Automatic software for controlling cryogenic systems

A technical discussion of the lessons learned during the seven years of software development/testing which occurred on the Liquid Oxygen System for the Space Shuttle at the Kennedy Space Center is given. Problems which were solved during these years came into four distinct phases: design/debug before simulation runs, verification using simulation with models up through Space Transportation System-1 launch, hardware usage from first launch to Space Transportation System-5 launch, and future use. Each problem/solution describes the apparent problem requirements/constraints, usable alternatives, selected action, and results.