having spent close to two decades working on computers with 3 level wire-wrap backplanes, I can state that wire wrap backplanes are not necessarily 100% reliable. Those G-D machines could pull the wire tight around the pins when cornering a long wire, and the Kynar insulation would eventually cold flow and short to the cornering pin. Depending on what was on that pin, this could cause all manner of havoc, and could be darn hard to locate.
It looks like the wiring in that backplane area is very open and I didn't immediately see any tight cornering, but it is certainly something I'd carefully look for.
The computer design review brings up the cold-flow problem, and it was taken into deep consideration. Here's an except from the review book:
The simple solution to this problem would be to obtain a wire with a greater "cold flow" resistance. However, these tougher wires present problems to the wire-wrap machine. They do not strip easily and have a mechanical memory which makes them spring up from the wire wrapping surface. It is, therefore, necessary to make a compromise between the toughness of the insulation and the ability of the wire-wrap machine to perform the wire-wrap operation successfully.
After more than a year of search for an acceptable wire, Raytheon has finally located a vendor who can produce a wire which is both tough and "wrappable."
@@mikestewart8928 Simple solution I saw was to leave some slack in the wire every time, so that it was not pulled flat on pins. Then they took the runs and laced them together with lacing twine in a loose loom. Lot of manual labour extra, but as they only made around 50 units not going to add too much cost wise.
IBM had a control unit, the 2821, that suffered from lots of cold flow problems which required rewiring the backplanes for all the wires that had been pulled tight around pins. Most other machines they built with wire-wrap had used procedural steps during wrap to avoid this problem
It could be a real problem. I worked on a machine at Burroughs that used wirewrap coax for all of the signal connections in the processor in the first prototype machine. The cold flow problems with that were so bad that we went to wirewrap twisted pair for the second prototype and the real machines.
On the other hand: Every single piece of equipment used went through a long series of testing before the launch. Especially with the electronics NASA tried to limit down the possibilities of random errors to as much as possible.
Eg. there was the concern that the chips might fail. Thus random samples of chips were taken and drowned into a liquid (I am not 100% sure but I think I remember it was Freon). The chips were weighted before and after that procedure. If the weight differed it meant that the casing was not 100% tight. If just 1 chip failed that test, the whole batch was thrown away.
I had no idea there was 2 dozen people watching and clapping at the power supply. Fitting that many people in the corner of a hotel room out of sight is a huge feat on it's own.
must have been room service and maid service
I was wondering if I missed something when I jumped ahead in the video and all of a sudden I heard clapping.
Excellent! I would be interested to have a PDF of the schematics you show in the video!
The raw scans of Don Eyles's AGC Handbook can be found here: archive.org/details/agc_handbook_jp2
They have been cataloged and indexed, along with Eldon's schematics, here: www.ibiblio.org/apollo/ElectroMechanical.html
Added a link to the schematics below the description.
They could’ve used my iPhone back in the day. My dad was management in TRW and helped design the redundant guidance systems in Apollo. They used it on ONE flight.
These images are priceless!
This is utterly fascinating...
While going thru the "AGC Handbook" noticed that the Signal Wiring diagrams were missing for modules 13 and 23 and a few other miscellaneous pages. Was the document you received incomplete? Thx for updating that backplane link. I probably missed it but is there a wiring diagram that covers the backplane?
Yeah, there were sadly a few missing pages in the Handbook. The good news is that we've since recovered the missing pages, as well as many many more revisions of the schematics, from the National Archives (in higher quality too!) The AGC schematics are split between boxes 459 and 460:
virtualagc.github.io/virtualagc/AgcDrawingIndexBox459.html
virtualagc.github.io/virtualagc/AgcDrawingIndexBox460.html
Search for the drawing numbers on those pages and you should be able to find anything missing from the Handbook.
