24 replies to this topic

[NHRob]

Well, I finally went ahead and built one of these for purposes of checking collimation and also for optical testing.

It was cheaper and easier to do than I had thought.  The main challenge is in making a proper size pinhole.

The size of the pinhole depends on the aperture you want to test.  In my case, it's not too crazy.  My main scope is my TSA-120, < 5" aperture.

Reading Suiter's book, I conclude that a hole of 50micron diameter or less will be fine, at a distance of at least 20X the focal length.

 

  To make the pinhole, I used thin copper foil I had at work. I flattened a small piece on a smooth piece of glass and then, using a fine sewing needle

I carefully poked small holes in the foil.  As you push the needle, you slowly spin it.  This creates a fairly nice, round hole.

Where I work, I have access to electronic microscopes with measuring capability.  With a bit of practice (1 hour) I was able to routinely create

nice pinholes of 40micron diameter and smaller.  I then just save these foil specimens.

 

The case I used is a small, plastic case large enough to fit a 9V battery in it.   

For the light source I used a Green, Ultrabright LED.  This I got from Digi-Key for $0.24 (#C503B-GAS-CB0F0791-ND).

It has a brightness rating of 53650 mcd.

 

I drilled a hole in one end of the plastic box and mounted the foil on the inside.  I installed a rocker switch and run the LED through a resistor off the 9V battery.

Pretty simple.  To protect the pinhole I just put a piece of masking tape across the hole in the case!

Attached Thumbnails

• 

Edited by NHRob, 31 August 2014 - 11:40 AM.

[NHRob]

Another view ....

 

In this case I have 2 pinholes  ... one about 80micron, the other about 40 micron.

For my testing I just place it about 100 feet away atop a fencepost and turn it on. 

The LED is bright enough that I can star test in daylight.

Attached Thumbnails

• 

Edited by NHRob, 31 August 2014 - 10:16 AM.

[Scott in NC]

Very nice, Rob!   Where did you get the plastic box from?

[NHRob]

IIRC, it was DigiKey.  I like ordering my electronics hobby stuff from them.  Prices are reasonable and they ship things promptly.

[DesertRat]

Nice work Rob!

 

Did you measure the voltage drop across the resistor before closing it up?  Curious if you limited the forward current to 20mA or pushed the led.

 

Glenn

[NHRob]

Hi Glenn,

 

  I measured the LED voltage drop to be 3.25V at 20mA.  I ended up using a 220 ohm, 1/2w resistor in series with the LED. This runs it close to 26mA.

[DesertRat]

Thanks Rob!  I used a very similar setup and a Mouser part # 630-HLMP-CM1A-450DD run at 30mA. 

 

These kind of diodes are too bright to look at - just in case anyone is tempted.

 

Glenn

[MKV]

Use a weak (1 mW) 3 Volt DC dot laser diode with the collimating lens removed. The laser's light cavity is only a few microns across, providing an ideal bright "pinhole". Run the laser in LED mode, not the lasing more. 

 

Mladen

[DesertRat]

Hi Mladen,

 

I've got some red and green laser diodes.  What voltage will prevent it from lasing?  Normally I run them with two 'C' cells on an optical bench.

 

The reason you want LED mode is to minimize speckle and interference effects?

 

Thanks!

 

Glenn

[MKV]

Hi Glenn, go to Radio Shack and get some potentiometers. As you dim the beam you can clearly see when the laser stops lasing. However, some diodes may be too dim in non-lasing mode. My understanding is that DVD recorders use non-lasing laser diodes, but I am not sure how safe they are. I would stick with 1 to 5 mW diodes max, preferably 1 mW.

 

Mladen

[Arjan]

Use a weak (1 mW) 3 Volt DC dot laser diode with the collimating lens removed. The laser's light cavity is only a few microns across, providing an ideal bright "pinhole". Run the laser in LED mode, not the lasing more. 

 

Mladen

Mladen, do you mean with "dot" that the cavity has a square shape? Usually they are significantly elongated.

