Mission Planning

The LM seemed to work in space, but how to make it gently land on the lunar surface?

Early on, Space Task Group member Donald Cheatham laid out the basic ideas for landing. Cheatham believed that the landing phase should take advantage of the crew's judgment, especially in the final moments of selecting a specific landing spot. For him, the problem came under the old rubric of ''handling qualities,'' although with few parallels in traditional, atmospheric flight. Cheatham set up a series of simulations in Houston, with a variety of controls and displays, and asked the astronauts to evaluate them according to the standard Cooper rating scales for pilot feel.9

Cheatham divided the landing into phases (figure 8.3), beginning from a point 50,000 feet above the moon. His phases remained salient for the rest of the program: the braking phase, which slowed the LM out of orbit, the approach phase, which allowed the commander to assess the landing site, and the landing phase, where the vehicle hovered to touchdown.

50,000 ft.

Approach/Visibility phase (P64) (-1.5 min)

Lunar surface

250 mi.

Approach/Visibility phase (P64) (-1.5 min)

High gate

Pitchover

High gate

Pitchover

Low gate

Low gate

Figure 8.3

LM landing phases. Not to scale. (Redrawn by author from Johnson and Giller, ''MIT's Role in Project Apollo, Vol. V,'' 182.)

Cheatham's ideas structured the engineering requirements for Grumman to build the LM, but as the vehicle took shape engineers raised numerous questions. How were they to transform these requirements into practical procedures, checklists, backup plans, and go/no-go decisions? This task fell to Floyd Bennett and his group in Houston's Mission Planning and Analysis Division (MPAD).

MPAD emerged in 1963 from part of the original Space Task Group and had responsibility for trajectories, orbital dynamics, navigation, and a variety of other technical functions associated with each mission. This high-morale group did the critical, difficult, but interesting calculations for the Apollo flights: they generated, defined, and analyzed trajectory data including rendezvous procedures, onboard navigation, and computer memory loads. Their offices (and hallways) stacked high with computer printouts, MPAD engineers analyzed the possible trajectories and the use of oxygen and water during Apollo 13. Bill Tindall, who applied such pressure to the IL software project, was one of the senior people in MPAD. MPAD did not directly manage the IL, but they defined the requirements and the trajectories for Apollo, which made them intimately involved in the software effort.10

After a series of reorganizations to keep up with the changing program, MPAD created a Lunar Landing Branch in December 1967, just a year and a half before Apollo

11, with twelve engineers, and Floyd Bennett as its head. It was no surprise that Bennett grew up to spend a career in aviation, for he shared a surname (although no family relationship) with the man who piloted Admiral Byrd around the earth's poles (and after whom the famous airfield in Brooklyn was named). The young Bennett joined NACA Langley straight out of engineering school in 1954, and then the Space Task Group in 1962. He began looking at lunar descent as part of the debate over the LOR decision. Bennett joined NASA Flight Operations in 1966, and after working on a variety of flight mechanics and rendezvous analysis positions, joined the Lunar Landing Branch as its first head.

Bennett made a logo for his group—a cartoon cowboy going over a steep cliff on his horse, with the cowboy saying ''Whoa, Whoa!'' (figure 8.4). Bennett's metaphor for landing was a fractious horse, which in his mind, represented the fuel-optimal descent. The mathematical solution for the trajectory targeted a point below the surface of the moon, which could result in a crash if the computer did not correctly switch to the landing phase at the right moment.

Bennett's plans and Grumman's hardware were in a constant give and take. ''This iteration resulted in much confusion and many investigative false starts,'' Bennett wrote, ''before the mission planners and system designers realized the extent to which the inputs of one affected the other.''11 The landing radar design proved particularly knotty—it depended on the reflective characteristics of the lunar terrain, the topography, and the attitude of the LM. Yet the radar signal absolutely needed to kick in to update the guidance solution, otherwise the astronauts could not attempt a safe landing. Faced with such problems, the designers were naturally conservative, but every ounce of conservatism introduced inefficiencies that cost pounds of fuel. In a system as close to the edge of performance as Apollo, excessive conservatism could push the goal out of reach.

Whatever Bennett and his group envisioned for the landing software, the IL engineers had to program. Enter Allan Klumpp, a mechanical engineering graduate of MIT and an expert in feedback control. Klumpp had been working at the Jet Propulsion Laboratory in Pasadena since 1959, when he was transferred to NASA headquarters to work in Joe Shea's systems group. Klumpp remembered seeing a presentation on Apollo guidance by Trageser and Hoag: ''It was the most high-powered presentation I'd ever seen... they had two slide projectors going at once.'' Klumpp decided that the IL, and not NASA headquarters, was where he wanted to work. Before long he moved to Cambridge. Klumpp enjoyed the peaceful nature of Apollo, for he had designed weapons systems before but was disturbed by the uses to which they were put in Vietnam. War and peace were never far apart in Apollo's technical culture.

At the IL, Klumpp was first assigned to simulate the view out the window during landing. He programmed a computer to plot pictures of what the astronauts would see each second. The assembled images then formed a rough movie of what the land-

Figure 8.4

Mission Planning and Analysis Division, Lunar Landing Branch logo with employees' signatures, depicting the fuel optimum descent as a bronco out of control going over a cliff. (Mission Planning and Analysis Division [MPAD], ''The End of a Great Era,'' June 15, 1990, JSC Archives, 336.)

Figure 8.4

Mission Planning and Analysis Division, Lunar Landing Branch logo with employees' signatures, depicting the fuel optimum descent as a bronco out of control going over a cliff. (Mission Planning and Analysis Division [MPAD], ''The End of a Great Era,'' June 15, 1990, JSC Archives, 336.)

ing might look like from the LM window. Before long, Klumpp was working on the guidance ''equations'' themselves—not just the mathematics, but also the logical flow charts for how the computer would control the vehicle during descent. A colleague, George Cherry, had built a stack of 3,000 punch cards that simulated a landing inside a mainframe computer, and soon Klumpp's version expanded to 6,000 cards as he struggled to translate the simulated landing into real-time instructions for an actual computer. Klumpp remembered that the astronauts originally wanted to fly the LM manually down from orbit. He programmed the flights on a simulator and the astronauts crashed every time. The landing would be critically dependent on the computer.12

Klumpp, along with his colleague, Don Eyles, implemented the landing equations in actual program code. A 1966 graduate of Boston University in mathematics, Eyles didn't even join the IL until the year before Apollo started flying. Yet the intense, bright Eyles soon found himself in the center of Apollo's most critical moments. When the astronauts in the LM interacted with the LM computer in the final phases of their descent, they were talking to a little piece of twenty-four-year-old Eyles's brain.

Telescopes Mastery

Telescopes Mastery

Through this ebook, you are going to learn what you will need to know all about the telescopes that can provide a fun and rewarding hobby for you and your family!

Get My Free Ebook


Post a comment