In the article ISS Solar Array Alpha Rotary Joint (SARJ)
Bearing Failure and Recovery: Technical and
Project Management Lessons Learned you can find your answer. Pretty much the problems were in underestimating the amount of use, which caused use of improper lubricant and a few other things. You can see in the paper they believe unclear requirements/specifications for the SARJ were given to everyone.
The following are a few brief quotes from the Lessons Learned section of the article:
For complex systems it is especially important that everyone have a clear, correct, and basic understanding of the entire system so they can see how their own detailed knowledge fits into the whole. For an engineered system like the SARJ, under development at that time, a frequently updated project description document would have been appropriate.
Such a document, however, does not seem to have existed.
The first problem was then the lack of proper mission documentation and management. This led to differences in the actual design of the SARJ. The actual design varied from the conceptual design, switching to the use of a gear.
This subtle difference prevents the rollers from self-aligning like a follower-castor wheel and can lead to roller tipping and race-ring pinching (Ref. 5). The conceptual design also took a different approach to the use of “lubricants” for the rollers. In the original design concept, the rollers were
intended to drive the ring, and sufficiently high sliding friction was thought to be necessary to prevent roller-race slippage (Ref. 3). Thus, surface traction materials were suggested that could ensure high friction and enduring smooth surfaces as opposed to the low-friction surfaces actually required by the
SARJ.
The article further discusses a misuse or miscommunication of gold foil as a solid lubricant.