NASASpaceFlight.com Forum
Robotic Spacecraft (Astronomy, Planetary, Earth, Solar/Heliophysics) => Space Science Coverage => Topic started by: Mongo62 on 11/27/2022 02:34 pm
-
JWST's successor: The Carl Sagan Observatory - a 12 METRE optical telescope searching for exo-Earth
https://www.youtube.com/watch?v=BIgQpXObjFI
-
I would like to see it fly, but it's not going to happen in only 12 years.
Something like it, but only half the size, is going to take 20 years to complete (though I look forward to it regardless).
The 10-page white paper (link in the video's description) was submitted to the Astrophysics Decadal Survey, which finished last year. Among many other things, the Decadal was charged with reviewing four, detailed proposals (~400 pages each, IIRC) for flagship space observatories. Two of those were LUVOIR and HabEx.* This paper seemed to be submitted in support of those two proposals, but specifically in favor of LUVOIR - which was bigger and segmented.
The Decadal did recommend a new flagship that was similar to the Sagan proposal, but half as large (still 2.5x larger than Hubble). They recommended a 6m ultraviolet-visible-infrared space telescope on the grounds that this would be the bare minimum required to meet the goal of directly imaging an Earth-sized planet orbiting at ~1 AU around a Sun-like star (as mentioned in the video). However, they also concluded that it would take 20 years and $11 billion dollars to complete the project (there was still some early tech development work to do, such as new coatings for the mirror, which would be different than Hubble or JWST, among other things). In other words, they concluded that LUVOIR (and by extension, the Sagan) was too ambitious, while HabEx was not ambitious enough. (Discusion regarding that descision can be found in the 2020 Astrophysics Decadal Survey (https://forum.nasaspaceflight.com/index.php?topic=41948.0) thread)
A 6m 'scope can still do the things that the 12m was described as doing in the video (though less spectacularly). Many of the arguments presented in the paper for the 12m design still apply to the 6m one. AFAIK, the 6m concept will benefit from the recent work done on JWST, but it will still take 20 years, rather than 12.
*HabEx proposed a 4m monolithic mirror, whereas LUVOIR came in two versions - one with an 8m mirror, and one with a 15.1m mirror. The Sagan proposal seems to have split the difference between "LUVOIR A" AND "LUVOIR B", while also being an even, 5x wider than Hubble. JWST is 6.5m.
-
I am not so sure... Most of the tooling used for JWST is already there, as well as the qualified personnel and procedures for the AIT. The question would be how long it would take to build another JWST today, to have a fair comparison.
I think it should be possible, supposing that the deployment of the 12m mirror can be done much like the deployment of the JWST mirror is done. Mirror blanks can be the same, changing the plating to something viable for visible. Cameras shouldn't be a big deal. It can even use similar design as the JWST cameras. Same with satellite platform...
-
Sagan is planned to be a visible-light telescope, which means the design has to be improved from JWST (better accuracy for mirror alignment, for instance). That's where a lot of that time will go.
-
But it helps a lot it won’t have to be deeply cryogenic.
-
I would like to see it fly, but it's not going to happen in only 12 years.
Something like it, but only half the size, is going to take 20 years to complete (though I look forward to it regardless).
The 10-page white paper (link in the video's description) was submitted to the Astrophysics Decadal Survey, which finished last year. Among many other things, the Decadal was charged with reviewing four, detailed proposals (~400 pages each, IIRC) for flagship space observatories. Two of those were LUVOIR and HabEx.* This paper seemed to be submitted in support of those two proposals, but specifically in favor of LUVOIR - which was bigger and segmented.
