In order to deal with this challenge, Manning’s team decided to go with a new approach. Rather than using liquid grease to lubricate actuators, as had been done on previous rovers, they used a molybdenum disulfide dry lubricant. The dry lubricant is good down to -135C, so the low temperatures would not be a problem. The actuator gears themselves would be made from titanium.
Unfortunately, during life testing, it became clear that this approach was not going to work. The particles of molybdenum disulfide apparently migrated away from key points on the gear teeth, and the thin layer of titanium oxide on the surface of the gears (which, under Earth conditions, is naturally replenished) broke down. Manning likens this to breaking away the hard candy shell of an M&M candy and exposing its “soft, chewy center.” The result was premature gear failure. The actuators could not survive even half of their intended lifetime.
This forced the team to redesign the actuators to use liquid grease. The gear material was changed from titanium to Vascomax, a high-strength maraging steel. This material is roughly twice as strong as titanium. However, it also has a thermal expansion coefficient that is roughly twice that of titanium. One consequence of this is that the gearset must have greater backlash in order to account for the increased thermal expansion. The team needed to develop strategies to deal with the greater backlash, and ultimately came up with what Manning refers to as “some neat tricks.”
A bigger problem was the fact that heaters would be needed in order to warm the liquid grease to a temperature at which it would be less viscous. Adding these heaters also meant adding thermal controls and fault protection, dramatically increasing the complexity of the system.
Most importantly, the heaters would require power. The Curiosity rover is powered by a radioisotope thermoelectric generator (RTG). At the core of this are about 10 lb of plutonium dioxide, surrounded by a number of small, solid-state thermocouples. The heat generated by the radioactive decay of the plutonium produces a voltage in the thermocouples. The generator outputs about 120W of electricity, nowhere near the thousands of Watts needed to heat the actuators.
Instead, the modest output of the generator is used to charge a lithium-ion battery. The rover design already included a battery to power other subsystems. With the added requirement of heating the actuators, the battery needed to be doubled in size.
The team worked around the clock to design solutions to these problems, to get the redesigned components manufactured, and to conduct testing to validate the new designs. Ultimately, however, it became clear that the 2009 launch date could not be met.
For Earth-bound engineers, missing a key schedule date might mean delaying a project until the next model year, but for Manning’s team, it meant waiting for the planets to align. The minimum-energy launch window from Earth to Mars only occurs once every 26 months. The launch had to be delayed until 2011.