Yes.Grumble wrote: ↑Thu Mar 17, 2022 11:10 pmThanks monkey. So to give the absolute lies to children version that numpties like me can understand, it may be worth sacrificing some of the ultimate sharpness at the central point in order to get a better overall image (and the best average for all the instruments)?monkey wrote: ↑Thu Mar 17, 2022 11:02 pmI'll have a go!Gfamily wrote: ↑
It's not clear, but as far as I can interpret from some of the descriptions, it's as though there are 4 dimensions to the alignment.
This image represents the basic alignment in 3D, which gives a good image on a flat (or gently curved) image plane.
However, AIUI the next phase is to perform 20 iterations of waveform adjustment, which involves tweaking pairs of mirrors at a time to get an even better alignment (I assume) for the other instruments.
The trouble is, I don't know what they mean. It's like knowing a phrase in a foreign language without really understanding it. I think I have a rough idea, but can't be sure.
If someone can provide a good translation, I'd be grateful.
They got all the mirrors pointing the right way using the selfie. This gives you a pretty good image, but it's not aberration* free, so it could be better. Aberrations cause the point spread function to be less than perfect, or if you like simple terms, make your focused spot go all smeary. This lowers image resolution and brightness.
The aberrations come from the wavefronts** from each segment arriving at the focus with different phases. To adjust the phase from each segment, they move them in and out very small amounts (because wavelength is small) till it's correct. e.g. If the light from one segment is arriving at the focus with advanced phase (too soon), you move the mirror outwards, so it gets there a little later and in phase. Positioning is done on a subwavelegnth scale. This is similar to adaptive optics, what you sometimes get in ground based telescopes***. The difference being on the ground they can do a wavefront measurement and correction on the fly. JWT can't do this measurement, so they use a complicated algorithm to optimise the image ). They will only have to do this once (if everything stays still), unlike the ground based AO, where the atmosphere keeps wobbling about and messing things up and you have to correct continuously. This is the step they have just completed.
As I said in another post, they have only corrected for the centre of the field of view (for one instrument). They will have a very nice image in this one spot, but as some aberations are dependent on the angle away from the optical axis, it might not be good near the edges, for example. The next step is to optimise the whole field, again by moving the mirrors in and out, but now they will sacrifice some of the image quality at the centre to improve it over the whole field. Part of this is making sure it's good in all the instruments, I believe they have an algorithm to decide what the best mirror position is for this. Like the field correction, they may sacrifice quality in one instrument, to improve in another.
*Aberrations are stuff like defocus and astigmatism, like what glasses correct, but much they go into higher orders than that, they are normally characterised by the Zernike polynomials - wiki clicky.
** The wavefront is an imaginary line that joins up all the points of a wave that have the same phase.
***And eye imaging, microscopy, and high powered lasers - wiki clicky.
There's always compromises with optics. You can have an awesome image in one case, but you normally have to sacrifice somewhere else. The compromises you make depend on what you want to do and how complicated you are willing to make things*.
*Putting a telescope in space is an example of making things complicated.