PROLOGUE — LM-1

January 22, 1968. It's a cold night in Cambridge.

The Lunar Module is spaceborne, launched four hours ago from temperate Cape Canaveral on a Saturn 1B booster. This is LM-1, later designated Apollo 5. It is the LM's first flight, a six-hour solo performance in low Earth orbit.

The Lunar Module is designed to carry two astronauts from lunar orbit to the surface of the Moon, support them while they explore, carry them back to orbit, and rendezvous with the Command Module that will return them to Earth. The LM is the first pure space craft, the first to disregard the requirements of atmospheric flight. There is no heat shield — no possibility of surviving reentry. That is why tonight the LM is flying without a crew. Commands will be sent remotely from Mission Control at the Manned Spacecraft Center in Houston.

Ungainly to some, distrusted by others because of the problems that bedeviled her construction, to us the LM is beautiful. We imagine her movements in orbit. An invisible puff from the maneuvering jets, she spins smoothly. Sunlight catches her delicate skin. Another quick puff and she stops on her mark. She is the prima ballerina and we know her intimately because we built her brain, the onboard guidance computer, and wrote the computer program, called SUNBURST, that is transforming her tin and transistors into an almost-living being.

Twenty men and women have crowded into a small room on the third floor of one of the nondescript buildings that house the MIT Instrumentation Laboratory. A squawk box connects us to mission control in Houston.

Next to me stands Allan Klumpp, who worked out the guidance equations that I programmed in the onboard guidance computer. It is our first flight too, our first chance to see our work in action. When I joined the Laboratory, 18 months ago, fresh out of college, this day seemed unimaginably far in the future. Now it is upon us.

It is also our first chance to screw up.

The LM will fire its descent engine twice. The guidance programs that Allan and I wrote will control the second burn. The same equation that is intended to guide the LM to a soft landing at a preselected spot on the Moon will be used tonight to satisfy target conditions just as stringently specified — but they represent a spot in empty space. After that the descent stage will be jettisoned and the ascent engine will fire to simulate the return to orbit from the lunar surface.

But before any of that comes a 38-second firing of the descent engine to simulate the burn that will place the LM in the right orbit for the start of the landing. The guidance computer is in Mission Phase 9. The LM is in the correct orientation for the burn. The countdown begins. If there were astronauts onboard, they would see the dwindling number of seconds until the time of ignition — TIG, pronounced “tig” — in the computer's display.

At TIG minus 30 seconds, powered-flight navigation begins. When in orbit or coasting between the Earth and the Moon navigation is leisurely, relying on computations and the occasional star sighting. For powered flight the accelerometers have to be read every two seconds, velocity updated, position recalculated — and that is just the beginning of it.

At TIG minus 7.5 seconds, attitude thrusters fire downwards. This is the ullage burn. It pushes fuel and oxidizer to the bottoms of their tanks to eliminate the bubbles that form in zero-G. The word is a brewer's term for the open space at the top of a keg of beer.

At zero, the Descent Propulsion System — DPS, pronounced “ dips ” — will come to life. We lean closer to the squawk box.

From Houston we hear “DPS on.”

Several seconds pass, far too few. “DPS off.”

Huh? But soon we begin to understand what happened. The Delta-V Monitor — a small part of the computer program, written by me, whose purpose was to monitor the velocity change imparted by the engine — had sent an off command because it thought the engine had failed. To give the engine time to come up to thrust the monitor checked three times at two-second intervals. But this time, on the third check, the engine still had not registered enough velocity change. The burn was over before it started.

Unknown to us, leak tests on the control valves between the fuel manifold and the descent engine had caused concern. Premature entry of fuel would have been dangerous. The decision was made just before launch, on the recommendation of the engine manufacturer TRW, to delay the pressurization of the propellant tanks. In fact, the signal to activate the explosive valves that released the helium that propelled the propellant was not sent until about 1.3 seconds after our software issued the engine-on command. The engine was slow to start because the fuel pump was activated late. We could have adjusted how long the monitor waited before it concluded that something was wrong with the engine — if we had known.

If NASA had tried again it would have worked because the manifolds were now charged, but the software was less flexible than it would be for the manned missions, and perhaps mission control was still honing the art and tenacity it later exhibited. In the NASA film about the mission you can hear flight director Gene Kranz tuning up his famous voice of command.

Houston turned off the onboard computer. The engine firings were performed under ground control. The LM passed her audition. Such a beautiful dancer, even with her brain turned off.

The mission was called a success but that was no consolation. We hung our heads in disappointment, and endured a public reaction that blamed the computer software, because, after all, the computer had issued the engine-off command. The mission report blamed the “slower than normal thrust buildup” and concluded that “the guidance system and the descent engine functioned as designed,” but the movie NASA made about the mission contained the statement:

The inability of an onboard computer to cope with a programming error was overcome by the infinitely more flexible mind of man.

