Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Concise thread about autocollimators+improvements

This topic has been archived. This means that you cannot reply to this topic.
434 replies to this topic

#351 johnnyha

johnnyha

    Fly Me to the Moon

  • *****
  • Posts: 6,564
  • Joined: 12 Nov 2006

Posted 28 January 2011 - 07:49 AM

Nils I agree with you in regards to wishful thinking. I have been following this thread with great interest. I wonder if the bevel itself (which is used with the template to gauge the center) is ground to the tolerances discussed here? Or the actual circumference or "absolute roundness" of the mirror blank? Is it PERFECT? Or for that matter the actual center of the polish strokes? Is the center hole in the hotspot punched out to this tolerance? In other words, is there any point in trying to achieve sub-1/10 millimeter tolerances for the "center" spot when it is not entirely clear when you achieve this, that it IS the absolute 100% perfect optical center? ...Or is it?

#352 Nils Olof Carlin

Nils Olof Carlin

    Vanguard

  • *****
  • In Memoriam
  • Posts: 2,227
  • Joined: 26 Jul 2004

Posted 28 January 2011 - 08:23 AM

johnnyha,

with machine-figured mirrors, the precision of figuring must depend on centering the blank (the periphery - ignore any asymmetry of the bevel) on the machine table - I have no guess here. With hand-figured mirrors, there may be even less precision! But anyway, while reports of badly centered spots are common, reports of actual decentering, large enough to call attention, seems rare to non-existing (I'd be interested to learn of any case).
I can't think of any method other than careful star-collimating under excellent seeing to find the optical center to any useful precision (I believe casual star-collimation as often recommended to finish instrument-collimation is a total waste of effort). I have no good data myself, but for a f/4.5 mm mirror, a 2 mm error would mean a Strehl ratio loss of 0.2 for a perfect mirror - should be readily discernible. A 1 mm error (there is half the error at the focal plane) probably discernible in good conditions. A few tenths of a millimeter - I would not believe so.
Perhaps a good research project for someone blessed with good seeing (and willing to waste a night of it on the altar of Science) and a good mirror? Done "blindly" by deliberate precision miscollimation (in different directions to balance a possible innate optical centering error) not known to the observer.

Or has anyone here done any systematic studies already?

Nils Olof

#353 UmaDog

UmaDog

    Vanguard

  • *****
  • Posts: 2,054
  • Joined: 15 Sep 2010

Posted 28 January 2011 - 12:44 PM

The bevel on my XX12 is appalling and this made placing the centre spot rather tricky. Did it OK in the end, I think.

#354 Jason D

Jason D

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 7,418
  • Joined: 21 Oct 2006

Posted 28 January 2011 - 01:35 PM

The bevel on my XX12 is appalling and this made placing the centre spot rather tricky. Did it OK in the end, I think.


Jam a geometrically-sound block against the mirror with the proper height then use the block edge as your reference.

Attached Thumbnails

  • 4346449-bevel.png


#355 johnnyha

johnnyha

    Fly Me to the Moon

  • *****
  • Posts: 6,564
  • Joined: 12 Nov 2006

Posted 29 January 2011 - 05:34 AM

Holy crimoley!

I have been having a long spate of bad seeing here, everything was just fuzzballs for months and months. Nothing over 120x. Then last night something Nils said jogged my memory.... Hmm. The last time I center-spotted my mirror was about four month ago. I had used the template for the first time and it placed the center spot about 1.5-2mm away from the engraved center of my OMI mirror. Could very well be my own human error but I had been reasonably careful so I went ahead and kept the center hotspot in this new position. Hmmmm...! Today I cleaned the mirror and moved the center hotspot back to the factory engraved center about 2mm away - and the difference in view is *dramatic*. I am looking at Saturn right now this morning at 250x and it is absolutely razor sharp. The mirror is performing as well as I have ever seen it, just gorgeous and super sharp. Wow. I'm very happy and the moral of the tale is, place the center spot *very* carefully a few mm can make a crazy amount of difference. What I thought was months of bad seeing (and maybe even incurable astigmatism) was just a very slightly shifted center spot. :foreheadslap: :p

#356 lightyear44

lightyear44

    Apollo

  • *****
  • Posts: 1,127
  • Joined: 29 Dec 2009

Posted 29 January 2011 - 05:59 AM

I have taken a tape-measure to the centre mark on my lightbridge in the past, and remember it being ok. But your post makes me want to double-check my measurements. Fascinating story, Johnny. -David.

