https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 22/02/13 9:31 AM, RichA wrote:

https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

If they can see this pattern, why can't they just move them all
together now with a few instructions to the actuators? Why should
it still take several months? (Actually two of them are almost
aligned already!)

Could it be that they just want them to be *not* aligned in order
to play with the individual mirror shaping first (to turn it into
18 very sharp points?)

--
Jos

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 13/02/2022 08:31, RichA wrote:

https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

Nothing like it at all. Hubble was a single mirror made incorrectly.
This one is so far performing exactly as intended.

If I was going to worry about anything it would be the scratch like
curved linear artefacts towards the bottom of the frame.

Incidentally the cluster of 5 images at bottom left is replicated as a
faint ghost to top left but curiously the bright pair at bottom centre
is not (even though it would appear to be in frame). ASCII art:

' .
.
' ..[missing in ghost]

'
'

The bit that looks like Orion with his belt collapsed to a single star.

This one is a multiple segment mirror with the segments not yet aligned.
It was designed that way so it isn't a surprise that the first light
image looks like 18 separate star images scattered around on the focal
plane. A pair are close together but the rest will need tweaking coarse
adjust first and then final focussing. Some are more nearly in focus
than others but they will only function correctly when they form a
coherent mirror surface with a single on axis focus and in phase.

The hexagonal symmetry is very obvious in the diffraction spikes.

--
Regards,
Martin Brown

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Sunday, February 13, 2022 at 4:27:29 AM UTC-7, Jos Bergervoet wrote:

Could it be that they just want them to be *not* aligned in order
to play with the individual mirror shaping first (to turn it into
18 very sharp points?)

I wasn't aware that they _could_ deform those mirrors, although if that
were true, it certainly would be a reasonable thing to do.

Also, the star they're using for it is one that will be too bright to look
at once the mirrors are aligned! But if that's so, how can they use it
to align the mirrors?

John Savard

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 22/02/13 12:35 PM, Martin Brown wrote:

On 13/02/2022 08:31, RichA wrote:

https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

...

The hexagonal symmetry is very obvious in the diffraction spikes.

The spikes should to a large extent cancel once the 18 images are
added coherently, I'd suppose.. Or isn't that expected?

--
Jos

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Sunday, February 13, 2022 at 4:27:29 AM UTC-7, Jos Bergervoet wrote:

If they can see this pattern, why can't they just move them all
together now with a few instructions to the actuators?

What amazes me is how they know which of the dots is made up of
light coming from which mirror. Have they taken 18 other pictures, each
one with just one mirror moved slightly?

Yes, once they know which dot is which, it seems they have enough
information to proceed directly towards a focused image.

However, there is more going on. One of the things is that they can only
move one mirror at a time, and that one very slowly, because moving the
mirrors creates heat... and they need the mirror to be very, very cold.

John Savard

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 13/02/2022 14:22, Jos Bergervoet wrote:

On 22/02/13 12:35 PM, Martin Brown wrote:

On 13/02/2022 08:31, RichA wrote:

https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

   ...

The hexagonal symmetry is very obvious in the diffraction spikes.

The spikes should to a large extent cancel once the 18 images are
added coherently, I'd suppose.. Or isn't that expected?

No. The hard hexagonal edges and the grid between mirrors will always
result in a hexagonal spiked cross in the star image. It doesn't really detract. Light grasp is way more important for seeing deep.
_
/ \ <<FT>> _\/_
\_/ /\

Our Fourier optics course one of the exam questions included the
diffraction patterns of capital letter apertures and the final answer:

NOW YOU HAVE IT

Someone who should have known better got "NOW YOU HVAE IT"!

O I Y T are fairly easy
E H N W are intermediate
A V are hard

--
Regards,
Martin Brown

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 22/02/13 5:09 PM, Martin Brown wrote:

On 13/02/2022 14:22, Jos Bergervoet wrote:

On 22/02/13 12:35 PM, Martin Brown wrote:

On 13/02/2022 08:31, RichA wrote:

https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

    ...

The hexagonal symmetry is very obvious in the diffraction spikes.

The spikes should to a large extent cancel once the 18 images are
added coherently, I'd suppose.. Or isn't that expected?

No. The hard hexagonal edges and the grid between mirrors will always
result in a hexagonal spiked cross in the star image.