With that perception hanging over us, Allan and I would have to wait 18 months more, less two days, before our programs would get another chance to guide the Lunar Module.

FOREWORD BY ASTRONAUT DAVID R. SCOTT 

I have often been asked, “What is the most dangerous part of walking on the Moon?”

My answer is, without hesitation, “Landing on the Moon.”

The Lunar Module must slow from about 3,800 miles per hour to a level hover at 100 feet above an unknown surface and descend through thick dust to a soft touchdown amongst craters, boulders and rocks that have never been seen on any photograph. The margin for error is essentially nil. There is no assistance from Mission Control; there is no back-up computer; and only 30 seconds of propellant reserve is available when the decision must be made to either land and shut down the engine, or command an abort back to lunar orbit. There will never be another chance to try again or to return to the Moon.

During the most critical phase of the landing, the LM must be controlled internally from 50,000 feet to touchdown through a single, very small, and very slow, computer, using software programs that were written, by hand, by a single engineer. That software must be absolutely infallible. When the guidance software for the lunar landing was to be developed, this daunting task was assigned to an engineer who had no formal computer training (none available) and who was only 23 years old! That engineer was Don Eyles, and the software he wrote was part of a program called Luminary.

Eyles was a member of an extended family of 400,000 dedicated engineers, pilots, scientists, managers and other essential members of a remarkable team that repeatedly relied on a single individual to invent, conceive, or imagine one of the vast number of essential elements of the mission that had to work. This amazing feat of the Apollo family was achieved during the 1960s, a period of national and internationalconflict, confusion, turbulence, and disappointment. And during this difficult period, and during Apollo, a contrast of cultures was blended into this remarkable family with no (or very little) conflict, confusion, or turbulence. As history will show, time and time again, members of this diverse and often culturally-opposed family stepped up to a challenge that seemed impossibleÑbut theirs was success after success, nearly beyond imagination.

In this book, Don Eyles has described his life and times among this family during this remarkable period. His descriptions are concise, coherent, and complete — the book is both entertaining and educational. We learn about the technology — computers, simulators, procedures and techniques. We learn about the people — throughout the book, Eyles relates his interactions with notable astronauts and NASA managers, and his impressions of their characters and capabilities. And we learn about the cultures — how he was assigned his tasks, how he worked night after night, and even how he might otherwise have gone to Woodstock.

Eyles highlights another unique aspect of the Apollo culture, a management style that facilitated open communication among astronauts, engineers, and scientists. He spent many hours in simulators with astronauts. He obtained an understanding of the demands and challenges of spaceflight and learned how to formulate complex computer instructions for the crews to use during normal and emergency flight operations. Concurrently, astronauts became familiar with the challenges of writing effective software for our small, slow computer. This teamwork was typical of Apollo and it marks Apollo as a major milestone in productive human relationships among culturally and intellectually diverse individuals.

Among his other achievements, Apollo 14 brought Eyles a special notoriety — including an invitation to the White House. Less than four hours prior to the planned descent from lunar orbit, the bright red Abort button on the LM's instrument panel illuminated and began sending a spurious signal to the computer that would have aborted Alan Shepard and Edgar Mitchell's landing at the moment the descent engine was ignited. Eyles had written the code that monitored that signal. At 1:00 in the morning he was called on to attempt to write a workaround for the problem, with no allowance for error. In less than two hours, not only did he write the workaround, but it was verified in simulators at both MIT and Houston, and then read up to the crew.

That clever workaround simply changed a few registers, first to fool the abort monitor into thinking that an abort was already in progress, and then to clean up afterward so that the landing could continue unaffected. The procedure required inserting 61 precise and sequential keystrokes on the computer keyboard (the DSKY) under severe time pressure. Mitchell executed the procedure flawlessly and it worked perfectly. Apollo 14 landed at its target point on time, and the Apollo program once again proceeded without a major setback.

An Apollo 14 failure, coming after the near-tragic failure of Apollo 13, would very likely have doomed my mission, Apollo 15, to a paper exercise. (Apollos 16 and 17 were already facing cancellation.) Eyles solved that problem — and thank goodness he did. He gave all 400,000 members of the Apollo family an opportunity to finish the Apollo program successfully and with pride and satisfaction — an opportunity to complete the history of the first humans on the Moon without another failure. In this remarkable book, Don Eyles gives the rest of us an opportunity to share and learn about a major part of this history. Read and enjoy.

David R. Scott
Commander, Apollo 15
Los Angeles March 2015