Artificial Stars for testing telescopes: the final answer?

[sophiecentaur]

I feel the need to make sure my Dobsonian scope is working at its best. I was considering a conventional (?) red LED collimator but I suddenly begin to see these Artificial Stars on sale. Would that be a better way to go? Anyone have experience of them? Looking at real stars tells me that the collimation is not too bad but it could be better. Which should I go for?

 

[davenn]

suddenly begin to see these Artificial Stars on sale.

g'day mate

got a link to these artificial stars ? ... haven't heard of themDave

 

[Drakkith]

I've never used an artificial star, only laser collimators. Sorry!

 

[sophiecentaur]

g'day mate

got a link to these artificial stars ? ... haven't heard of themDave

This link tells you all you need to know about the principles. They don't need to be particularly expensive but there are some very pricy ones (of course).
My problem would be how to do any adjustment at one end of a 1m Dobs whilst looking at the eyepiece at the other end. Using a laser collimator could be more straightforward, I think. (I found a movie with two geezers using one and it looked a doddle.)

I've never used an artificial star, only laser collimators. Sorry!

Thanks for the reply. I think they must be a recent addition to the market place.
If I go for a laser option then the choice is quite wide but Amazon etc. reviews are varied on all of them. I never know how seriously to take some of the comments you can read on Amazon, though. Some people are so inept and a lot of poor reviews are largely the result of klutzes trying to use fancy equipment.

 

[Drakkith]

I wish I could be of more help. My SNT I used my collimator for is so poorly made that it merely picking it up uncollimated it.
I actually haven't used it in a long time and I never became any good at collimation when I was using it.

 

[russ_watters]

Star collocation is a time consuming PITA, but it produces great results. I've never heard of an "artificial star" for it.

 

[sophiecentaur]

Star collocation is a time consuming PITA, but it produces great results. I've never heard of an "artificial star" for it.

Well, you see, I am a keen new boy to this topic and I am reading everything I can find. The first ad I saw was only a few weeks ago.

 

[Andy Resnick]

I feel the need to make sure my Dobsonian scope is working at its best. I was considering a conventional (?) red LED collimator but I suddenly begin to see these Artificial Stars on sale. Would that be a better way to go? Anyone have experience of them? Looking at real stars tells me that the collimation is not too bad but it could be better. Which should I go for?

Those devices seem to be based on the (aptly named) 'star test' method of optical alignment- you are simply aligning the optics to obtain the best possible Point Spread Function (PSF). It's perhaps the easiest method (certainly the oldest), but is also semiquantative and requires experience to rapidly converge to optimal alignment. I use the star test regularly, whenever I need to align a microscope objective- this is trivial with the star test, but that's because the source doesn't have to be located far from the lens. With a telescope, you have to place the 'star' far enough away so that it is unresolved and within the focus range of the telescope- AFAIK, star tests are generally not used for aligning telescope mirrors.

Since you are not testing individual mirrors, but aligning one with respect to another, do you have access to the prime focus? Another good alignment technique is 'autoreflection' or 'retroreflection'. Here, you place a mirror or ball bearing (the ball bearing center is located at the focal plane) at the focus and monitor the light reflected back through the optical system, performing alignment to make the incident and reflected light coincide. This requires more equipment (an interferometer, for example), but is also fairly easy to perform.

Here's a random idea, not sure if it will work: remove the eyepiece, use a laser pointer to send light 'backwards' through the telescope, and aim the telescope at a retroreflecting cube/corner reflector. Then align the mirrors until the reflected beam overlies the incident. If you carefully orient the telescope you may be able to use a simple flat mirror instead of a retroreflector.

 

[sophiecentaur]

Those devices seem to be based on the (aptly named) 'star test' method of optical alignment- you are simply aligning the optics to obtain the best possible Point Spread Function (PSF). It's perhaps the easiest method (certainly the oldest), but is also semiquantative and requires experience to rapidly converge to optimal alignment. I use the star test regularly, whenever I need to align a microscope objective- this is trivial with the star test, but that's because the source doesn't have to be located far from the lens. With a telescope, you have to place the 'star' far enough away so that it is unresolved and within the focus range of the telescope- AFAIK, star tests are generally not used for aligning telescope mirrors.