If so, where can you buy such laserdiodes?

[MKV]

Arjan, most laser diodes come with a description of the signal. Not all of them are elliptical. Even the ones with an elongated signal are still only 2 by 6 microns at the origin. Six microns is the size of the Airy disc of an f/4 telescope, So, that's a point source for all practical purposes. Check eBay for selections. Below is an example of a signal from 8 feet of my 5 mW green laser bought on eBay. I have red ones with cirular and elliptical signals. I use both. Hope this helps.

 

 

Mladen

 

 

 

Attached Thumbnails

• 

Edited by MKV, 02 September 2014 - 11:44 AM.

[Arjan]

OK, thanks. I found a nice article supporting this here.

I always thought the aperture was much larger than that, order 100um in one direction and less than a micron in the other.

[Pinbout]

Darren Drake wrote a nice piece on using scope to scope testing.

 

http://www.cloudynig...null-test-r1586

[MKV]

 

Darren Drake wrote a nice piece on using scope to scope testing.

 

Here's the method I use. 

 

1. Using an Orthoscopic eyepiece, I focus on  a star.

2. I then place a small circular aperture in the focal plane of the EP. Since the focus is set to infinity, the aperture also corresponds to the telescope's focal plane.

4. The aperture simply blocks out most of the stray ilght.

5. I then place a laser diode into a small cap that fits over the EP.

6. The Laser beam enters the eyepiece as a parallel beam and is focused as a spherical wave to an Airy disc. For a 6 mm EP, It's about 2.4 μm.

7. The Airy disc then expands into a spherical wavefront which travels through the telescope backwards and exists as a parallel beam.

8. The beam axis is then centered with the optical axis of a test mirror or a finished OTA, and focused to form am artificial star.

 

This method allows star testing in close quarters. It operates in controlled temperature. The device used is light, and inexpensive. The method assures the light source is precisely at the collimating telescope's focus, and that the light source is of the size of an Airy disc, and can be made as bright as the laser's maximum intensity. It is probably as close to an artificial star test as it gets.

 

Mladen

Attached Thumbnails

• 
• 
• 
• 
• 

Edited by MKV, 03 September 2014 - 02:11 AM.

[NHRob]

Mladen,

 

  Very nice.  The only issue is that you are testing a system with two telescopes' sets of optics.  The collimating telescope will need to be of very high quality so as not to 

introduce its own spherical aberrations.

[MKV]

Mladen,

 

  Very nice.  The only issue is that you are testing a system with two telescopes' sets of optics.  The collimating telescope will need to be of very high quality so as not to 

introduce its own spherical aberrations.

 

So very true,  Rob. It depends on the reason to use an artificial star. I use it for collimation or exact focus determination. It's much easier to accomplish this in a room then outside. If you're testing for optical quality, then the reference collimating telescope must be of exceptional quality.

 

But if you're using an artificial star outside, you have air currents to deal with, especially close to the ground. So, it's not exactly a perfect situation, and the scope has to equilibrate temperature-wise before it can be tested, which may take some time.

 

If quality is your aim, then autocllimation is probably the best artificial star test setup.
 

Mladen

[TG]

Use a weak (1 mW) 3 Volt DC dot laser diode with the collimating lens removed. The laser's light cavity is only a few microns across, providing an ideal bright "pinhole". Run the laser in LED mode, not the lasing more. 

 

Mladen

 

OK, explain this: why is the collimating lens removed? I'd think you need a parallel beam to produce a small image at the focal plane of the eyepiece.

 

Thanks,

 

Tanveer.

[MKV]

 

OK, explain this: why is the collimating lens removed? I'd think you need a parallel beam to produce a small image at the focal plane of the eyepiece.

 

Hi Tanver, parallel beams by definition do not form an image. Convergent beams do.

 

If your artificial star is going to be a laser diode, you remove the collimaitng lens so the diode can act as a point source. 

