Judson Mitchell said:
I am trying to wade through Suiter's book on Star Testing and was wondering if someone would be willing to condense it to a point where a few of the features of mirrors are clarified through star tests. Most interest would be turned edges, over correction, undercorrection, collimation issues, and surface roughness. My reading will ultimately reveal these images, but others as well may benefit, plus it would be beneficial to have that information in a single location. Virtually none of the people I observe with use star tests to any extent if at all.
The followup could be how these features affect EP performance. Anyone care to take a stab at this.
Sure. Although I own and admire Suiter's book, I agree that it can be a bit intimidating. Playing around with Aberrator probably will help (although I've only done a little bit of that myself).
To understand the star test, it pays to spend a little time considering what it is you're looking at. Suppose you were to put a thin sheet of ground glass right into the field stop of the eyepiece. You would capture an image of the star you're testing on, and as you racked the focuser back and forth, that image would get smaller as you approached focus, and then bigger again after you went through it. That's exactly what you're looking at in the star test, but magnified by the eyepiece.
Now forget diffraction for a moment; just think rays of light. When a telescope is optically perfect, light rays from the star--a point source for all intents and purposes--are focused by the objective and converge to a point at the focal plane. The converging and diverging cones of light are exactly symmetrical; the only difference is which way they're facing. That's why, as Suiter says, the intrafocal and extrafocal patterns look exactly the same in a perfect optic.
What's more, the cones are everywhere evenly illuminated; no part of the disc is brighter or dimmer than any other part. You can see this for yourself if you draw out the light rays, looking from the side in a "profile" perspective; the rays don't "bunch up" anywhere.
Next, let's put in an aberration--say, spherical aberration. Remember, we're still just using light rays. When a telescope exhibits spherical aberration, light rays from the periphery focus too close to the objective, and those from the center focus too far from the objective.
When you adjust for best focus in such a telescope, the light rays from the periphery have already converged and are already diverging, whereas those from the center haven't converged yet. The cones are
not symmetrical; it's even impossible to define exactly where one ends and the other begins, since there isn't a unique point of focus.
If you draw out the rays of light for such a telescope, with the outside rays converging closest to the objective, and the central ones converging further out, you'll see that the light rays
do bunch up. Inside focus--that is, closer to the objective--it's the peripheral rays that are bunching up, so the pattern is bright on the edge, dim at the center. Outside focus, it's the opposite: the central rays are bunching up, so the pattern is bright at the center, dim at the edge.
If a telescope is overcorrected for spherical aberration, the central rays converge too close to the objective, and the peripheral rays converge too far out. So the intrafocal pattern is brighter at the center than at the edge, whereas the extrafocal pattern is the other way around--just the opposite of what you get with (undercorrected) spherical aberration. A mildly turned-down edge will be similar to that, but with the difference very obvious at the edge of the pattern.
If the telescope has a rough surface, the light rays bunch up randomly all over the place, and you end up with what is not so affectionally termed a "dog biscuit," for its uneven texture.
Finally, let's bring diffraction back into the mix. What happens? As far as the star test is concerned, not much--diffraction mostly just "quantizes" the light into rings and a central point. So a perfect optic has evenly illuminated rings on both sides of focus, rather than a completely flat disc. A telescope with spherical aberration has a bright outer ring inside of focus, and a bright center outside of focus, and one that's overcorrected goes the other way around. And so on.
Hope that's of some assistance to you in reading Suiter. To be honest, I don't think that most people star test rigorously after they've used a telescope for some time; the things that change from night to night--collimation, seeing, maybe pinched optics--are pretty easy to see either in focus, or else defocused, maybe, but at lower powers than are necessary to run for the star test. But I did spend a lot of time at the beginning, running the star test often, and understanding what the heck I was seeing.
It's true that the test can be grossly misinterpreted. Generally speaking, if your telescope ever star tests well, it's very likely a good telescope, even if it tests poorly at other times. The test is simply very sensitive to all kinds of aberrations. It is even possible in principle to assign a particular wavefront error amount for spherical aberration, and those observers who do, often do so with great force of authority, but I haven't met anyone that I'd trust to the level of precision they offer.