The Decadal did recommend a new flagship that was similar to the Sagan proposal, but half as large (still 2.5x larger than Hubble). They recommended a 6m ultraviolet-visible-infrared space telescope on the grounds that this would be the bare minimum required to meet the goal of directly imaging an Earth-sized planet orbiting at ~1 AU around a Sun-like star (as mentioned in the video). However, they also concluded that it would take 20 years and $11 billion dollars to complete the project (there was still some early tech development work to do, such as new coatings for the mirror, which would be different than Hubble or JWST, among other things). In other words, they concluded that LUVOIR (and by extension, the Sagan) was too ambitious, while HabEx was not ambitious enough. (Discusion regarding that descision can be found in the 2020 Astrophysics Decadal Survey (https://forum.nasaspaceflight.com/index.php?topic=41948.0) thread)
A 6m 'scope can still do the things that the 12m was described as doing in the video (though less spectacularly). Many of the arguments presented in the paper for the 12m design still apply to the 6m one. AFAIK, the 6m concept will benefit from the recent work done on JWST, but it will still take 20 years, rather than 12.
*HabEx proposed a 4m monolithic mirror, whereas LUVOIR came in two versions - one with an 8m mirror, and one with a 15.1m mirror. The Sagan proposal seems to have split the difference between "LUVOIR A" AND "LUVOIR B", while also being an even, 5x wider than Hubble. JWST is 6.5m.
Can the 6 meter UV-Vis-IR scope be monolithic? Seems like that could potentially help a lot with cost control, plus being able to launch on either New Glenn or Starship.
We have new launch vehicles, I think it’d be a very good idea to leverage them for large cost reductions by going for monolithic before going to an even bigger segmented telescope. And maybe build like 2 or 3 of them to take a more risk tolerant development posture (with the bonus of having multiple scopes if everything goes well).
-
Can the 6 meter UV-Vis-IR scope be monolithic? Seems like that could potentially help a lot with cost control, plus being able to launch on either New Glenn or Starship.
We have new launch vehicles, I think it’d be a very good idea to leverage them for large cost reductions by going for monolithic before going to an even bigger segmented telescope. And maybe build like 2 or 3 of them to take a more risk tolerant development posture (with the bonus of having multiple scopes if everything goes well).
Wavefront control is harder on a large monolithic scope. At the precision modern astrophysics requires, the surface of the scope has to be adjusted by nanometers or less to get a useful, coherent image. That’s easier to do on a bunch of smaller, thinner, lighter, less rigid mirrors than on one large, thicker, heavier, more rigid mirror. We don’t even know how to get the precision to do that — by like an order of magnitude or so — for UV/V for JWST-sized or larger segmented scopes. Monolithic would make that challenge yet even harder.
It’s also easier to manufacture and mount a bunch of small mirrors, there’s some modest manufacturing scale benefits to multiple small mirrors, small mirrors are less likely to deform in ways we don’t want them to in microgravity, etc. NASA astrophysics is doing its due diligence by revisiting the segmented versus monolithic trade for HWO. But it’s hard to see how monolithic comes out ahead. CERs for segmented have been better than monolithic for a couple decades now.
-
Then why do ground telescopes typically have monolithic mirrors up to 8 meters or so? And they’re WAY cheaper than JWST, etc. And ground monolithic scopes, if what you were saying was the full story, would have it even worse as it has to slew in the presence of Earth gravity.
I just don’t buy it. I suspect it just wasn’t proposed much and didn’t win the trade studies is because the capability to launch large and heavy monolithic apertures did not exist. “Capabilities determine requirements, regardless of what the systems engineering textbooks say.” —one of Akin’s Laws.
-
So I’ll ask again. Is there any information I can find online about this 6 meter space telescope? A PI name? A project name? A link to a paper? Anything??
-
Then why do ground telescopes typically have monolithic mirrors up to 8 meters or so?
AFAICR, they have not since the necessary processing power became available/cheap enough in the 90s. I think the last big monolithic optical scope commissioned for operations was the LBT in 2005. Keck (1993/1999), Hobby-Eberly (1996), S. African Large Telescope (2005), Gran Telescopio Canarias (2007), and Large Sky Area Multi-Object Fiber Spectroscopic Telescope (2009) have all been segmented. The Thirty-Meter Telescope and the ELT will both be segmented, too. The OLT was (ridiculously) monolithic but got cancelled.
And they’re WAY cheaper than JWST, etc.
Any ground scope will be way cheaper than any comparable space scope.
And ground monolithic scopes, if what you were saying was the full story, would have it even worse as it has to slew in the presence of Earth gravity.
