The Early Missions

To qualify Apollo for a lunar landing, the early missions had to test the systems. Basic hardware, performance profiles, reliability, abort modes, and a host of other parameters needed to be tried, measured, and analyzed. Guidance and navigation, of course, were critical to these tests. Could the astronaut control the spacecraft? Could the computer? Could the ground controllers? Could the crew track the stars and landmarks on the earth or moon? Early Apollo missions collected data on gyro drift, accelerometer performance, and ability of the inertal platform to keep stable track of position and velocity. They also began to refashion the pilots' relationship to their craft in ways that would mature during the landings themselves.

After the fire, the second unmanned flight, AS-501 (Apollo 4) was launched in November 1967, intending to fly into high orbit and reenter the earth to simulate a lunar return. An AGC controlled operations on board for five hours. It determined position and velocity, controlled numerous attitude maneuvers, and fired two major burns of the big service propulsion system (SPS) engine on the service module. Ground tracking sent two state vector updates to the computer via telemetry, necessary because without people on board the vehicle could not take its own navigational fixes. The SPS engine burned for four and a half minutes, sending the vehicle into an orbit that would simulate a lunar return trajectory and reenter at 36,500 feet per second.

Unfortunately, a ground controller in Australia sent a turn-on command to the engine after the computer had already generated the command on board. Receiving the instruction, the AGC switched modes and began only taking commands from the ground. This necessitated ground control to issue the shutoff command to the engine, which it did, but 13.5 seconds late, resulting in a 200 feet per second overspeed on reentry, which actually created a more stringent test for the heat shield. Onboard computer control ended at 23,000 feet when the parachutes opened. Despite the extra reentry velocity, the computer controlled the landing to less than two miles from the aim point.

Apollo 5 flew in January 1968, an unmanned test of the LM with no command module. This was the first flight for the Grumman vehicle, the first time its engines would be powered in flight. It was also the first Block II computer flight, and hence the first use of the digital autopilot, using the SUNBURST program. About four hours after launch, the goal was to have the AGC aboard the LM fire its engine at 10 percent thrust for about thirty seconds, and then to go up to full power for twelve seconds.

When the LM began to fire its descent engine, the computer thought the ignition was late, and prematurely shut it down.89 ''Utter chaos took place in the Mission Control Center,'' recalled Jim Miller, who was rope mother for the flight. ''Everybody was climbing all over everybody to find out what happened. Totally preventing anybody from finding out what happened.'' Houston sent a signal to turn off the computer altogether and assumed remote control, bypassing the computer. Miller thought that the mission controllers didn't understand the details of the software, or the subtleties and flexibility of running a software-controlled spacecraft. ''I knew right away what was going on,'' Miller said. ''Nobody had asked MIT anything____They just knew better and took over.'' Miller suggested a way to correct the situation, but the flight directors decided otherwise, because it had never been tested in simulation.

Mission Control commanded a backup guidance system to take over, then issued the burn and the staging commands, and the mission succeeded. The LM turned around, fired its engine, and separated the descent stage. Then the LM's computer took over again, but it had not been informed that the descent stage was no longer there, so its stabilization system was calibrated for a much higher mass. The thrusters hissed and puffed, nearly going unstable.

The cause of the problem, though called a ''software error,'' was actually an error of communications between organizations. ''I had been told by the guy that wrote the descent-burn software that it had to have a very narrow window in the startup,'' Miller recalled, and that there was a serious problem if the thrust didn't build up immedi-ately.90 But a series of pressurizations and valve closures had to occur before the engine even began to light up. The computer mistook this delay for a slow thrust buildup and shut the engine down just as it was firing up. Internally, NASA acknowledged that ''the premature shutdown was the result of incomplete systems integration, and not the result of improper function of individual systems.''91 In this case, ''incomplete integration'' stood for a miscommunication between two individuals.

Publicly NASA used the incident to further a different agenda. Rather than acknowledging the lack of communication that led to the problem, they praised the advantages of having humans on board. ''If men had been aboard to fly the vehicle,'' George Low told the press, ''the flight test might have been a different story____I don't like to fly unmanned missions . . . with hardware designed to carry men.'' Later Sam Philips blamed the trouble on ''overly conservative computer programming.''92

IL programmers now began to face the public and official misunderstanding of their work. Because software did so much to integrate the system and depended on so many technical and organizational interfaces, social and managerial problems could be obscured by blaming the code. ''We who programmed the LM's computer hung our heads in disappointment,'' Don Eyles recalled, ''and endured a public reaction that did not distinguish between a 'computer error' and a mistake in the data.''93 Miller did feel, however, that the incident ''really shook up the Mission Control people who realized that their abilities to handle things was much lower than they thought.'' Computers were introducing new realities into spacecraft flight and mission control, generating tensions between the people who programmed and those who used the programs.

Apollo 6, the final unmanned mission, also aimed to simulate a lunar return. It had trouble on the way up: two engines in the Saturn booster's second stage cut off prematurely and could not be restarted, causing a more elliptical orbit than planned. Then the third stage could not be started to simulate a lunar trajectory insertion. The reentry came in too slow, at 32,800 not 36,000 feet per second. A known bug in the reentry software, which was not expected to be important at the intended higher speed, steered the capsule fifty miles short of its intended landing spot. Overall, the guidance and navigation worked well: the alignment remained under inertial control throughout the mission and the onboard state vector remained accurate to within two miles of ground tracking.94

Apollo 7 was the first manned Apollo flight and the first Block II spacecraft (no Block I flew with people in it). Most of the primary objectives related to the astronauts working with the computer to control the spacecraft. Of the flight's nine goals, the first one read, ''Demonstrate GNCS [guidance-navigation-control system] performance.'' Others included demonstrating the inertial system's coarse and fine alignment, determining orbital parameters by earth landmark tracking, and exercising the attitude hold modes, both automatic and manual.95

Launched on October 11, 1968, Apollo 7 spent almost eleven days flying 164 orbits while running the SUNDISK program on its computer. The AGC fired the service propulsion system (SPS) rocket six times, using digital autopilot for steering, and the computer monitored the Saturn boost into orbit. The crew practiced a command module rescue of the lunar module, pretending the spent Saturn stage was the LM. They tracked the stage with the sextant, from which the computer estimated range. ''This may have been the first in-flight use of Kalman estimation formulations,'' Hoag reported with pride. The astronauts turned the guidance and navigation system on and off several times and easily realigned the IMU with star sightings. During the flight, the computer experienced three ''restarts,'' in response to ''three abnormal procedures.'' These were likely due to erroneous keyboard entries, so the IL refined the logic that allowed the operators to cancel commands if they hit the wrong buttons. Some problems arose with visibility through the scanning telescope due to particles surrounding the spacecraft, and the more critical difficulty of finding good horizons for tracking so close to earth (where the atmosphere blurred the crisp horizon line). Reentry started manually, then switched over to automatic control, and the computer brought the capsule to within a mile of its aim point on the ocean (all subsequent entries were under automatic control).96 Nevins remembered this flight as a major step in NASA's and the astronauts' faith in the computer and automation. ''Performance exceeded our expectations and hopes,'' Hoag wrote to his team in a ''how did we do'' memo that would become a regular event.

While a technical success, Apollo 7 exposed tensions between the astronauts' ideal of mission control and those in Houston. Through much of the mission commander Wally Schirra proved uncooperative and at odds with flight controllers on the ground. ''Wally was legitimately in command of the spacecraft,'' wrote his fellow crewmember Walter Cunningham, ''but he attempted to expand that authority over the entire mis-sion.''97 Whatever their control over the machine itself while in orbit, astronauts still had to contend within the larger organizational system, one with its own distribution of power. None of the Apollo 7 crew ever flew again.

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