The Project Icarus Blog addresses the problem of maintenance. This is another fundamental issue for an interstellar mission. And I would like to make some comments on this operational aspect of the mission.
The need of maintenance comes from the fact that the performance, the reliability and the safety of the spacecraft has to be insured for a given lifetime in a given environment. If each equipment and the system meet this requirement, there is no need of maintenance. If we know that a component will fail during the mission span, it will need replacement and a maintenance strategy is required.
Thus, the first question is to know if the components can fail during the mission. To answer, on need to know both the characteristics of the equipment and the nature of the environment in which it will evolve. “to know” has here a probabilistic meaning. First problem here: we do not have a precise knowledge of the interstellar environment: The Voyager probes have reached the borders of the solar system and give some hints of its nature but what about the “deep” interstellar medium? The analysis should in the end give us the failure rate of each component.
The second question is the type of strategy to prevent these failures. There is a whole span of strategies with two ends: at one end, the classical “no maintenance” strategy used in spacecraft engineering: all components used have a sufficient lifetime under the given constraints to comply with the mission requirements: no spare, no maintenance. At the other end is the full science-fiction regenerative spacecrafts: like a human body, it can take resources from its environment and permanently replace its defective parts (I think there is this kind of spacecraft in Hamilton’s Night’s dawn trilogy). Thus you can place the cursor wherever you want between the two extremes: an ISS-like automated robotic arm, an army of micro- or nano agents healing the components at the microscopic level or anything else your imagination or your sci-fi books can provide.
The choice of strategy is, as usual, the result of a trade-off between the cost of the mission (cheaper-faster type or Cassini type), the TRL of the maintenance method, the reliability and a bunch of other intertwined factors. You can make a list of all factors, give them a weight factor, create an optimization algorithm, implement it in a code which will give you the right maintenance strategy based on a rigorous approach. Okay, I am just joking, it does not work like that. The right solution will be given by the intuition of engineers cultivated by their knowledge of the existing techniques and methods.
I think there are five categories of components which need different treatments:
- structures (envelope, shields, pipes)
- active mechanical parts (pumps, motors)
- electrical components (RF, antennas, power supply)
- electronics (avionics)
I will take a bit time in a next post to take a tour of the promising technologies in these different areas.
To finish, I would like to address one major issue of the maintenance system: its own maintenance. This problem falls into two parts: first, how is its reliability insured? second, how is its reliability demonstrated? In my opinion, it is a very difficult question that we can illustrate with an example: when assembling a rocket on the ground and that a problem appears on, say, a pump. The pump is replaced: methods to assemble the pumps have to be respected: right constraint on the screws, good grounding and so on. It is achieved by a high skilled workman. But it is not enough to follow the process, it also must be proven that the right process was followed: a controller is present to check that, measurements are carried out to be sure that the process used really lead to the right assembly, that the build baseline corresponds to the design baseline. And sometimes, it is not the case (contamination, bad soldering) and it requires a panel of experts to determine if the pump will still function. All these processes are backed up by a huge configuration control insuring that the exact real configuration of the spacecraft is known and conform to the theoretical one. This is what happens on the ground. Now, how to do that, billions of kilometers away? How will you design the maintenance system in charge of all this processes: fixing, controlling, measuring justifying, registering. And will you prove that it is able to do it with the required reliability?