A couple weeks ago I was at a party where out of the corner of my eye I noticed what looked like a giant phone book sitting open on a table. It was printed with perforated green and white paper bound in a binder who’s cover looked a little worse for the wear. I had closer look with my friend James Kinsey. What we read was astonishing; Program 63, 64, 65, lunar descent and landing. Error codes 1201, 1202. Comments printed in the code, code segments hastily circled with pen. Was this what we thought we were looking at? And who brings this to a party?
This is a project that is about a year and a half in the making, but [Fran] is finally digging into the most iconic part of the Apollo Guidance Computer and building the most accurate reproduction DSKY ever.
The Apollo Guidance Computer was a masterpiece of engineering and is frequently cited as the beginning of the computer revolution, but it didn’t really look that interesting – it looks like a vastly overbuilt server blade, really. When everyone thinks about the Apollo Guidance Computer, they think about the DSKY, the glowey keypad interface seen in the blockbuster hit Apollo 13 and the oddly accurate disappointment of Apollo 18. It’s the part of the Apollo Guidance Computer the Apollo astronauts actually interacted with, and has become the icon of the strange, early digital computers developed for NASA in the 60s.
There are a few modern DSKY replicas, but all of them are exceedingly anachronistic; all of these reproductions use seven-segment LEDs, something that didn’t exist in the 1960s. A true reproduction DSKY would use custom electroluminescent displays. These EL segments are powered by AC, and transistors back then were terrible, leading to another design choice – those EL segments were turned on and off by relays. It’s all completely crazy, and aerospace equipment to boot.
Because of the custom design and engineering choices that seem insane to the modern eye, there isn’t much in the way of documentation when it comes to making a reproduction DSKY. This is where [Fran] tapped a few of the contacts her historical deconstruction cred earned when she reverse engineered a Saturn V Launch Vehicle Digital Computer to call upon anyone who would have access to a real Apollo-era DSKY.
The first contact was the Kansas Cosmosphere who was kind enough to send extremely detailed photographs of the DSKYs in their archives. It would have been extremely nice to have old documentation made when the DSKYs were rolling off the assembly line, but that information is locked away in a file cabinet owned by Raytheon.
[Fran] got a break when she was contacted by curators at the National Air and Space Museum’s Garber facility who invited her down to DC. She was given the grand tour, including the most elusive aircraft in the museum’s collection, the Ho 229, the dual-turbojet Nazi flying wing. At the Garber facility, [Fran] received permission to take apart two DSKYs.
The main focus of [Fran]’s expedition to the Air and Space Museum was to figure out how the EL displays were constructed. The EL displays that exist today are completely transparent when turned off because of the development of transparent conductors.
The EL displays in the DSKY were based on earlier night lights manufactured by Sylvania. After looking at a few interesting items that included Gemini hardware and early DSKYs, this sort of construction was confirmed.
With a lot of pictures, a lot of measurements, a lot of CAD work, and some extremely tedious work, [Fran] was able to create the definitive reference for DSKY display elements. There are 154 separate switchable element in the display, all controlled by relays. These elements are not multiplexed; every element can be turned on and off individually.
Figuring out how the elements were put together was only one part of [Fran]’s research. Another goal was to figure out the electrical connections between the display and the rest of the DSKY. There, [Fran] found 160 gold pins in a custom socket. It’s bizarre, and more like a PGA socket than like the backplane connector [Fran] found in the Saturn V computer.
Even though [Fran]’s research was mostly on the EL panel inside the display, she did get a few more insights with her time with the DSKYs. The buttons are fantastic, and the best keys she’d ever used. This is just part one of what will be an incredibly involved project, and we’re looking forward to what [Fran] looks into next.
[Fran] has already made a name for herself in some retro cool historical aviation and computer circles by tearing down a flight-ready spare of a Saturn V launch vehicle digital computer, the computer that was responsible for getting all flights to the moon into low earth orbit. Now she’s ready for another project, and again, this is something that hasn’t been done in 40 years. She’s building a DSKY, the control panel for the Apollo Guidance Computer
The Apollo Guidance Computer is a well-documented piece of computing history, with homebrew versions all over the web. The DSKY is only one small part of the AGC, but it is by far the most famous module. Being the only user interface for the AGC, it’s the only part of the AGC that gets all the screen time in Apollo 13, the travesty on BluRay that was Apollo 18, and is the only device that bears any physical resemblance to its real-life counterpart in a number of AGC simulators.
That’s not to say DSKY builds haven’t been attempted before; there are a few out there using LEDs and off-the-shelf buttons for the build, but the DSKY from the mid-60s is much, much cooler than a bunch of LEDs and light pipes. The eery green numbers are actually EL displays. Guess how those displays are controlled? Relays. It’s a masterpiece of technology, made even more impressive in that the folks at MIT who built the thing didn’t have anything better to build the display with.
Because of her deconstruction efforts with the Saturn V LVDC, [Fran] was invited down to the National Air and Space museum in the middle of Washington DC. There, she saw everyones favorite ugliest spacecraft, the Apollo LEM, along with an incredible assortment of paraphernalia from aviation history. The Wright Flyer – yes, the original one – is hanging from the ceiling next to the Spirit of St. Louis, and X-15 rocket plane, right above the command module Columbia from Apollo 11. Copies of probes currently rolling over Mars are on display, and you can walk through a training model of Skylab. If you’ve never been, spend half a day there, then take the metro out to the Udvar-Hazy center, where you’ll find all the stuff they couldn’t fit in the downtown collection like a Space Shuttle and a Concorde.
This is only the first part of [Fran]’s vlog documenting the construction of a copy of the DSKY, and we haven’t even seen the inner guts of the most famous part of the AGC yet. She’s been working on this for a while now, and there’s no doubt she’ll finish the job and come up with the best replica of a DSKY ever.
[Fran] has been researching the Saturn V Launch Vehicle Digital Computer – the computer that flew all the Apollo flights into orbit and onwards towards the moon – for a while now. Even though she’s prodded parts of the LVDC with x-rays and multimeters, this is the first time she’s committed to a little destructive testing.
After [Fran] took a flight-ready LVDC spare to the dentist’s office for x-raying and did an amazing amount of research on this artifact from the digital past, there was only so much she could learn without prying apart a few of these small, strange chip packages. Not wanting to destroy her vintage LVDC board, she somehow found another LVDC board for destructive reverse engineering.
This new circuit board was a bit different from the piece in her collection. Instead of the chip leads being soldered, these were welded on, much to the chagrin of [Fran] and her desoldering attempts. After removing one of these chips from the board, she discovered they were potted making any visual inspection a little difficult.
While [Fran]’s attempts at reverse engineering the computer for a Saturn V were a bit unsuccessful, we’ve got to hand it to her for getting this far; it’s very, very likely the tech behind the LVDC was descended from ICBMs and would thus be classified. Documenting the other computer used in every Apollo launch is an impressive feat on its own, and reverse engineering it from actual hardware, well, we can’t think of anything cooler.