Apollo Expeditions to the Moon



Apollo in January 1967 was adjudged almost ready for its first manned flight in Earth orbit. And then disaster. A routine test of Apollo on the launching pad at Cape Kennedy. Three astronauts - Grissom, White, and Chaffee - in their spacesuits in a 100-percent oxygen environment. A tiny spark, perhaps a short circuit in the wiring. It was all over in a matter of seconds. Yet it would be 21 months before Apollo would again be ready to fly.

By April 1967, when I was given the Apollo spacecraft job, an investigation board had completed most of its work. The board was not able to pinpoint the exact cause of the fire, but this only made matters worse bccause it meant that there were probably flaws in several areas of the spacecraft. These included the cabin environment on the launch pad, the amount of combustible material in the spacecraft, and perhaps most important, the control (or lack of control) of changes.

Apollo would fly in space with a pure oxygen atmosphere at 5 psi (pounds per square inch), about one-third the pressure of the air we breathe. But on the launching pad, Apollo used pure oxygen at 16 psi, slightly above the pressure of the outside air. Now it happens that in oxygen at 5 psi things will generally burn pretty much as they do in air at normal pressures. But in 16 psi oxygen most nonmetallic materials will burn explosively; even steel can be set on fire. Mistake number one: Incredible as it may sound in hindsight, we had all been blind to this problem. In spite of all the care, all the checks and balances, all the "what happens if's", we had overlooked the hazard on the launching pad.

A photo of large volume of books on bookshelves
The pedigree of just one Apollo spacecraft took this many books. A mind-numbing degree of documentation contributed to reliability, safety, and success. lf one batch of one alloy in one part was found to be faulty, for example, a search could show if the bad material had found its way into other spacecraft, to lie in wait there.

A photo of astronaut Wally Schirra inspecting the new hatch of the capsule
Inspecting the new hatch, Wally Schirra makes sure his crew cannot be trapped as was the crew that died in the terrible Apollo spacecraft fire. Opening outward (to swing freely if pressure built up inside), the new hatch had to be much sturdier than the old inward-opening one. The complicated latch sealed against tiny leaks but allowed very rapid release.

Most nonmetallic things will burn - even in air or 5 psi oxygen - unless they are specially formulated or treated. Somehow, over the years of development and test, too many nonmetals had crept into Apollo. The cabin was full of velcro cloth, a sort of space-age baling wire, to help astronauts store and attach their gear and checklists. There were paper books and checklists, a special kind of plastic netting to provide more storage space, and the spacesuits themselves, made of rubber and fabric and plastic. Behind the panels there were wires with nonmetallic insulation, and switches and circuit breakers in plastic cases. There were also gobs of insulating material called RTV. (In Gordon Cooper's Mercury flight, some important electronic gear had malfunctioned because moisture condensed on its uninsulated terminals. The solution for Apollo had been to coat all electronic connections with RTV, which performed admirably as an insulator, but, as we found out later, burned in an oxygen environment.) Mistake number two: Far too much nonmetallic material had been incorporated in the construction of the spacecraft.

A photo of before and after short-circuit fire test
After the fire, flammability and self-extinguishment were key concerns. In the test setup at right a wiring bundle is purposely ignited, using the white flammable material within the coil near the bottom to simulate a short circuit (left). Picture at right shows the aftermath: a fire that initially propagated but soon extinguished itself. It took great effort and ingenuity to devise materials that would not burn violently in the pure-oxygen atmosphere. lf a test was not satisfactory and a fire did not put itself out, the material or wire routing was redesigned and then retested.

A photo of a sample of heat-shield material tested at high temperature
Seared at temperatures hotter than the surface of the Sun, a sample of heat-shield material survives the blast from a space-age furnace. Machines used to check out Apollo components were as demanding as those in the mission itself, because a mistake or miscalibration during preflight trials could easily lay the groundwork for disaster out in unforgiving space.

There is an old saying that airplanes and spacecraft won't fly until the paper equals their weight. There was a time when two men named Orville and Wilbur Wright could, unaided, design and build an entire airplane, and even make its engine. But those days are long gone. When machinery gets as complex as the Apollo spacecraft, no single person can keep all of its details in his head. Paper, therefore, becomes of paramount importance: paper to record the exact confiugration; paper to list every nut and bolt and tube and wire, paper to record the precise size, shape, constitution. history, and pedigree of every piece and every part. The paper tells where it was made, who made it, which batch of raw materials was used, how it was tested, and how it performed. Paper becomes particularly important when a change is made, and changes must be made whenever design, engineering, and development proceed simultaneously as they did in Apollo. There are changes to make things work, and changes to replace a component that failed in a test, and changes to ease an astronaut's workload or to make it difficult to flip the wrong switch.

A photo of the command module floating in water
Meant to fly in a vacuum, and to survive fiery reentry, the command module had also to serve as a boat. Although its parachutes appeared to lower it gently, its final impact velocity was still a jarring 20 mph. Tests like this one established its resistance to the mechanical and thermal shocks of impact, and its ability to float afterward.

A photo of the command module hitting land
Hitting land was possible, even though water was the expected landing surface. For this, a shock-absorbing honeycomb between the heat shield and the inner shell was one protection, along with shock absorbers on the couch supports. A third defense against impact was the way each couch was molded to its astronaut's size and shape, to provide him with the maximum support.

Mistake number three: In the rush to prepare Apollo for flight, the control of changes had not been as rigorous as it should have been, and the investigation board was unable to determine the precise detailed configuration of the spacecraft, how it was made, and what was in it at the time of the accident. Three mistakes, and perhaps more, added up to a spark, fuel for a fire, and an environment to make the fire explosive in its nature. And three fine men died.

A photo of the command and service modules in a huge test chamber
Through the portal of a huge test chamber, the command and service modules can be seen in preparation for a critical test: a simulated run in the entire space environment except for weightlessness. In this vacuum chamber one side of the craft can be cooled to the temperature of black night in space while the opposite side is broiled by an artificial Sun. Will coolant lines freeze or boil? Will the cabin stay habitable?

Previous Next Index