That some effet remains doesn't contradict that those spikes will "to a
large extent" cancel, if the 18 images are added coherently. Assuming
the gaps between the mirrors are much narrower than the diameter of the individual elements, I would expect that ratio also to be the reduction
factor for the spikes, approximately..

It doesn't really
detract.

Why do you believe so? In a Fourier analysis with narrow gaps in the
input function you get a nice reduction factor as described above (of
course smoothly approaching zero if the gaps are narrowed to nothing).

Do you think they already rounded the edges to such an extend that
only higher order effects remain visible to begin with? (That might
make the simple Fourier picture less predictive, I guess..)

--
Jos

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 13/02/2022 11:27, Jos Bergervoet wrote:

On 22/02/13 9:31 AM, RichA wrote:

https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

If they can see this pattern, why can't they just move them all
together now with a few instructions to the actuators? Why should
it still take several months? (Actually two of them are almost
aligned already!)

Remember they have 18 mirrors making 18 images but they don't at present
know which one is which. The next step they will move probably 8 or 12
of them slightly in distinct compass directions to ID them without
changing the pattern too much. Then they move them closer together but
not quite touching and in the same pattern as the mirrors themselves.

They can get the raw images approximately superimposed to first order.

Then comes the really difficult part of making a holograph of the
composite mirror surface and fine tuning the individual focus settings
until all the panels are exactly in phase for the paraxial ray.

I can't recall how long it took them to do it for the Jodrell Bank dish refiguring but it was a few months.

Could it be that they just want them to be *not* aligned in order
to play with the individual mirror shaping first (to turn it into
18 very sharp points?)

Getting it perfect involves deliberately introducing some phase errors
and moving the scope through focus taking images as you go. Once they
get a firing solution they will move them all at once to the final
setting but that will take some time and a lot of computer power on the
ground to interpret the large datasets it produces.

--
Regards,
Martin Brown

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Mon, 14 Feb 2022 09:18:39 +0000, Martin Brown
<'''newspam'''@nonad.co.uk> wrote:

On 13/02/2022 11:27, Jos Bergervoet wrote:

On 22/02/13 9:31 AM, RichA wrote:

https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

If they can see this pattern, why can't they just move them all
together now with a few instructions to the actuators? Why should
it still take several months? (Actually two of them are almost
aligned already!)

Remember they have 18 mirrors making 18 images but they don't at present
know which one is which.

They know which segment created each of the individual images.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 22/02/14 4:50 PM, Chris L Peterson wrote:

On Mon, 14 Feb 2022 09:18:39 +0000, Martin Brown
<'''newspam'''@nonad.co.uk> wrote:

On 13/02/2022 11:27, Jos Bergervoet wrote:

On 22/02/13 9:31 AM, RichA wrote:

https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

If they can see this pattern, why can't they just move them all
together now with a few instructions to the actuators? Why should
it still take several months? (Actually two of them are almost
aligned already!)

Remember they have 18 mirrors making 18 images but they don't at present
know which one is which.

They know which segment created each of the individual images.

Yes, because there was also a version of that image with annotated
segment names. Still, I'd think that individual segment adjustment
will be done before they join the segments (I mean, first doing the
slight tweaking of the berilium frames to get sharp points and then
moving the points together.)

--
Jos

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Wed, 16 Feb 2022 12:04:40 +0100, Jos Bergervoet
<jos.bergervoet@xs4all.nl> wrote:

On 22/02/14 4:50 PM, Chris L Peterson wrote:

On Mon, 14 Feb 2022 09:18:39 +0000, Martin Brown
<'''newspam'''@nonad.co.uk> wrote:

On 13/02/2022 11:27, Jos Bergervoet wrote:

On 22/02/13 9:31 AM, RichA wrote:

https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

If they can see this pattern, why can't they just move them all
together now with a few instructions to the actuators? Why should
it still take several months? (Actually two of them are almost
aligned already!)

Remember they have 18 mirrors making 18 images but they don't at present >>> know which one is which.

They know which segment created each of the individual images.

Yes, because there was also a version of that image with annotated
segment names. Still, I'd think that individual segment adjustment
will be done before they join the segments (I mean, first doing the
slight tweaking of the berilium frames to get sharp points and then
moving the points together.)

