James Webb Telescope

[RecoveringYuppy]

Thanks smartcooky.

 

[Puppycow]

NASA livestream:

https://www.youtube.com/watch?v=IBPNi7uGgWM

 

[Puppycow]

They're explaining why there's no cameras on the telescope. There's really no light on the dark side, and it's incredibly cold too. An off-the-self camera couldn't operate in temperatures that cold, and would need artificial light to see anything anyway. On the other side, the hot side, it would be very bright and shiny, and of course very hot. You would need many cameras too if you want to see all parts of the telescope.

 

[theprestige]

Am I the only one who's a little depressed that they actually felt they had to explain this?

 

[Puppycow]

The 4th layer has been fully tensioned. On to the 5th, and final layer.

 

[Ziggurat]

The Earth and the Moon wouldn't look like anything, not being illuminated by the sun.

Atmospheric scattering would create a visible glow around the edges of the earth, so you could see it if you're off L2. If you're at L2, then the earth is producing an annular eclipse of the sun, and the sun itself will wash that out, but the silhouette of the earth against the sun would be quite dramatic.

Assuming you're properly at L2, the enshadowed disk of the moon would be invisible within the enshadowed disk of the earth, or vice versa.

Only during a solar or lunar eclipse, otherwise the moon isn't in line with the sun and the earth. The earth-L2 distance is about 930k miles, and the earth-moon distance is about 240k miles. So at half moon, the moon will be at about 14 degrees away from earth (and the sun). That's enough that you should be able to see a sliver of illumination on the moon.

The larger disk would occlude some or all of the disk of the sun.

The earth is smaller than the sun when viewed from L2.

Getting all three in one shot would be a mild achievement in applied geometry,

Not really. Just wait for a half moon (happens twice a month). Balancing the exposure is probably more of an issue (the sun is so much brighter than the moon), but that can probably be dealt with easily enough by combining multiple exposures, as is pretty standard practice in astronomy. Alternatively, it wouldn't be hard to catch the moon and the earth both partially eclipsing the sun, with the moon above or below the earth. The tilt of the moon's orbit is why we don't get solar eclipses every lunar cycle.

But it wouldn't look like much

Oh, I don't know. A lot of people really liked seeing the pale blue dot of the Earth against Saturn's rings from Cassini.

and wouldn't be worth going out of our way for.

I'll agree there.

 

[RecoveringYuppy]

Atmospheric scattering would create a visible glow around the edges of the earth, so you could see it if you're off L2.

The orbit JWST is headed for would also see a bit of crescent Earth too wouldn't it? If I understand the orbit correctly I think Earth would be a constant crescent sliver with the crescent rotating across the planet every 6 months.

BTW there is also the issue that we are talking about infrared instruments. Neither the "dark" side of the Moon nor the Earth would be dark in that range of the spectrum.

 

[theprestige]

The orbit JWST is headed for would also see a bit of crescent Earth too wouldn't it? If I understand the orbit correctly I think Earth would be a constant crescent sliver with the crescent rotating across the planet every 6 months.

BTW there is also the issue that we are talking about infrared instruments. Neither the "dark" side of the Moon nor the Earth would be dark in that range of the spectrum.

Nitpick: I wasn't talking about infrared instruments, since they would be burned out by such a view. I was talking about ynot's hypothetical gopro or whatever.

 

[smartcooky]

Assuming you're properly at L2, the enshadowed disk of the moon would be invisible within the enshadowed disk of the earth, or vice versa. The larger disk would occlude some or all of the disk of the sun

Nope. JWST will never be "properly at" L2, it will be in a halo orbit "around" L2, so the Earth will never occlude any part of the sun's disk

https://jwst-docs.stsci.edu/files/97976978/97976979/1/1596073035306/JWSTorbit.jpg

"The distance of JWST from the L2 point varies between 250,000 and 832,000 km, as shown in Figure 1. The period of the orbit is about 6 months. The maximum excursion above or below the ecliptic plane is 520,000 km. The maximum distance from the Earth is 1.8 million km, and the maximum Earth-Sun angle is <33°."​

 

[smartcooky]

Tensioning is complete. All layers now fully tensioned, so the sunshield is ready to do its work

Next up, deployment of the secondary mirror tomorrow or the day after.

