Jump to content


Photo

Star Test on SCT

  • Please log in to reply
23 replies to this topic

#1 Julio

Julio

    Messenger

  • *****
  • Posts: 452
  • Joined: 18 Jan 2007
  • Loc: Pembroke Pines ,FL

Posted 13 January 2013 - 09:58 PM

Hello everyone, been doing some reading about star testing and the science behind deciphering the diffraction patterns. I don’t have the Suiter book and have been struggling attempting to determine the quality of my optics, both CATs. I don’t have an artificial star to test this on, so I took a shot on a real one. It certainly appears easier than it really is; the nature of star testing requires experience and knowledge surpassing mine at the moment. I took the advice of another fellow CN member and slightly defocused a star at 350X. The test is for a SCT and I’m curious as to what my fellow cat owners think. I took some videos afocal on an Alt Az mount so please excuse the shaking, any information would be appreciated. I pulled the EP 5mm-6mm or so to produce the intra and extra focal patterns that showed roughly 4-5 diffraction circles, seeing 3-4/5.

Intra focal

http://s99.beta.phot...s1893e72d.mp...

Extra-Focal

http://s99.beta.phot...sbc1d7f18.mp...

#2 Eddgie

Eddgie

    Voyager 1

  • *****
  • Posts: 12488
  • Joined: 01 Feb 2006

Posted 14 January 2013 - 10:11 AM

First and most important... The star test is a very specific process involving several tests for several different types of errors.

The test you submitted is inconclusive because you have only presented a single set of images and unless by accident, perhaps did not set the spacing properly.

The tests for zones and turned edges are also done in and outside of focus, but for zones, you need much more defocs than in the picture.

The test for smoothness requires good seeing or artificial star.

And using a real star is fine. I prefer it.

Once again, the methods are very precise, and here is the reason why I am going to strongly qualify what I am going to tell you, and you should not panic. The star test is so sensitive that it shows the very smallest error in the system.

And this. The SCT design is only optimal at the stated focal lenght. For a C11, this will be at 2800mm, which is the configuration when using a 1.25" visual back and 1.25" diagonal. Even putting a 2" diagonal can change the result for the Spherical Abberation test. It is that sensitive.

Your image shows a hint of sphrical abberation.

Quantifying it though is amost impossible.

The reason is because the Spherical Abberation test reqires that your images be taken with an exact amount of inward and outward focuser travel.

When you do this, what you are looking for is to compare the ratio of the secondary shadow vs the overall Fresnel disk.

If for example, your system has a 33% obstruction, then at exactly 10 wavelenths inside and 10 wavelengths outside of focus, the seconary shadow should appear to be 33% the size of the fresnel pattern.

If there is SA, it will appear larger on one side of focus and smaller on the other side of focus. How much larger/smaller is used to calculate how much SA is present.

But again, an exact read requires exactly matching amounts of In/Out focus..

Again, on your scope, it appears that inside, the shadow is a bit smaller than outside. That would suggest some SA.

I also see a hint of a zone in the extra-focal image, but it is not defocused enough to tell (the test for zones requires about twice as much defocus as shown here). Outer zones in C11s an C14s are common though and rarely bad enough (on their own) to cause much of a problem.

Again, here is a big big big disclaimer... Because you did not follow the process exactly, the reading is of course subject to error, and based on the probability that you did not use exact amounts of defocus, I would be remiss in attempting to estimate the error, though it does not appear to be severe at all. If you compare the secondary shadow sizes though, you can see that they do appear slighly unballanced.

If you want to do the image again, here is the process.

First, configure the scope for nominal focal length. This means the standard 3.25" to SCT adapter, the standard 1.25" visual back, and the standard 1.25" diagonal. Failure to do this can influence the test negatively.

Next, use a 10mm Plossl or similar focal lenght eyepeice.

Now, find a medium bright star.

Here is the tricky part. At f/10, you want to take your images with 4.4mm of travel ( .1732 inch). It is not essential to be "Exact", but you want to be as close as you can get.

The easiest way to do this test is to move the eyepiece and the focuser in the exact method I am going to describe.

First, find something to use as a thin "Feeler" guage. A thin piece of metal or wood is fine. Trim or shave this piece of wood to 4.4mm thichness so that you have a thin strip of wood or metal 4.4mm long (again, this is for f/10 scopes).

Focus your star as best as possible.

Now, not touching the focuser, slip the eyepiece out towards you and use your feeler guage to set the travel of the eyepiece by placing the flat side against the eyepiece barrel and space the bottom shoulder of the eyepeice housing exactly 4.4mm from the top of the eyeppiece holder.

