4 ms·

Can anyone elaborate on these single points of failures? For example, are they actually single points of failure? Or is it a bit of exaggeration for marketing p

by bmitc 5y ago

Can anyone elaborate on these single points of failures? For example, are they actually single points of failure? Or is it a bit of exaggeration for marketing purposes (a bit of under promise, over deliver)?

terramex 5y ago
There is a bit of exaggeration, for example non-explosive actuators that needed to be released for sunshield to deploy have two redundant electric circuits for their deployment but are considered 'single points of failure' as compound part. Propulsion system is also considered SPoF but there is some redundancy built in, like there are two independent sets of thrusters feeded from one fuel tank. Some other parts don't and cannot have redundancy due to design.
ProAm 5y ago
They are single points of failure.
- bmitc 5y ago
  That's not much of an elaboration. Further, I am doubtful, because something always goes wrong. So if a project that has seen decades of delays and billions of dollars of budget overruns has suddenly invented engineering and processes that yields zero failures, count me surprised. I just wish there was more elaboration of things they are able to accommodate for as things inevitably pop up and not this hyper focus on a number of supposed single point failures.
sailingparrot 5y ago
A lot of them are "optional" but strongly preferred. E.g. if one or two of the mirror wings failed to deploy, JWST would still be usable, but with a degraded resolution. If one or two layers of the sunshade failed to correctly deploy, JWST is still usable, but the temperature of the mirrors would be higher than expected, and thus the spectrum of IR light it can image would be reduced. If the momentum flap fails to deploy, then more fuel is needed to keep the JWST at the right attitude/position, shortening the lifespan of the mission. But some others were make or break: deployment of the solar panels, deployment of the secondary mirror and a few others. If those had failed then the mission would be over.
colechristensen 5y ago
It’s probably an internal list of serious things that could go wrong. There’s a lot of bike shed discussion here about what a “single point of failure” should include.

by bmitc 5y ago

terramex 5y ago

There is a bit of exaggeration, for example non-explosive actuators that needed to be released for sunshield to deploy have two redundant electric circuits for their deployment but are considered 'single points of failure' as compound part. Propulsion system is also considered SPoF but there is some redundancy built in, like there are two independent sets of thrusters feeded from one fuel tank. Some other parts don't and cannot have redundancy due to design.

ProAm 5y ago

They are single points of failure.

bmitc 5y ago
That's not much of an elaboration. Further, I am doubtful, because something always goes wrong. So if a project that has seen decades of delays and billions of dollars of budget overruns has suddenly invented engineering and processes that yields zero failures, count me surprised. I just wish there was more elaboration of things they are able to accommodate for as things inevitably pop up and not this hyper focus on a number of supposed single point failures.

sailingparrot 5y ago

A lot of them are "optional" but strongly preferred. E.g. if one or two of the mirror wings failed to deploy, JWST would still be usable, but with a degraded resolution. If one or two layers of the sunshade failed to correctly deploy, JWST is still usable, but the temperature of the mirrors would be higher than expected, and thus the spectrum of IR light it can image would be reduced. If the momentum flap fails to deploy, then more fuel is needed to keep the JWST at the right attitude/position, shortening the lifespan of the mission. But some others were make or break: deployment of the solar panels, deployment of the secondary mirror and a few others. If those had failed then the mission would be over.

colechristensen 5y ago

It’s probably an internal list of serious things that could go wrong. There’s a lot of bike shed discussion here about what a “single point of failure” should include.