Yes, above certain sizes, monolithic mirrors become impractical because they become so massive to maintain their shape that moving them becomes hard to impossible. Better to break that collecting area up into pieces that can maintain their individual shapes without adding so much mass.
Another issue is thermal mass. Keeping a mirror the same temperature as the ambient air is critical to avoid convection in the air above the mirror. This is easier with a lot of small, lightweight mirrors than with one, big, heavy mirror.
I just don’t buy it. I suspect it just wasn’t proposed much and didn’t win the trade studies is because the capability to launch large and heavy monolithic apertures did not exist.
No, monolithic apertures were considered for the Next-Generation Space Telescope, which became JWST:
https://ui.adsabs.harvard.edu/abs/1998SPIE.3356...69W/abstract
“Capabilities determine requirements, regardless of what the systems engineering textbooks say.” —one of Akin’s Laws.
Akin is right, but you have to know enough about the system to understand what the driving constraint is, and the driving constraint on space-based IR/V/UV scopes since HST has been wavefront control, not collecting area. I can collect all the photons I want, but if I don’t collect the right photons at the right time, the data is just gibberish, no matter how much of it I have.
Bigger payload bays on launchers are not the impediment to advancing astrophysics in this area, and the last decadal recommendations for HWO show it. They were fine with another JWST-sized scope, but recognized that wavefront control in the V/UV bands would be several orders of magnitude more difficult and went after that limiting obstacle.
So I’ll ask again. Is there any information I can find online about this 6 meter space telescope? A PI name? A project name? A link to a paper? Anything??
It was just a white paper for the decadal:
https://assets.pubpub.org/q51s85xw/21598545491312.pdf
-
If I remember correctly, wave front control has two components: the curvature of the mirror and the smoothness of the surface. I seem to remember that on another thread it was said that for a large mirror in the 12 meter range, there has to be a significant improvement in the surface smoothness/roughness of the mirrors compared to Webb to work in the Visual and UV range with enough resolution. Improving this kind of manufacturing tolerances even an order of magnitude beyond the state of the art might prove to be a bigger challenge than the technical hurdles that were overcome by the Webb team. The next telescope needs a risk reduction program before going ahead with full scale development. After that, there would be a better idea of just how capable a telescope it would be possible to build before committing to a design.
-
I am not so sure... Most of the tooling used for JWST is already there, as well as the qualified personnel and procedures for the AIT. The question would be how long it would take to build another JWST today, to have a fair comparison.
I think it should be possible, supposing that the deployment of the 12m mirror can be done much like the deployment of the JWST mirror is done. Mirror blanks can be the same, changing the plating to something viable for visible. Cameras shouldn't be a big deal. It can even use similar design as the JWST cameras. Same with satellite platform...
By "cameras", I assume you mean all the detectors on jwst. Those were all just as difficult to build as the mirrors, if not harder. The instruments on telescopes rarely get any press, but they are often a bigger deal than things like the mirrors are.
-
If I remember correctly, wave front control has two components: the curvature of the mirror and the smoothness of the surface. I seem to remember that on another thread it was said that for a large mirror in the 12 meter range, there has to be a significant improvement in the surface smoothness/roughness of the mirrors compared to Webb to work in the Visual and UV range with enough resolution. Improving this kind of manufacturing tolerances even an order of magnitude beyond the state of the art might prove to be a bigger challenge than the technical hurdles that were overcome by the Webb team. The next telescope needs a risk reduction program before going ahead with full scale development. After that, there would be a better idea of just how capable a telescope it would be possible to build before committing to a design.
HWO needs both different mirror coatings and different mechanisms for controlling mirror deformation to achieve the required precision for wavefront control in the V/UV. Candidates for each exist, and a technology program has been proposed to tackle them (and other challenges).
-
If I remember correctly, wave front control has two components: the curvature of the mirror and the smoothness of the surface.
(snip)
As an optical engineer who built imagers for astronomer’s space missions, it is at least not clear what you mean.
Wavefront (one word) control does not deal with or impact surface roughness. That is a fundamental aspect of the polished mirrors on which the requirement scales with wavelength. So a UV system working down to 100 nm needs roughly one tenth the roughness (and particulate contamination) of a Near Infrared system covering down to 1 micron (1000 nm), although the specifics depend on the particular application.