Get this as close as possible. This is why the feeler guage works so well. Exact and easy to repeat.

Now, take your first image. This will be about 10 wavelenghts of defocus.

Now, without touching the eyepeice, use the telescope focuser to re-focus on your star. Again, do not slip the eyepice. Reach best focus using the telescope focuser.

Once you have reached best focus using the telescope focuser, now you can slip the the eyepeice back to its shoulder. This will put you exactly the same distance inside focus as you were outside focus. Now you can take your second image.

This might sound tedious to do, but it is in fact the right way to star test, and for most scopes, the easiest way to get the exact in/out focus you need to make the test.

Finally, take an image inside and outside with about twice as much defocus. This is the way you test for zones and turned edges. The amount of defocus for this test does not have to be exact at all. About twice what you used for SA is fine, but again, doesn't need to be even close to exact.

Once you have your images, you can PM me and we can talk about modeling them if you want to get a more precise assessment.

I am going to recommend a program called "Abberator 3.0." This program and these images will allow you to get a very accurate assesment of the quality of your instrument.

But if you want assistance, I would be happy to do it.

See, star testing is more than just "Defocusing" and seeing that the rings look alike. Even a bad scope can pass a "Gee, the rings look about the same to me" kind of star test.

The truth only comes to those that know how to star test properly, which is easy to do, but very methodical.

And if you don't want to do all of this, that is fine with me, but for others that want to learn, you will get far more real data than all of the subjective data on optical quality when you learn to test for yourself.

#3 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 8053
  • Joined: 08 Aug 2007
  • Loc: La Union, PI

Posted 14 January 2013 - 12:24 PM

Yes, Eddgie is correct. For a star test, the images are not large enough for some aberrations and too large for others. However, what they do show looks fine. Primarily, the scales in your images are good for collimation, which looks fine, too.

#4 orion61

orion61

    Aurora

  • *****
  • Posts: 4500
  • Joined: 20 Oct 2007
  • Loc: Birthplace James T Kirk

Posted 14 January 2013 - 12:26 PM

Well done Eddgie! Very clear, and concisely written

#5 Steve OK

Steve OK

    Viking 1

  • -----
  • Posts: 843
  • Joined: 22 Sep 2007
  • Loc: OKC, OK

Posted 14 January 2013 - 01:15 PM

Thanks from me, too, Eddgie!

This is going in my notebook!

Steve

#6 Julio

Julio

    Messenger

  • *****
  • Posts: 452
  • Joined: 18 Jan 2007
  • Loc: Pembroke Pines ,FL

Posted 14 January 2013 - 01:22 PM

Ed i really appreciate your methodical approach to this and your insight. Ill do just that as soon as conditions improve down here, its been very cloudy muggy and transparency has been sporadic. Ill start designing the EP mechanism.I ordered a Hubble 5 star to try this on, i got 3 cats so i figure ill make good use of the artificial star. I performed a non scientific test prior to posting here with more defocus in hopes that someone would enlighten me so ill post the link here. Again it is not a proper method simply an addendum to what i performed on this exact procedure, but with more defocus several days prior.

http://www.cloudynig...5617251/page...

#7 Eddgie

Eddgie

    Voyager 1

  • *****
  • Posts: 12488
  • Joined: 01 Feb 2006

Posted 14 January 2013 - 01:49 PM

Well, the other test does show what looks like the typical large Celestron "S" zone.

This zone is very common. I think it is from the use of sub aperture polishers, most likely the same one used for the C8s.

When a sub aperture polisher is used, the center and outside diameter are polished in two passes.

This almost always leaves a zone. My C14 has one, my C11 had one, and many other larger Celestron scopes have one.

What makes me think that it is a sub-aperture polisher is that it appears further out, almost to the outside edge of the C9s that show it, a bit further in on the C11s, and much further in on the C14s.

Minor zones are usually considered "Cosmetic" because while they show up in the star test, they generally don't do much damage to the image. The reality though is that all errors add together, and a little zonal error by itself is usually not a problem, you have to consider the total effect of all abberations. This is the real issue with SCT quality. Usually you suffer death by a thousand cuts... No one serious error, but a collection of minor errors none by itself critical, total up to enough to make one scope OK and another one excellent.

But this zone appears minor. I would be interested in seeing your repeat of the SA test though. Again, if there is only a bit of SA and a bit of a zone, the scope can still easily fall into the Good catagory, and maybe even excellent.

This though is why the star test has to be done as a series of individual tests for each kind of error, and each error has to be estimated so that you can add them all together.