 

Mladen

Edited by MKV, 03 September 2014 - 11:14 PM.

[gregj888]

Tanveer,

 

I'll give you a little different answer.  A true artificial star emits parallel “rays”.   To be useful though, you need to cover the full optic.  A common way to do this is to place a point source at the focus of one telescope and image it with the telescope under test.  There are issues with the quality of the first scope but ATMs have done this for years.  This strategy will work in general with the Laser and the collimator that comes on it, if your optic under test is on the order of 1/8” or about 3mm in diameter.

 

By removing the lens and running the diode in superlum mode (non-lasing) you get the equivalent of a very small pin hole emitting light in a petty nice cone.  This allows much larger optics to be tested at the center of curvature or focal point of your choosing.  You can still place this diode at the focus of a telescope if you want a parallel ray source to test an optic at infinity focus.  The small spot when brought to focus is general diffraction limited with good optics.

 

Anything I can get to focus and produces a diffraction disk is close enough for me to call an "artificial Star", but you need to be aware that it may not be at the infinity focus which can be an issue for some optical systems.

 

A few other uses for these diodes:
- I use a laser diode, no collimator but another larger lens, projecting it through a flip mirror to test CCD/CMOS camera.  A

   reference camera is on the other leg of the flip mirror.
- Foucault tester source (I save the lens a put it in place for alignment)
- Source for a Twyman–Green interferometer (how I took the picture of the deformable mirror in the AO thread below).

- source for a monochromatic Lott test

[TG]

@Mladen, @Gregg: thanks for the explanations, I missed the bit about the diode itself being only a few microns in size. I'd always assumed it would be a few mm across. I've used an A-P refractor as a collimator to Roddier test an Intes Mak using the eyepiece + red laser combination. For testing the A-P itself, the red laser won't do so a green laser diode at 50m seems promising.

Tanveer

[MKV]

 

Anything I can get to focus and produces a diffraction disk is close enough for me

Gregg, just because you can get an Airy disc doesn't mean the wavefront is corrected. You can observe diffraction discs even if a significant amount of residual OPD is present.  If you're using your artificial star for collimation purposes you have more leeway in that regard. But if you wish to "star test" your telescope you don't.

 

Mladen

Edited by MKV, 04 September 2014 - 03:51 PM.

[MKV]

Tanveer, just so there is no confusion, what Gregg described as a true artifical star is exactly what my method illustrates in pictures above. It's a basic optical beam expander. It's important to understand that the laser used in this case is not without the lens. To the contrary, you need a collimated beam to enter the eyepiece. The lensless diode I mentioned earlier was in reference to the not-so-true artificial star most ATMs use, namely a small light source placed at some distance in your home or your back yard. That's not a true artificial star, but a lensless laser is a better choice because it provides a smaller source then an LED, a monochormatic source, a much brighter source, and because. ultimately, it is a much more coherent source then an LED.

 

regards,

Mladen

[gregj888]

Mladen, Yes, agree completely.  We have a decent 12"  flat for Autocollimation  if we are doing "real testing" :-)  Using Autocollimation we know all the errors are from the test scope (the flat is real good) and we aren't inducing too many other errors.    I like you EP setup... nice.

 

 

 

Anything I can get to focus and produces a diffraction disk is close enough for me

Gregg, just because you can get an Airy disc doesn't mean the wavefront is corrected. You can observe diffraction discs even if a significant amount of residual OPD is present.  If you're using your artificial star for collimation purposes you have more leeway in that regard. But if you wish to "star test" your telescope you don't.

 

Mladen

 

[MKV]

 

Mladen, Yes, agree completely.  We have a decent 12"  flat for Autocollimation  if we are doing "real testing" :-)  Using Autocollimation we know all the errors are from the test scope (the flat is real good) and we aren't inducing too many other errors.    I like you EP setup... nice.-- Greg

Thanks Greg. I use that setup only for two reasons: (1) indoor collimation and (2) setting the infinity focus. 

 

Mladen