Telescope Reviews: Band-Pass of Single and Double Stacked Filters

I don't know what exactly was the original difficulty/disagreement, so I'll try to give an overview of using etalons.

A fabry-perot etalon is basically just two parallel optical surfaces, making up an interference filter. But the peak wavelength shifts as the angle between the beam and etalon is changed. So you want all your light to pass through the etalon at the same angle (i.e. parallel beam).

That's easy if you're observing a point source (e.g. a star). The light from the point source is parallel to begin with, so you can just mount the etalon in front of the telescope. But a big etalon is very expensive, so you would use a telescope to concentrate the light, collimate the beam again (i.e. make it parallel), and pass it through a small etalon:

But the sun is not a point source, it is in fact a very large extended source. So if the etalon is tuned to H-alpha at the center of the sun, it's not tuned properly for the limb. As a result, there is a wavelength shift from disk center (blue rays in diagram) to the limb (red rays):

Note here, that not only is the red ray at an angle, but this angle is increased by the telescope/collimator setup (i.e. larger at the etalon than in front of the telescope). This is unavoidable; if you try to reduce the size of the beam, the ray angle increase - that is, the image is magnified. (This is the same reason why there is a minimum magnification for a given telescope aperture - if you go lower, the light won't all fit through your pupil.) The smaller the etalon, the more severe the tilt of the red rays, and the more severe the wavelength shift. Or to think of it another way - for a given size etalon, the larger the telescope, the more severe the wavelength shift (because the light must be concentrated/magnified more to fit the given etalon).

So how much tilt can we tolerate? I won't go into the math, but at 1/4 of a degree, it becomes borderline unbearable. Which is the radius of the Sun, so we might as well just put the etalon in front of the telescope. This also simplifies the telescope design, so that's why Coronado, etc. do this:

This works pretty well for visual observation. It gives excellent bandpass at the center of the field of view, and tolerable view towards the edge of the field. If you want to observe a prominence on the limb, you just move the telescope and center it in the eyepiece. But it's not so great for photography or scientific observations; for those purposes, you want a very uniform view across the field.

So how do you make sure the rays from the limb passes through the etalon at the same angle as the rays from disk center? Ideally we want something that behaves like this:

But there is no such magical optical device. It's against the laws of thermodynamics, because such a device would be able to take a diffuse light source, turn it into a perfectly parallel beam, which can then be focused into a tiny spot that gets hotter than the original light source.

SO what's the best thing that we can actually build? It's called a telecentric lens. A telecentric lens is not a collimating lens, but a focusing lens. What makes it "telecentric" is the fact that the bundle of red rays is NOT tilted; it's at the same angle as the bundle of blue rays.

Because the rays from a given point (e.g. the two blue rays) are not parallel, there will be some wavelength shift. The shift depends on which part of the telescope the ray passed; that means even if you observe a point source, the image will be made up of rays whose peak wavelength is slightly shifted by different amounts. In other words, the bandpass is broadened, or smeared. But at least the rays from the limb have the exact same amount of broadning.

Note that the etalon is near (or at) an image plane. So the size of the etalon translates to the size of the field of view. You can reduce the bandpass broadning by reducing the telescope aperture, or by increasing the magnification of the telescope/telecentric system. Though if you increase the magnification, you'll need a larger etalon to cover the same field of view.

That's the very brief overview of the issues involved. Please feel free to ask questions so I know which areas I should explain in further detail.

Edited by Kobayashi (11/11/09 02:30 PM)