NASA has a video out which shows the process. First they will align
the mirrors so the individual images are distributed to spatially
match the mirror pattern. Then they will move them again so that all
the images are in the same spot. Then the integrated adaptive optics
system will take over and move the mirrors automatically to manage
wavefront control.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 16/02/2022 11:04, Jos Bergervoet wrote:

On 22/02/14 4:50 PM, Chris L Peterson wrote:

On Mon, 14 Feb 2022 09:18:39 +0000, Martin Brown
<'''newspam'''@nonad.co.uk> wrote:

On 13/02/2022 11:27, Jos Bergervoet wrote:

On 22/02/13 9:31 AM, RichA wrote:

https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

If they can see this pattern, why can't they just move them all
together now with a few instructions to the actuators? Why should
it still take several months? (Actually two of them are almost
aligned already!)

Remember they have 18 mirrors making 18 images but they don't at present >>> know which one is which.

They know which segment created each of the individual images.

Yes, because there was also a version of that image with annotated
segment names. Still, I'd think that individual segment adjustment
will be done before they join the segments (I mean, first doing the
slight tweaking of the berilium frames to get sharp points and then
moving the points together.)

I would expect them to move each image to fairly close proximity to the
centre of the field and on the same side or exact opposite side of dead
centre for the purposes of more easily solving the phase equations.

Each off axis mirror segment will only perform optimally when it is
aligned to make an image at or near the optic axis. It makes no sense to
try and sharpen them up until all the crude position adjustments are
complete.

If you want to see the sorts of diffraction patterns it will have try a hexagonal mask on your APO or more realistic a zigzag hexagonal mask.

--
Regards,
Martin Brown

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Wed, 16 Feb 2022 15:23:51 +0000, Martin Brown
<'''newspam'''@nonad.co.uk> wrote:

On 16/02/2022 11:04, Jos Bergervoet wrote:

On 22/02/14 4:50 PM, Chris L Peterson wrote:

On Mon, 14 Feb 2022 09:18:39 +0000, Martin Brown
<'''newspam'''@nonad.co.uk> wrote:

On 13/02/2022 11:27, Jos Bergervoet wrote:

On 22/02/13 9:31 AM, RichA wrote:

https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

If they can see this pattern, why can't they just move them all
together now with a few instructions to the actuators? Why should
it still take several months? (Actually two of them are almost
aligned already!)

Remember they have 18 mirrors making 18 images but they don't at present >>>> know which one is which.

They know which segment created each of the individual images.

Yes, because there was also a version of that image with annotated
segment names. Still, I'd think that individual segment adjustment
will be done before they join the segments (I mean, first doing the
slight tweaking of the berilium frames to get sharp points and then
moving the points together.)

I would expect them to move each image to fairly close proximity to the >centre of the field and on the same side or exact opposite side of dead >centre for the purposes of more easily solving the phase equations.

As I understand it, they don't manually deal with the phase. The
telescope does that autonomously. All they need to do is get each
mirror aligned so that they all place the stellar image at the same
point.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 16/02/2022 18:32, Chris L Peterson wrote:

On Wed, 16 Feb 2022 15:23:51 +0000, Martin Brown
<'''newspam'''@nonad.co.uk> wrote:

On 16/02/2022 11:04, Jos Bergervoet wrote:

On 22/02/14 4:50 PM, Chris L Peterson wrote:

On Mon, 14 Feb 2022 09:18:39 +0000, Martin Brown
<'''newspam'''@nonad.co.uk> wrote:

On 13/02/2022 11:27, Jos Bergervoet wrote:

On 22/02/13 9:31 AM, RichA wrote:

https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

If they can see this pattern, why can't they just move them all
together now with a few instructions to the actuators? Why should
it still take several months? (Actually two of them are almost
aligned already!)

Remember they have 18 mirrors making 18 images but they don't at present >>>>> know which one is which.

They know which segment created each of the individual images.

Yes, because there was also a version of that image with annotated
segment names. Still, I'd think that individual segment adjustment
will be done before they join the segments (I mean, first doing the
slight tweaking of the berilium frames to get sharp points and then
moving the points together.)

I would expect them to move each image to fairly close proximity to the
centre of the field and on the same side or exact opposite side of dead
centre for the purposes of more easily solving the phase equations.