And this.. If you use an artifical star, it needs to be placed about 20 times the focal length to work for the SA test. I bit closer is Ok, but 20x is best to eliminate a false reading. Some of this depends on the pinhole size though. The smaller the pinhole the better.

As you can see though, for 20 times the focal length, you may need to be up to 180 feet away from your point source.

This is why I just use a real star. No need to worry about it.

#8 Julio

Julio

    Messenger

  • *****
  • Posts: 452
  • Joined: 18 Jan 2007
  • Loc: Pembroke Pines ,FL

Posted 14 January 2013 - 04:55 PM

Interesting Ed, where do you notice the zone ? On the second set of links i provided? Is it in the intra focal dark band (ring) seen right in the right middle aspect of the defocused star?

#9 Eddgie

Eddgie

    Voyager 1

  • *****
  • Posts: 12488
  • Joined: 01 Feb 2006

Posted 14 January 2013 - 06:10 PM

It is visible in one of the later images with more defocus than the ones at the beginning of the thread.

An "S" zone is detected by looking at the Fresnel pattern with about 20 wavelents of defocus (it does not have to be exact at all, not even close).

For zones, what you are looking for is that the rings between the inner and outer bright rings look evenly flat. If you see some of the faing rings as appearing brighter or darker than the neighbors, this is the zone.

It is not always easy to see exactly where a zone is, but in general it is reflected by this aspect of the pattern.

The rings across the entire Fresnel pattern being evenly thinner on one side and evenly thicker on the other, but all being the same brightness on one side or the other is how you see a turned edge (TE_) which is a form of zone.

So, you are looking for the even brightness of the rings between the inner and outer rings on the same side of focus.

#10 orion61

orion61

    Aurora

  • *****
  • Posts: 4500
  • Joined: 20 Oct 2007
  • Loc: Birthplace James T Kirk

Posted 14 January 2013 - 07:08 PM

Yes I saw it about 2/3 of the way out on the Extra focus image, mot major from what I saw.
I have certainly seen a LOT worse. :waytogo:

#11 Julio

Julio

    Messenger

  • *****
  • Posts: 452
  • Joined: 18 Jan 2007
  • Loc: Pembroke Pines ,FL

Posted 15 January 2013 - 01:58 AM

Interesting data, i appreciate everyone's insight. I will update you guys when i get a good seeing night. Lately it has been 2-3/5 seeing with clouds rolling by and twinkling of the fainter stars. Even with all this i can see the E and F stars in the trap with the scope in question every time ive used it. Of course this is normal out of 6+ inch scopes.

#12 Eddgie

Eddgie

    Voyager 1

  • *****
  • Posts: 12488
  • Joined: 01 Feb 2006

Posted 15 January 2013 - 08:42 AM

As I mentioned, this zone is quite common on larger Celestron scopes. I doubt that by itself it does any meaningful harm, but that is only by itself. It all adds up though. If there are no other issues (TE, SA, Roughness, etc) then the zone won't matter.

But all to often, these other errors do show up in Celestron (and Meade) SCTs and again, usually none of them is serious by itself, but because they all add together, it turns them into scopes that vary widely in quality.

Most are good to very good, and many are downright excellent.

Some though are only passable, and honestly, I can't say I have ever seen a truely bad one.

But the problem is that assessing the quality when multiple errors exist is complex. You have to measure each error then approximate the damage to the MTF curve which is then added to the curve from the other optical abberations.

SA is the major killer in an SCT though, if it is bad enough. And it is rarely that bad, but if it is in the middle and there are other problems (roughness, zone, turned edge, etc) it makes for an SCT that has some more easiy reduced contrast performance. Enough that a good observer will see the differnce.

#13 mattflastro

mattflastro

    Vendor - Astrovideo Systems

  • -----
  • Posts: 622
  • Joined: 31 Jul 2009
  • Loc: Brevard County , FL

Posted 15 January 2013 - 08:52 AM

As I mentioned, this zone is quite common on larger Celestron scopes. I doubt that by itself it does any meaningful harm, but that is only by itself. It all adds up though. If there are no other issues (TE, SA, Roughness, etc) then the zone won't matter.

But all to often, these other errors do show up in Celestron (and Meade) SCTs and again, usually none of them is serious by itself, but because they all add together, it turns them into scopes that vary widely in quality.

Most are good to very good, and many are downright excellent.

Some though are only passable, and honestly, I can't say I have ever seen a truely bad one.

But the problem is that assessing the quality when multiple errors exist is complex. You have to measure each error then approximate the damage to the MTF curve which is then added to the curve from the other optical abberations.