#357 Starman1

Starman1

    Vendor (EyepiecesEtc.com)

  • *****
  • Vendors
  • Posts: 48,151
  • Joined: 23 Jun 2003

Posted 29 January 2011 - 10:33 AM

That mirrors a comment I made a few years back: that I was amazed at how much the seeing improved once I learned to use the autocollimator to achieve good alignment.

Nils Olof's earlier comments about how the mispositioning of the center marker will influence the Strehl ratio of the final image is certainly apropos.
Say you have a superlative-quality mirror and your collimation isn't right--you will never see what the mirror is truly capable of.

Seeing the most out of a mirror is the best justification for collimation accuracy, center marker positional accuracy, cooling fans, and coma correctors.
When everything is as good as it can be, and you see superb seeing conditions also, it's magic.

Burned into my brain is an image of Jupiter through my 12.5" at 228X, wherein the amount of detail seen was close to a truly superb photograph, only sharper. White swirls in the GRS, multiple small storms in the EQ band, and polar shading broken up into a mackerel sky pattern of storms and eddies. All in color.
I could not even have begun to draw what I saw.

And my centermark is probably off by about 0.5mm.

And therein lies my interest in making the center marker accuracy as high as possible--if having a .5mm error results in near-perfect star images and that kind of detail on Jupiter, what would be the results in similar seeing of having an even higher accuracy on the center marker's position?

#358 Jason D

Jason D

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 7,418
  • Joined: 21 Oct 2006

Posted 29 January 2011 - 11:07 AM

I had used the template for the first time and it placed the center spot about 1.5-2mm away from the engraved center of my OMI mirror. Could very well be my own human error but I had been reasonably careful so I went ahead and kept the center hotspot in this new position.

I recommend the technique I described in multiple posts earlier.
1- Place blocks around the mirror with about the same height -- maybe little lower
2- Place the template and place weights on the top of the blocks to hold the template in place
3- When it is time to press, press a little vertically and ensure the template is still centered before committing the center spot.

John, I do not doubt you centered the template but I believe when it was time to press and commit, the template slipped.

Jason

#359 UmaDog

UmaDog

    Vanguard

  • *****
  • Posts: 2,054
  • Joined: 15 Sep 2010

Posted 29 January 2011 - 11:46 AM

Argh! I just photographed my mirror and measured the spot position along two axes.
It's 0.06 mm from centre along L/R but 3.35 mm away along U/D.
I suppose I'd better go back and do it again...

#360 Jason D

Jason D

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 7,418
  • Joined: 21 Oct 2006

Posted 29 January 2011 - 12:04 PM

3.35 mm away along U/D

It is not clear how such a large error can be introduced.
Yes, you need to redo it.
3.35mm error will introduce ~1.6mm PAE with is significant.
If you redo it, consider the method I have described.

#361 UmaDog

UmaDog

    Vanguard

  • *****
  • Posts: 2,054
  • Joined: 15 Sep 2010

Posted 29 January 2011 - 12:10 PM

I don't think I have a nice square block but I shall try weighing down on one side only. Also, I think it's of help that I've measured the error and can check if I see it using the template.

#362 Jason D

Jason D

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 7,418
  • Joined: 21 Oct 2006

Posted 29 January 2011 - 12:14 PM

It does not have to be a "nice" block. A "nice" block only helps to get around the bevel readability.

In my case, I used 2x4 wood blocks (1.5" width). Weigh one side but shift the template 0.1 away from the weighted side. The 0.1mm shift is a recommendation but not a requirement.

#363 johnnyha

johnnyha

    Fly Me to the Moon

  • *****
  • Posts: 6,564
  • Joined: 12 Nov 2006

Posted 29 January 2011 - 12:53 PM

That mirrors a comment I made a few years back: that I was amazed at how much the seeing improved once I learned to use the autocollimator to achieve good alignment.


