John Doe wrote in :

"Jorge R. Frank" wrote:
And you'll have to send replacements more frequently, since you'll
have no real idea why they keep failing if you can't bring the failed
ones back to the ground intact.

The next step in space evolution would be to have the tools and
documentation to allow the ISS crewmembers to perform the forensic
analysys of failed components. (assuming CMGs could be fitted through
airlock hatch, or at least a CMG with creative use of an empty MPLM
with arm.

Well, no... that's two or three steps away.

The *next* step is to shift the design paradigm away from achieving
reliability by means of redundancy at the system or subsystem level, and
towards achieving reliability by redundancy at the component level. System
and subsystem redundancy is fine for short trips but can be vulnerable to
common-mode failures in the long term, and these types of vulnerabilities
can be very hard to wring out during testing.

That paradigm shift implies a different set of "virtues" for subsystem
design: everything must be designed for crew servicing, and to the extent
possible, subsystems should use a minimum number of common components. That
in turn will minimize the number of tools and spare components that must be
carried

Only then, I believe, will onboard forensics be truly practical.

--
JRF

Reply-to address spam-proofed - to reply by E-mail,
check "Organization" (I am not assimilated) and
think one step ahead of IBM.