SA is the major killer in an SCT though, if it is bad enough. And it is rarely that bad, but if it is in the middle and there are other problems (roughness, zone, turned edge, etc) it makes for an SCT that has some more easiy reduced contrast performance. Enough that a good observer will see the differnce.

SA might not be bad by design but users make it really that bad if not worse. Adding 4 inches to the backfocus by inserting all sorts of gizmos like focal reducers, coma correctors, filter wheels , robofocusers and camera adapters adds 1/4 lambda of SA. These scopes were not designed with imaging in mind or at least not with the kind people are doing today.

#14 Eddgie

Eddgie

    Voyager 1

  • *****
  • Posts: 12488
  • Joined: 01 Feb 2006

Posted 15 January 2013 - 02:57 PM

Actually, the were indeed designed for imageing. The best performance is at about 4" of back focus.

When they were designed though, film was in use, and it was common to have to crop the corners.

Many imagers at the time also did their own lab work because the best film at the time was Gas-hypered (saturated to increase sensitivity) Tech Pan film. It was very easy to process, and this was important because most film processing labs simply did not know what to do with it because the film was always "Black". Many imagers would process their own Tech Pan because it was pretty easy to do.

I got back several rools where they said that I must have forgotten to take the cover off the camera lens. :roflmao:

I agree though, that these scopes should deliver their best performance when used at the native focal lenght.

And yet people do routinely use them at much longer focal lengths than the design length.

See, while people obsess about optical quality (I know I do), the brutal reality is that unless an error is pretty severe, the damage at the focal plane is not always easy to see.

The damage being done by adding only a little back focus to and SCT will be completely undectable in anything but a star test.

Add maybe 300mm though, and the damage should be more evident. For someone wanting the absolute best performance from their SCT, they should stay under 200mm of back focus just to be safe. They get 100mm already, and this gives them 100mm to play with. This means a 2" diagonal won't induce enough SA to be detectable on the in focus image.

The C14 has a native 150mm of back focus (was designed for 2" diagonal use), but it also has tight baffles. It will start loosing aperture when the back focus goes over about 200mm.

And this is a far more serious issue with a lot of back focus on SCTs. Most of them start to loose effective aperture when the back focus goes over between 180mm and 200mm.

So, the change in spherical abberation to me is not as meaningful as the loss of aperture you can experience using a lot of back focus.

#15 Julio

Julio

    Messenger

  • *****
  • Posts: 452
  • Joined: 18 Jan 2007
  • Loc: Pembroke Pines ,FL

Posted 16 January 2013 - 03:02 AM

Ed I went ahead and tried what you suggested, I used a screw a nut and washers to create a well measured 4.5mm spacer for the EP. I followed your directions and took some shots with 10 wave of defocus to asses SA. Again this is a-focal, im a visual guy and don’t have fancy cameras. I stacked the video frames for some clarity; Fresnel patterns are more obvious visually on both sides of focus. Seeing was 3/5, used a 10mm Plossl only using the diagonal and 1.25 visual back with 4.5mm of in and out focus travel. I find all this quite interesting, thanks for the great information. Ill attempt this again as soon as seeing improves, the stars on my well collimated SCTs look fuzzy with no defined airy disk even after scope cool down.

Extra 4.5mm spacer

http://i99.photobuck...zpse4c199d0.png

Intra 4.5mm Spacer

http://i99.photobuck...zps59b843d7.png


http://i99.photobuck...kedImg_zps57...

http://i99.photobuck..._zps2dbe1760...

#16 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 8053
  • Joined: 08 Aug 2007
  • Loc: La Union, PI

Posted 16 January 2013 - 06:43 AM

I'd have to let Eddgie correct me if I am wrong, but there seems to be some under correction. That's pretty much expected and normal in a system that uses spherical surfaces (SCTs, MCTs, even APOs.) The bright outer ring outside focus, I believe, is the marginal zone (which contains a lot of light energy from the larger outer mirror surface) leaving focus and expanding outward. The inner bright ring is the more narrowly converging paraxial zone just leaving focus (with lesser energy allowing the shadow to peek through a bit further from the primary, hence larger at break out.)

I believe the shadow inside focus is also constrained in size by the marginal zone coming to focus quite rapidly and nearer the primary's surface. It remains well defined until swamped by the collapsing, bright marginal rays. Outside of focus, this marginal ray (containing lots of energy) is already spread out while the inner rings are formed by the central (paraxial rays with less energy) are just beginning to spread away from focus. I believe this lesser energy allows the shadow to break out and become larger than equal intra defocus. So, for under correction, the shadow will be smaller inside and larger outside when it breaks out at different distances from focus. The worse the (under) correction, the more energy at the center of the pattern outside focus and the longer it takes to allow the shadow to break out.