Since you are not testing individual mirrors, but aligning one with respect to another, do you have access to the prime focus? Another good alignment technique is 'autoreflection' or 'retroreflection'. Here, you place a mirror or ball bearing (the ball bearing center is located at the focal plane) at the focus and monitor the light reflected back through the optical system, performing alignment to make the incident and reflected light coincide. This requires more equipment (an interferometer, for example), but is also fairly easy to perform.

Here's a random idea, not sure if it will work: remove the eyepiece, use a laser pointer to send light 'backwards' through the telescope, and aim the telescope at a retroreflecting cube/corner reflector. Then align the mirrors until the reflected beam overlies the incident. If you carefully orient the telescope you may be able to use a simple flat mirror instead of a retroreflector.

You seem to be more or less describing how the laser collimators on sale are used. They have a laser and 45degree semi silvered mirror in place of the eyepiece and you look at the resulting ring pattern from the light coming back on the two way path. Reviews seem to suggest that some of the cheaper collimators are not reliable and need collimating before you start!
The artificial stars have a small enough source to mimic a star, at a distance of 20m or so (allegedly). When I look at a very unfocussed real star, the pattern looks pretty symmetrical with just a hint of the four lines of the secondary support. That must mean the system is fairly well aligned, I guess.

 

[Andy Resnick]

You seem to be more or less describing how the laser collimators on sale are used.

Maybe... I have used an autocollimator for alignment but didn't make the connection here. Autocollimators and associated alignment tools are rather pricy, especially when you need sub arcminute accuracy.

The artificial stars have a small enough source to mimic a star, at a distance of 20m or so (allegedly). When I look at a very unfocussed real star, the pattern looks pretty symmetrical with just a hint of the four lines of the secondary support. That must mean the system is fairly well aligned, I guess.

Hard to say- you began this thread with "Looking at real stars tells me that the collimation is not too bad but it could be better." I'm not disputing the statement, but wondering on what basis you made the claim.

 

[sophiecentaur]

but wondering on what basis you made the claim.

On a very flimsy basis, perhaps. I was just looking at the symmetry of the pattern and comparing it with the 'bad' examples on various places I found. As the Newtonian telescope is so basic, I assumed that was enough. In fact, I couldn't thin of anything else that could be at issue. If the sub reflector is at near enough the right distance from the primary, I assumed the dominant factor would have to be symmetry. (?)

 

[Drakkith]

Well, as long as the light source used in the test is a close approximation of a point-source, I don't see why this wouldn't be a perfectly valid way of collimating. Just make sure the spherical mirror you have to use is of good quality.

 

[Andy Resnick]

On a very flimsy basis, perhaps. I was just looking at the symmetry of the pattern and comparing it with the 'bad' examples on various places I found. As the Newtonian telescope is so basic, I assumed that was enough. In fact, I couldn't thin of anything else that could be at issue. If the sub reflector is at near enough the right distance from the primary, I assumed the dominant factor would have to be symmetry. (?)

Newtonian reflectors suffer from coma- could this be what you noticed?

 

[Drakkith]

Newtonian reflectors suffer from coma- could this be what you noticed?

That's a good thought. If the star isn't near-center in the field of view then coma may cause some asymmetry.

 

[sophiecentaur]

It's a parabolic mirror, which would be better(?).
Anyway, thanks for your thoughts, chaps. I'll have to do some looking around the Astronomy forums. I don't hold up much hope though because, unlike PF, they just don't seem to have the 'traffic' and this is a new topic.

 

[Drakkith]

It's a parabolic mirror, which would be better(?).

Huh?

 

[sophiecentaur]

I understood that coma in parabolic mirrors is less/ eliminated.

Huh?

A well justified "Huh?", I think. I rushed round. re-reading some bits on Spheres vs paraboloids. Paraboloid eliminates Spherical aberration on axis (hence parabolic reflectors in radio reflectors) but coma is there, off axis. Slower reflector, the less coma - hardly surprising because the angles are less.

 

[Drakkith]

I understood that coma in parabolic mirrors is less/ eliminated.

Good lord no! Parabolic mirrors have some of the worst coma you will ever come across.

A well justified "Huh?", I think. I rushed round. re-reading some bits on Spheres vs paraboloids. Paraboloid eliminates Spherical aberration on axis (hence parabolic reflectors in radio reflectors) but coma is there, off axis. Slower reflector, the less coma - hardly surprising because the angles are less.

Yep. Slower mirrors have less coma, for exactly the reason you mentioned.