The guts of NASA’s pioneering Apollo computer were handwoven like a quilt

Image result for The guts of NASA’s pioneering Apollo computer were handwoven like a quiltThis is the 14th in an exclusive series of 50 articles, one published each day until July 20, exploring the 50th anniversary of the first-ever Moon landing. You can check out 50 Days to the Moon here every day. 

When the sophisticated electric car that Americans flew to the Moon set out across the lunar surface, it rolled on wheels that had an interior of titanium and whose tire treads were made of wire mesh. The treads flexed to allow the lunar rover to race across the Moon’s dusty surface; the treads let some of the dirt into the tire, and then as it rolled around, the mesh flexed open and the dust simply fell out.

It was a brilliant solution to keeping traction in a slippery, gritty surface in one-sixth gravity. And the mesh that made it possible? It was made of piano wire—woven by hand.

When the Apollo spacecraft floated down to the Pacific Ocean on three distinctive orange and white parachutes, each one 83.5 feet across and 7,200 square feet of material—bigger than the floor space in two homes—all the sewing was done by hand, by women at black Singer sewing machines. Each chute required 2 million individual stitches to assemble.

And then there was the Apollo flight computer. In two previous installments of #50DaysToTheMoon, we’ve explained how astonishingly advanced the technology of the Apollo computer was, and how MIT’s pioneering use of integrated circuits in the spaceship computer laid the foundation of the digital age.

The Apollo computer, like the lunar rover and the parachutes, was cutting-edge technology for the era, so much so that it, too, leaped ahead of the ability to manufacture it in an automated way.

The technology, tools, and materials required to fly to the Moon in the 1960s often stretched the limits of what was available, and once it was invented, actually manufacturing it for spaceflight often proved even harder. But here’s the thing: That didn’t slow anyone down. If the lunar rover wheels had to be woven by hand, so be it.

Core rope memory from the Apollo Guidance Computer. [Photo: NASA/Wiki Commons]

If the internals of the Apollo computer itself needed to be hand sewn, so be it.

And so, like the lunar rover wheels and the parachutes, the circuits and programs of the Apollo flight computers were also woven by hand, by women at a Raytheon factory in Waltham, Massachusetts. They sat at sophisticated looms, using long needles with wire attached to them instead of thread, carefully weaving the wiring that was the programming of the computers.

The software was, in fact, hardware.

It was an astonishing process that was tedious yet required absolute attention and precision. Every single 1 and 0 in the computer’s memory required a wire in exactly the right place. A single mis-wired strand meant the computer’s programs wouldn’t work properly—and might fail at some critical, potentially disastrous moment.

The women who did this work had, in fact, been recruited from nearby textile factories. The memory holding the programming for a single Apollo flight computer—for the entire mission—was a total of just 73 kB, far less than the memory many emails require today. Yet it took dozens of women in Waltham eight weeks to assemble it, meticulously, by hand.

Raytheon set up a dedicated group just to weave Apollo computer memory, and the company found out how just how hard the task was. In the mid-1960s, there was a strike at the Waltham factory, and supervisors and managers tried to do the work the women had been doing. When the strike was over, every module the supervisors had wired was scrapped.

This kind of computer memory was called “rope core memory,” because it ended up looking like collections of densely bundled ropes made of wires. For Apollo, it had two critical advantages: It was impervious to damage. It couldn’t be accidentally erased or altered by an astronaut or a technician, or by an electrical problem. The software to fly to the Moon was hardwired into the computer. (In a frightful test of this, Apollo 12 was launched into a bank of rain clouds and was promptly struck by lightning twice in the space of 16 seconds just after leaving the launch pad at Cape Kennedy; the spaceship’s flight computers were undamaged.)

Just as important, in the mid-1960s, in the transition from vacuum tubes to transistors, and from transistors to early computer chips, handwoven rope core memory was far and away the densest computer memory available. It packed between 10 and 100 times more kilobytes into the available weight and space than any other kind of memory. On spaceships where keeping the weight down was a constant battle, that was indispensable.

The only problem was that it required an eight-week manufacturing lead time. With testing and then moving the computers to the spacecraft at the Cape, that meant a flight’s software had to be finished and perfect many months before the flight. Even in the 1960s, programmers wanted to keep tweaking right to the last moment. When they learned that they had to be done months in advance, says MIT’s Ramon Alonso, who helped invent the programming logic for the Apollo computer, “The reaction was, ‘What? I can’t walk up to the launch pad and change whatever the program is like?’ ”

The Apollo guidance computer was the first operational computer to use rope-core memory. By the time U.S. lunar modules starting landing on the Moon, the technology had moved on to the much easier to handle—and much cheaper—computer chips and floppy disks that dominated computing for the next several decades.

So the Apollo flight computer was not only the first computer of any significance to use handwoven rope core memory. It was also the last.

Charles Fishman, who has written for Fast Company since its inception, has spent the last four years researching and writing One Giant Leap, a book about how it took 400,000 people, 20,000 companies, and one federal government to get 27 people to the Moon. (You can order it here.)

For each of the next 50 days, we’ll be posting a new story from Fishman—one you’ve likely never heard before—about the first effort to get to the Moon that illuminates both the historical effort and the current ones. New posts will appear here daily as well as be distributed via Fast Company’s social media. (Follow along at #50DaysToTheMoon.)


Author: Ayaan