 

[Trebuchet]

Tensioning is complete. All layers now fully tensioned, so the sunshield is ready to do its work

Next up, deployment of the secondary mirror tomorrow or the day after.

And with that accomplished, JWST will be able to do useful work even if the wing mirrors don't deploy. Better if they do, of course.

 

[Ziggurat]

And with that accomplished, JWST will be able to do useful work even if the wing mirrors don't deploy. Better if they do, of course.

Secondary mirror still needs to deploy. Won't do anything at all without that.

 

[Senex]

As someone who is getting on in years I would love to know a few more secrets of the universe. I doubt my best wishes will help but I doubt they will hurt.

 

[crescent]

How many single-point-failure steps left to go?

(ETA: In my head at least, deploying the sunshield seemed to be just about the most complicated thing. Good to have that done.)

 

[theprestige]

How many single-point-failure steps left to go?

(ETA: In my head at least, deploying the sunshield seemed to be just about the most complicated thing. Good to have that done.)

Serious engineering question: Is it still considered a single point failure if there's redundant pathways to success?

For example, many modern tank turrets have both an electrically-driven mechanism for turning the turret, and a manual mechanism for turning the turret. But there's still only one turret and one rotating joint. Would that be considered a single point of failure (to turn the turret)?

 

[Pixel42]

How many single-point-failure steps left to go?

A NASA tweet I saw said 75% of the 344 have been retired.

 

[smartcooky]

A NASA tweet I saw said 75% of the 344 have been retired.

And at least half of that 75% are the ones they were most concerned about.

 

[smartcooky]

Serious engineering question: Is it still considered a single point failure if there's redundant pathways to success?

For example, many modern tank turrets have both an electrically-driven mechanism for turning the turret, and a manual mechanism for turning the turret. But there's still only one turret and one rotating joint. Would that be considered a single point of failure (to turn the turret)?

That depends on the failure mode.

In the example you give, a drive motor failure would not be a single point of failure if the redundancy included a manual rotation system (like landing gear on an aircraft). However, the breaking of a drive cog or worm gear that was common to both the motor drive system and the manual system of rotation would be a single point failure.

There was a great example of a design allowing a single failure to cause the failure of redundant systems in United Flight 232 that crashed in Sioux City Iowa in 1989. This was a Lockheed L1101, a Tri jet (two engines under the wings, one engine in the tail).

The hydraulic system was triple-redundant - three complete and separate systems for the control surfaces - if one failed there were two more to take over. However the problem with the design was that the hydraulic lines for all thee systems came very close to each other near the tail engine, and when a fan disintegrated in that engine, the shrapnel took out all three of those lines, and left the crew with no control of the aircraft.

This is not a true "single point failure" more a design flaw (like your tank turret) where the engineers did not foresee the possibility that a single failure might knock out all available redundancy too.

 

[Ziggurat]

That depends on the failure mode.

There are also usually hypothetically possible failure modes which we don't practically have to worry about. In the tank example, a turret rests on a sort of ring track that it can rotate on. If this ring track buckles significantly, the turret won't work, and there's no backup. But that ring track isn't going to buckle significantly under any conditions which aren't basically destroying the tank.

With the JWST, if the strut on the upper arm for the secondary mirror snaps, it's over, the whole thing fails. But the upper arm strut isn't going to just fracture. The joint might have problems, the motor that's supposed to extent it might have problems, but the strut itself just snapping isn't something they're likely including in their list of possible single point failures.

 

[alfaniner]

Tensioning is complete. All layers now fully tensioned, so the sunshield is ready to do its work

Next up, deployment of the secondary mirror tomorrow or the day after.

I'm glad I got caught up here after the fact. I think waiting through each tensioning would have been stressful. I can only imagine what it must be like for the engineers!

I will try to check in on main mirror deployment, though. (Friday?)