There sadly wasn't a wiring diagram that covers the backplane... and (amazingly) not even one that covers the internal inter-tray connectors. All of that information was stored in a physical book called the "AGC Wirelist", of which there were very few copies. Instead of digitizing those onto microfilm, they instead scanned little notices that said to get a hard copy from MIT (here's an example: archive.org/stream/apertureCardBox459NARASW_images/apertureCardBox459NARASW#page/n1691/mode/1up )
It's not clear at the moment if MIT still has any copies of the wirelists, but hopefully they do, and finding out for certain has been on my list of things to do.
Is there a complete set of schematics available anywhere. I am in the process of building a hardware gate accurate AGC but I only have schematics A01 - A29. By the way I tried the backplane tool but could not get it to work. It just showed a fixed image.
The raw scans of Don Eyles's AGC Handbook can be found here: archive.org/details/agc_handbook_jp2
They have been cataloged and indexed, along with Eldon's schematics, here: www.ibiblio.org/apollo/ElectroMechanical.html
Logic circuit built from wirewraped NOR gates. Impressive!
wow, wish samsung would make power supplies for their TVs with that level of reliability !
Lol
They would be happy to if you would actually buy it but would you pay triple the price if the only feature it had over the cheaper versions was a better power supply? You'd be better off buying the cheap one and when it dies, buy another one.
@@wingracer1614 I'd rather buy one tv for the next 20 or so years, that's much less ressources used
I mean, sure, a TV built to this standard would cost probably $100,000, so just be prepared to open your wallet really, really deep!
Great result with those supplies!
How do you weld a wire to a turret terminal? I'm guessing some kind of resistance welding but that's a tight space to work in.
What is Mike's background? He looks young but the work he has done with the AGC emulation looks like a lifetime's effort =)
I'm an embedded software engineer, and have been doing flight software for various spacecraft from various newspace companies for the past ~8 years. My AGC hobby project started as an on-and-off thing of a lot of thinking and reading about five years ago, and a bit over three years ago I started in on an effort to simulate and then build a gate-accurate replica from Eldon's schematics. Roughly a year after that the project began to grow rapidly in scope, after I got in touch with Don Eyles who has, over the past couple of years, contributed for scanning over half of the program listings we have today, as well as all of the Tray B schematics for the AGC itself. And here we are today... :)
It sure does feel like a lifetime, though, haha.
What are some of the problems or difficulties you have run into so far?
I love watching your video's, thanks for sharing my friend!😉
Fantastic stuff. Any plans once it's up and tested to hook it up to emulators of the peripherals, load up the software, and run the whole shebang in a closed loop? Maybe even do whole simulated missions complete with failure scenarios? I'd tune into that show.
Yes, we hope to do something like that. It depends on how well everything works and how ambitious we are. The peripherals do I/O with pulses at weird levels, so we'd need to make some emulation interfaces. One problem with doing a simulated mission is that it would happen in real-time, so it would be very slow to perform unless we can jump to the parts where something happens.
Holy shit. Did I just come across this video. Thank you almighty AI god.
I sat through a lecture a few years ago about PWM. It was at a Seattle Makers thingy. Lots of remote control "robots" on display. Did you know PWM was invented "around the turn of the millennium?" Hilarious.
That's so much fun, made me smile ear to ear. Not sure if it's legal doing it in a hotel room seems kind of odd.
Don't forget to change all the capacitors in the power supply, i'm joking :-)
The RUclips comments wouldn't be complete without a suggestion to change all the capacitors :-)
Yahhhh new video! thanks!
Don't worry about making them too long! Longer videos are better!
these guys are my new heroes ..way much cooler than ghostbusters!!
Man, those things were built like tanks, weren't they?
Like spaceships.
@@CuriousMarc Built to military specs. Space Tanks.
how did they weld electronic components without damaging them??
They used spot welding; basically a quick zap on each joint. It wasn't like a guy with an arc welder joining steel beams together. Here's a documentary on the assembly of the AGC showing the welding taking place: ruclips.net/video/ndvmFlg1WmE/видео.html
I love this... a huge thank You!!
Again... thank you, thank you, thank you!
The closest we had to the internet back in those days was gaming Ma Bell for a free telephone call.
Did you have to solder wires to accomplish the strapping of the power supplies? Can't see what they are soldered to. Fascinating technology! Seems like this computer connected directly to lots of external things. I had thought it would be used more for manual calculations, but seems like it was doing real time control based on sensor inputs. Thanks for making these videos.
Some of the pins on the bottom of the module accomplish the configuration. ON the backplane where the power supply is plugged, wires are routed to the pins in the pattern needed to select either 4V or 14V
We used some jumper wires we crimped with mostly-compatible terminals to do the strapping.
Yes, the guidance computer is doing real-time control. It receives data from the gyroscopes, accelerometers and radar to determine the spacecraft's position. Meanwhile, it controls those devices as well as the engine and RCS jets to keep the spacecraft on course. There's a diagram of its inputs and outputs here: archive.org/details/acelectroniclmma00acel/page/n165
I'm slightly confused. If this came out of the first man-rated lunar module, then why was it populated with a non flight-rated unit, and why is it being repeated that it was just a development unit? Unfortunately Google isn't much help right now, googling anything about "first" and "apollo" or "lunar module" just yields results about Ryan Gosling.
Grumman produced a set of test versions of the LM as well as the machines destined for the specific missions such as Apollo 10, 11, etc. These earlier machines were called Lunar module Test Articles and our AGC was in LTA-8. LTA-8 was used at the Manned Spacecraft Center (now Johnson center) in the thermal/vacuum chambers.
These chambers were the only ones that NASA owned which were large enough to hold an LM, CM, or CM+SM. They would subject the spacecraft to temperatures ranging from hundreds below zero to hundreds above, while in a vacuum, to verify that the spacecraft interiors would remain habitable. Astronauts sat in the spacecraft during the tests, simulating a mission.
LTA-8 was basically the same design as the production LMs used for manned flights, but was subjected to abuse on the ground to ensure that the production ones were safe for flight. heroicrelics.org/space-ctr/lta-8/index.html
Our AGC was the 14th of 15 prototypes before the first flight models were produced, although it has the same circuits as the production models. It was installed in LTA-8 and used for the various tests. There was some overlap in production, but the major difference in the AGC we restored is the use of aluminum for the chassis instead of the lighter magnesium chassis in the flown computers.
The first man-rated lunar module (LTA-8) was used for testing on Earth. Astronauts practiced in this lunar module inside a giant vacuum chamber for over 48 hours, using the module for life support. (This is why a man-rated module was needed.) Since the module was never flown, this AGC didn't need to be rated for flight. This is also why LTA-8 is on display in Houston rather than being in space. The first video discussed this a bit more and details on how the lunar module was tested are here: spacecenter.org/attractions/starship-gallery/lunar-module-lta-8/
They man-rated it on earth, in a giant vacuum chamber in Houston. It not intended to fly. Good info on how they did it in this link: www.nasa.gov/feature/50-years-ago-two-critical-apollo-tests-in-houston .
1:34 Anyone remember when hal 9000 got disconnected and all his logic memory was taken out, Now I know where stanley kubrick got his ideas from.
Great work!!! ... you are very lucky people, I would love to work with you there.
Loving this!
Good job, love this project!!
About the power supplies - I heard the word oscillator. If this is a DC to DC power supply from1970 would you need an oscillator. Then I see on the 'scope that there is a square wave. There was also mention of the ripple (shown on the scope) being caused by the caps charging and discharging. All this points to this being an early switching power supply. Cool. True?
I'm curious what were the attempted PCB's manufactured from? I know that wood fiber PCBs which were used in the 1960s for consumer and some professional products were notorious for humidity problems.
@Jonas The Movie I would think Mica would be too fragile. I'm referencing thin multi-layer mica washers and power transistors insulators (TO-3) where slight flexing of the material can cause the layers to start separating.
Thanks
And keep them long. The longer the better. Would be nice if less edited and 1 hr each. Love every video you do. Much love for large super computers
Thats crazy; so much wires 😮
That one dude never ever blinks.