johnnydeep wrote: ↑Sat Jul 16, 2022 3:11 pm
VictorBorun wrote: ↑Sat Jul 16, 2022 2:52 pm
johnnydeep wrote: ↑Thu Jul 14, 2022 1:41 pm
Whether JSWT uses mirrors and or lenses to focus the light onto the detectors I'm not sure.
It's ok to use flat "lenses" to filter the light.
It's not ok to use concave or convex lenses to focus the image, because it would destroy multi-chromatic image resolution.
A good Newtonian reflector should avoid focusing with lenses.
What is the shape of the image field focused on the sensor?
Well, it does not resemble a hexagon, criss-crossed into 18 hexagon pieces and crossed with 3 radiuses — the shadows of the 3 holders of the second mirror.
What those things do with the image is create diffractional anti-shadows — 8 spikes around every Milky Way's star and every mid-IR-quasar
(those beasts has just been discovered by JWST):
So Webb doesn't use any lenses to focus light on detectors? I wish I could find a reference for that somewhere.
As for that image explaining where the diffraction spikes are coming from, that is amazingly illustrative, but I still don't really understand it. The conclusion seems to be that each set of parallel edges (either on the 18 hexagonal mirrors or from the secondary mirror support struts) end up creating diffraction spikes perpendicular to them! But why perpendicular? I'm sure there's a "obvious" explanation, which I happen to be too dense to see.
• Focusing light is done two ways (excluding gravitational lensing
), and both lenses (refractive) and mirrors (reflective) manage all sorts of imaging situations. In fact, both lenses and mirrors are used in Webb
. NIRCam uses sets of collimator
lenses before the FPA
(focal plane array). MIRI and NIRSpec
use only mirrors to "focus" light for imaging
. However, the MIRI LRS (low-resolution spectrometer) does introduce a transmissive dual prism (Zn and Ge substrates) to disperse the light, which is very reasonable and easier to implement for low-resolution spectrometry.
A quick search landed these articles, and I think they suffice to answer your first question: Webb employs a hybrid of reflective and refractive optical components depending on application.
• Regarding diffraction, mathematics formally answers your question. It's a bit difficult to explain if you're not familiar with the physics, and unfortunately wrt Webb, edge diffraction dominates and is a bit more complicated than single slit diffraction. However, the principles are the same. Said simply:
→ The fringe pattern below the edge is caused by interference of the edge-diffracted light and the unblocked light, and
→ The reason why the fringes are parallel to the edge is buried in the theory
Scientific Reports wrote:
Among these, Fresnel’s approach based on Huygens’s principle is well known26,27
. It describes the diffraction field in terms of superposition of two waves: one wave propagates through the diffracting object without any perturbation (called the geometrical wave) and the second wave originates from every point of the illuminated boundary of the object or edge (called the boundary diffracted wave).
Note, the figure ignores diffraction at the side edges (assume the block is infinitely long). For a finite long block and a light beam wider than the block, you would see identical fringes on the left and right sides of the block with a transition at each sharp corner.
Note, below is a single-slit diffraction illuminated with collimated light without a focusing lens after the slit
. Inserting a lens will squish the (vertical) height of the diffraction fringes down to be diffraction-limited.
FYI, Wolfram has a single-slit simulator you can play with: Single-Slit Diffraction Simulator
Diffraction fringes form along an axis perpendicular to the edge (linear or curved)
Anecdotally, light incident on a slit (Webb mirror-segment separations), and equivalently, narrow opaque edges (secondary mirror support) will spread away
from the slit or edge. Because the edge lengths are much longer (>> wavelength) than it is wide, visible diffraction dominantly occurs in one axis along the narrow-width direction, i.e. perpendicular to the edges.
→ Think of a single slit illuminated by collimated light (e.g. star light). Looking at the transmitted light on a card near the slit, you'll see a wavelength dependent, periodic slit-like pattern (fringes) having diminishing brightness the further from the slit you look. The visible fringes are much longer than they are wide. Ok, that's up close. Now put a lens immediately after the slit and look at the focal plane of the lens (like a telescope) and you'll see a narrow set of fringes (i.e. not the length of the slit anymore) similarly propagating away from the slit. As stated, this occurs because diffraction imparts an angle to the transmitted light only
in the direction perpendicular to the long axis of the slit.Light is negligibly diffracted along the slit's long axis because any edges are too far apart to significantly diffract the light.
Now, consider a simple refractor. The only diffracting aperture is at the entrance of the telescope. Here, light diffracts perpendicular to the circular edge (i.e. radially
. So, at the telescope's prime focus
, you'll similarly see periodic diffraction rings
This is the Airy pattern
. In reality, basic astronomical telescope designs with a circular primary mirror and secondary mirror supports reveal a diffraction pattern combining the Airy pattern with support vane diffraction.
→ The JWST is definitely more complicated, but the same principles apply.
I hope this increases your understanding of diffraction, at least up to, but excluding, the theory.
You do not have the required permissions to view the files attached to this post.