In any case, the inner and outer rings are not that noticeably different in intensity. Correction seems pretty good. The way I try to remember normal under correction: more energy in the /outer/ rings inside of focus, and the opposite, more energy in the /inner/ rings the outside focus. The more these vary, the worse the correction. It's hard to put a figure on the correction seen, but I'd bet it's better than 1/4 wave.

There is a slight zone at about 30%D. The light ring outside focus is replaced by a dark ring inside focus. Not to worry, they are common and probably contribute almost nothing to RMS.

My two cents based on what's shown.

#17 mattflastro

mattflastro

    Vendor - Astrovideo Systems

  • -----
  • Posts: 622
  • Joined: 31 Jul 2009
  • Loc: Brevard County , FL

Posted 16 January 2013 - 08:45 AM

Actually, the were indeed designed for imageing. The best performance is at about 4" of back focus.

I didn't say they have SA when used at 4" of back focus.
I said you ADD 1/4 wave of SA if you ADD 4" extra of back focus on top of the original design back focus. Do a raytrace and you will see .

#18 Eddgie

Eddgie

    Voyager 1

  • *****
  • Posts: 12488
  • Joined: 01 Feb 2006

Posted 16 January 2013 - 09:23 AM

Actully, you said:

These scopes were not designed with imaging in mind...



I was responding mostly to this statement.

#19 mattflastro

mattflastro

    Vendor - Astrovideo Systems

  • -----
  • Posts: 622
  • Joined: 31 Jul 2009
  • Loc: Brevard County , FL

Posted 16 January 2013 - 02:13 PM

Actully, you said:

These scopes were not designed with imaging in mind...



I was responding mostly to this statement.


You left out of the quote exactly what mattered most .
I said :

"These scopes were not designed with imaging in mind or at least not with the kind people are doing today."

You know very well I was referring to the fact that the 105mm back focus was designed in the days of 35mm film cameras whereas nowadays people attach to them microfocuser, filter wheels, adaptive optics , flip mirrors or diagonals, focal reducers/field flatteners/correctors .
These in most cases add up to way more than 105mm and lead to increased SA . Additional 4" of backfocus adds an additional 1/4 wave of SA.

#20 Julio

Julio

    Messenger

  • *****
  • Posts: 452
  • Joined: 18 Jan 2007
  • Loc: Pembroke Pines ,FL

Posted 16 January 2013 - 02:57 PM

Sure hope you're right Norme regarding better than 1/4 wave P/V correction.Im going to continue these tests when i receive my artificial star.

#21 Eddgie

Eddgie

    Voyager 1

  • *****
  • Posts: 12488
  • Joined: 01 Feb 2006

Posted 16 January 2013 - 05:06 PM

Well, I apologize for any misunderstanding on my part.

#22 mattflastro

mattflastro

    Vendor - Astrovideo Systems

  • -----
  • Posts: 622
  • Joined: 31 Jul 2009
  • Loc: Brevard County , FL

Posted 17 January 2013 - 11:08 PM

Well, I apologize for any misunderstanding on my part.

No need to apologize for such a small thing , I should be apologizing as well but on a more serious note your star testing skills and knowledge are amazing .

#23 Julio

Julio

    Messenger

  • *****
  • Posts: 452
  • Joined: 18 Jan 2007
  • Loc: Pembroke Pines ,FL

Posted 18 January 2013 - 11:55 AM

I guess the lingering question is does the scope meet the difraction limited spec? :question:

#24 Eddgie

Eddgie

    Voyager 1

  • *****
  • Posts: 12488
  • Joined: 01 Feb 2006

Posted 18 January 2013 - 12:07 PM

Unless you have some reason to suspect that there is a serious error, then I would say that it is improbable that it has a flaw so serious that the scope is no longer diffraction limited (which in itself is rather vague, but is generally understood to equate to a Strehl of .8).

While your test was not done to the standard necessary to give a meaningful quantification of the errors I mentioned (maybe a bit of SA and a small zone) neither looked serious.

Your choices are to do more testing and attempt to quantify the error, or just start using your telescope and forget about it. I did not see a serious error in the scope.

If you want to see more tests, I recommend you go to this web site. It has plenty of star tests with different errors shown, and as you will see, your scope's test looks to show a scope with typical SCT quality. Not bad, but not great, and and almost always much better than diffraction lminted .8 Strehl) from Celestron.

Astro-Foren Test Directory.. Never seen a Celestron below .8, and most are .9...

There is also the Roddier test.

But you are wasting your time. Nothing in your test suggests the telescope is a lemon.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics