Выбрать главу

In the meantime, other engineers prepared the equipment we’d use for this ambitious flight. Although I wouldn’t be driving it myself, I found the work on the lunar rover fascinating. The rover was designed to address a fundamental need: without some kind of vehicle, astronauts would not be able to explore as much of the moon’s surface. But the rover wasn’t the only design that had been put forward.

During one of my visits to North American Aviation, I remember talking to some of the engineers over lunch, and they mentioned something fun they had out on the backyard that I could try if I wanted. When I agreed, they took me to the most bizarre flying vehicle prototype I had ever seen, a small, flat, circular platform, with a four-foot pole sticking out of the top. On the top of the pole was a set of bicycle handles, looking a little like a pogo stick. The engineers put me in a protective suit, strapped me into a safety harness, then asked me to stand on the platform and grip the handlebars. They quickly explained that one handle had a throttle control, and when I twisted it I would activate an air hose that blew down at high pressure from under the platform, counterbalancing my weight. It would probably be unstable, they warned me, so the harness was there in case the platform started to tip over.

I turned the throttle control and, wow, that thing was a kick! It took a lot of getting used to, trying to balance on a carpet of air as I revved it up and slithered and shimmied around. The vehicle was possible to master after some practice, but it took skill. With no control system to keep the platform stable, I had to use my natural instincts to stay upright. It was tough but great fun to fly.

Sadly, it was just a rough prototype of an idea, and nothing like that ever flew to the moon. It would have been a great experience to fly over the lunar surface surveying large areas much faster than walking would allow. But the budget cuts that whittled away the Apollo program meant NASA abandoned ambitious plans such as lunar flyers. Fortunately the lunar rover idea survived, although in a stripped-down, basic version of earlier designs. We also had to wait until the end of 1969 for the green light to develop and build the final version.

The Boeing Company had only about eighteen months to design and build the first car to drive on the moon. It was a crazily short amount of time to come up with something so innovative, especially since the lunar rover needed to fold up like a pretzel on the side of the lunar module for the journey to the moon’s surface. When it reached the moon, the rover had to be unfolded again and ready to drive in a short amount of time. It amazed me how fast Boeing came up with a working vehicle.

But not everything went smoothly. I distinctly recall one time that Boeing was having a real problem getting the rover’s electrical system to work. Fortunately, General Motors was Boeing’s prime subcontractor for the vehicle, and the astronauts had some great contacts within that company because of our Corvette deal. On this occasion a discreet phone call was made to Ed Cole, the president of General Motors and a good friend, describing what was not working. He immediately understood the seriousness of the problem and what needed to be done. The prototype car underwent some General Motors tests that Boeing may not have thought to do, and soon afterward the problems were fixed. Although NASA managers generally frowned on astronauts being cozy with the captains of industry, on occasions like this our personal relationships helped move the program forward and cut through a lot of red tape.

Our crew with full-size mockups of the lunar rover and lunar module

But the rover was of more concern to Dave and Jim. They would be the ones driving it on the surface, so they spent a huge amount of time together training with it. There were other new pieces of equipment that directly affected me, and I wanted to focus my time on them instead.

The Apollo service module had been modified after Apollo 13 to make it much safer. I’d had a number of conversations about it with Jack Swigert, while the engineers and technicians tried to work out what had gone wrong on his mission. Jack had followed the flight plan to the letter, and yet an oxygen tank had exploded. The two of us had worked hard on the spacecraft’s malfunction procedures, so we were concerned that a normal procedure might somehow have caused the damage. It was almost a relief when we learned the oxygen tank had an undetected flaw: an easy fix. There was nothing that Jack could have done to prevent the explosion.

For Apollo 15, there would be even bigger changes to the service module. While I orbited the moon alone, I would operate an entire bay of instruments built into the side of the service module. We called it the SIM bay, or Scientific Instrument Module, and it contained a huge amount of scientific equipment to study the moon in great detail. For example, I would have two different cameras to extensively photograph the surface. A high-resolution panoramic camera would take long, thin photos, capturing objects as small as three feet on the lunar surface while I flew overhead. I also had another, wider-range camera that I could use to help cartographers create a detailed map of the moon. To help calibrate the photos, a laser beam would fire at the surface so we could tell exactly how high up I was, and therefore the distances and feature sizes in the photographs.

I would also have instruments that could detect gamma rays, alpha particles, and X-rays, all of which could tell us a great deal about the composition of the moon’s surface. If volcanic gases were also still escaping from the moon, even in minute quantities, we should be able to measure them. When added to the photographs I would take with a handheld camera through the spacecraft windows, I would be an independent scientific laboratory, able to gather data on the moon from above like no human ever had before. At the end of the mission, I’d even get to launch a tiny satellite, which we would leave in lunar orbit to continue making discoveries.

Explaining the SIM bay operations to the press using a model of the command and service module

It was exciting stuff for me and for the scientists who hoped to unlock more of the moon’s secrets. We were moving into far more complex areas than previous missions, doing work that at times sounded more like science fiction. Soon I would be skimming across an alien world, studying and recording it in huge detail. I couldn’t wait.

I worked with the scientists who designed the SIM bay experiments and with the flight planners who integrated our activities into a flight plan with an organized timeline. We ensured that my orbital operations fit well with the surface work, not just in an operational sense, but also in a scientific sense. It would be a powerful combination: Dave and Jim collecting rocks on the surface while I recorded the chemical composition of an entire region from orbit.

This individual training meant I saw less of Jim and Dave as we began to concentrate more and more on the unique elements of our mission. The two of them spent a lot of time on Long Island working with the lunar module, as well as practicing the activities they would conduct on the lunar surface. Even when we trained together, I was often alone in the command module simulator, talking to them in the lunar module simulator. We were probably only in the same simulator for 20 percent of the training time. That made sense, as there were only a limited number of times, such as launch and reentry, when we would work as a trio. For maneuvers, midcourse corrections, and changing our orbit around the moon, it made more sense for me to train solo.