As I understand it, they don't manually deal with the phase. The
telescope does that autonomously. All they need to do is get each
mirror aligned so that they all place the stellar image at the same
point.

They may not do it manually any more but it is definitely using the same holographic dOTF phase retrieval algorithm as Jodrell Bank for segments.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4814230/

I would be a little surprised if the on board computer had enough
compute power to do this sort of analysis in real time.

--
Regards,
Martin Brown

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Thu, 17 Feb 2022 11:42:43 +0000, Martin Brown
<'''newspam'''@nonad.co.uk> wrote:

On 16/02/2022 18:32, Chris L Peterson wrote:

On Wed, 16 Feb 2022 15:23:51 +0000, Martin Brown
<'''newspam'''@nonad.co.uk> wrote:

On 16/02/2022 11:04, Jos Bergervoet wrote:

On 22/02/14 4:50 PM, Chris L Peterson wrote:

On Mon, 14 Feb 2022 09:18:39 +0000, Martin Brown
<'''newspam'''@nonad.co.uk> wrote:

On 13/02/2022 11:27, Jos Bergervoet wrote:

On 22/02/13 9:31 AM, RichA wrote:

https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

If they can see this pattern, why can't they just move them all
together now with a few instructions to the actuators? Why should >>>>>>> it still take several months? (Actually two of them are almost
aligned already!)

Remember they have 18 mirrors making 18 images but they don't at present >>>>>> know which one is which.

They know which segment created each of the individual images.

Yes, because there was also a version of that image with annotated
segment names. Still, I'd think that individual segment adjustment
will be done before they join the segments (I mean, first doing the
slight tweaking of the berilium frames to get sharp points and then
moving the points together.)

I would expect them to move each image to fairly close proximity to the
centre of the field and on the same side or exact opposite side of dead
centre for the purposes of more easily solving the phase equations.

As I understand it, they don't manually deal with the phase. The
telescope does that autonomously. All they need to do is get each
mirror aligned so that they all place the stellar image at the same
point.

They may not do it manually any more but it is definitely using the same >holographic dOTF phase retrieval algorithm as Jodrell Bank for segments.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4814230/

I would be a little surprised if the on board computer had enough
compute power to do this sort of analysis in real time.

Why? AO systems used in dozens of telescopes manage the wavefront
analysis and mirror control in realtime, without extraordinary
computers. The system has to be realtime, or it won't work.

The JWST mirror control system is a local closed loop.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Wed, 16 Feb 2022 15:23:51 +0000, Martin Brown
<'''newspam'''@nonad.co.uk> wrote:

On 16/02/2022 11:04, Jos Bergervoet wrote:

On 22/02/14 4:50 PM, Chris L Peterson wrote:

On Mon, 14 Feb 2022 09:18:39 +0000, Martin Brown
<'''newspam'''@nonad.co.uk> wrote:

On 13/02/2022 11:27, Jos Bergervoet wrote:

On 22/02/13 9:31 AM, RichA wrote:

https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

If they can see this pattern, why can't they just move them all
together now with a few instructions to the actuators? Why should
it still take several months? (Actually two of them are almost
aligned already!)

Remember they have 18 mirrors making 18 images but they don't at present >>>> know which one is which.

They know which segment created each of the individual images.

Yes, because there was also a version of that image with annotated
segment names. Still, I'd think that individual segment adjustment
will be done before they join the segments (I mean, first doing the
slight tweaking of the berilium frames to get sharp points and then
moving the points together.)

I would expect them to move each image to fairly close proximity to the >centre of the field and on the same side or exact opposite side of dead >centre for the purposes of more easily solving the phase equations.

They have now adjusted all of the mirrors in order to bring each image
into a pattern on the sensor that corresponds to the spatial pattern
of the mirrors themselves. Next they'll adjust them so they overlap,
which will be the end of the alignment process.

https://blogs.nasa.gov/webb/2022/02/18/webb-team-brings-18-dots-of-starlight-into-hexagonal-formation/

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Sunday, 13 February 2022 at 03:31:15 UTC-5, RichA wrote:

https://www.dpreview.com/news/7026993143/james-webb-space-telescope-sees-its-first-star-using-all-18-of-its-primary-mirror-segments

https://phys.org/news/2022-02-webb-team-dots-starlight-hexagonal.html

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)