Mars Climate Orbiter Launch Delay The battery state of charge (SOC) was estimated or directly measured in a variety of different ways depending on the mission in question. In any case, the role of the battery module was the same: determine the battery state of charge, increase the charging rate if the SOC was too low, decrease the charging rate if the SOC was too high. This software depended on measurements read initially as voltages, then converted through linear equations into battery temperatures and pressures. These values were used to determine the state of charge through another formula. Unfortunately, the conversion factors for the sensor measurements and the state of charge calculation needed to be measured directly from the actual hardware, which hadn't been built yet when the battery software was written. Since the simulation people had no idea what values to supply for voltages to the battery software, running the EPS task in the early months of the program was a disaster. The battery module would read garbage data from the sensors, determine that the battery was at a wildly implausible state of charge, and reboot the spacecraft. The spacecraft messaging and configuration system hadn't been written yet, so, in the interests of being able to test anything else, the software was turned off by default with no way in place to activate it. By the time the Mars Climate Orbiter was set to launch, it was discovered that the software was still deactivated with no way to turn it on, and the spacecraft was on the launch pad! The launch was delayed 24 hours while a patch was written that activated the battery monitor and set its parameters to reasonable values. Once this was done the battery software ran perfectly. This delay was overshadowed by the notorious metric conversion error that led to MCO's destruction.

Related Technologies

• C Programming Language
• Mars Missions
• Spacecraft
• VxWorks
• IRIX
• Solaris
• Embedded Development