On Earth, wavefront control involves quickly countering the fluctuations in atmospheric path lengths but in the vacuum of space, it about maintaining the figure of the mirror.
For JWST, that means each hexagonal segment has three redundant adjusters to keep its piston, tip and tilt matched to the other segments around it.
Each JWST segment does also have a mechanism to adjust curvature. However, this is unusual, and possibly unique, for what may be the only big segmented mirror in space. (Anyone know of another with curvature control? Keck doesn’t have it.)
Wavefront control can also be done at an internal image of a large monolithic primary by adapting current technology to the space environment.
People have also discussed (but maybe not “proposed”) “bed of nails” figure adjusters for a large (8 meter diameter) thin monolithic mirror in space, but this would have had to be sliced in thirds, like JWST, to fit in the available payload shrouds. This limit will soon be overcome.
-
If I remember correctly, wave front control has two components: the curvature of the mirror and the smoothness of the surface.
(snip)
As an optical engineer who built imagers for astronomer’s space missions, it is at least not clear what you mean.
Wavefront (one word) control does not deal with or impact surface roughness. That is a fundamental aspect of the polished mirrors on which the requirement scales with wavelength. So a UV system working down to 100 nm needs roughly one tenth the roughness (and particulate contamination) of a Near Infrared system covering down to 1 micron (1000 nm), although the specifics depend on the particular application.
On Earth, wavefront control involves quickly countering the fluctuations in atmospheric path lengths but in the vacuum of space, it about maintaining the figure of the mirror.
For JWST, that means each hexagonal segment has three redundant adjusters to keep its piston, tip and tilt matched to the other segments around it.
Each JWST segment does also have a mechanism to adjust curvature. However, this is unusual, and possibly unique, for what may be the only big segmented mirror in space. (Anyone know of another with curvature control? Keck doesn’t have it.)
Wavefront control can also be done at an internal image of a large monolithic primary by adapting current technology to the space environment.
People have also discussed (but maybe not “proposed”) “bed of nails” figure adjusters for a large (8 meter diameter) thin monolithic mirror in space, but this would have had to be sliced in thirds, like JWST, to fit in the available payload shrouds. This limit will soon be overcome.
Maybe I misunderstood that part. I remember going to a presentation put on by a group of scientists and engineers at the University of Wisconsin back around 1979 or 1980 and they were developing an instrument called WUPPE (Wisconsin Ultraviolet Photo Polarimeter Experiment) for the Hubble telescope. It got yanked out on the first Hubble repair mission to make room for the corrective optics and they got no useful observations for all their work. That had to really suck for them. During their presentation they talked about surface roughness as a function of maintaining resolution at the wave lengths being observed that had something to do as a function of focal length. I remember on scientist in particular mentioning that based upon the focal length of Hubble that the surface roughness could be no worse than lambda divided by 76 for the frequency of light being observed. It's been close to half century since I sat in on that presentation and I was a lowly undergrad at the time, but I think he said that the surface would have to get much smoother for larger telescopes with shorter frequencies of light being observed to get a well focused reflection on an instrument like theirs at the resolution required. I remember the guy putting up some formulas on the chalkboard and the lambda divided by 76 stuck in my head.
-
If I remember correctly, wave front control has two components: the curvature of the mirror and the smoothness of the surface.
(snip)
Each JWST segment does also have a mechanism to adjust curvature. However, this is unusual, and possibly unique, for what may be the only big segmented mirror in space. (Anyone know of another with curvature control? Keck doesn’t have it.)
Hubble is not segmented, but does have actuators to change shape. I believe these were included since they were not completely confident their compensation for zero gee operation was good enough. These were considered for fixing the mirror shape problem, but did not have enough range. From an article on fixing the Hubble optics (https://www.cbsnews.com/news/an-ingenius-fix-for-hubbles-famously-flawed-vision/):
Actuators attached to the back of the mirror could be adjusted to make very slight changes to its shape, but the problem appeared to go well beyond that.