:lol:

Another thing that tipped me off was, I had recently started using the Paracorr and I noticed it was displaying some field curvature or something, the image was shifting in parts of the field. After I shifted the center spot on the mirror the Paracorr performed great and for the first time I am able to describe the view as near "refractor-like". Incidentally my new 10mm Ethos was amazing in the Paracorr at 200x, the Trapezium including the E and F star were perfectly sharp and visible all the way to the edge, with very little distortion if any at the edge. Saturn's Cassini Division was a very thin black line, the rings are only slightly tilted yet but I could see glimpses of what I call the "record groove" effect. the disc showed some detail and beautiful color, and Saturn's tiny moons were sprinkled like salt around the planet, it was mesmerizing. Again thanks to everyone in this thread, especially Nils for jogging my memory with his comment on 2mm shifted center spot causing noticable strehl loss.

#364 Vic Menard

Vic Menard

    Cosmos

  • *****
  • Posts: 7,631
  • Joined: 21 Jul 2004

Posted 29 January 2011 - 03:01 PM

Johnny,
have you measured the OMI indicated center spot placement?

...The last time I center-spotted my mirror was about four month ago. I had used the template for the first time and it placed the center spot about 1.5-2mm away from the engraved center of my OMI mirror. Could very well be my own human error but I had been reasonably careful so I went ahead and kept the center hotspot in this new position. Hmmmm...! Today I cleaned the mirror and moved the center hotspot back to the factory engraved center about 2mm away - and the difference in view is *dramatic*...

Just wondering if you've determined whether it was "human error" or is the engraved center actually 2mm off of the mirror blank's measured center?

#365 UmaDog

UmaDog

    Vanguard

  • *****
  • Posts: 2,054
  • Joined: 15 Sep 2010

Posted 29 January 2011 - 03:03 PM

The photos I take show a shift of about 2 to 3.7 mm in my spot position. I'm pretty sure I'm calculating it correctly.

However, I'm having enormous difficulty verifying that with the Catseye template. The reason is that the mirror slightly larger than 12" and the coating is not a consistent distance from the bevelled edge. No matter how much I re-measure, the current spot position seems to be right one to under a 1 mm.

EDIT
Ah! It turns out that the spot isn't so far off after all. What matters, it seems, is that the camera is pointing straight at the middle of the mirror. If it isn't, the calculated spot position is off. Using an older image where the camera was positioned in the focuser and the scope was collimated, I derive a spotting error of about 0.48 mm. I've gone on to verify that with the mirror removed from the scope. Based on a previous thread, I think this puts me within tolerance even for a Paracorr. So I'm calling it a day on this one!


#366 johnnyha

johnnyha

    Fly Me to the Moon

  • *****
  • Posts: 6,564
  • Joined: 12 Nov 2006

Posted 29 January 2011 - 03:32 PM

Vic I tried looking straight down the center from directly overhead with the template and it did look like the engraved center was off a hair from the template center, and then I think when I pressed it down I pushed it off a bit more. At the time I had called OMI and spoke to someone there about the engraved center spot, they said it should be the same as physical center, and he did say it would be a good idea to defer to the engraved spot. I have another theory which is the warp of the template, it was sent curled up in a tube and it flattened out pretty good but may still have had a slight curve to it. Even looking down the center from directly overhead I can't be 100% sure I'm within a mm... another good reason to use blocks or something to hold the edges. I don't have the mirror out to recheck right now but I'm guessing more human error than a misplaced engraved center mark. :p

#367 Vic Menard

Vic Menard

    Cosmos

  • *****
  • Posts: 7,631
  • Joined: 21 Jul 2004

Posted 29 January 2011 - 04:13 PM

...At the time I had called OMI and spoke to someone there about the engraved center spot, they said it should be the same as physical center...

This is pretty much what I suspected--thanks for the confirmation!

#368 Jason D

Jason D

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 7,418
  • Joined: 21 Oct 2006

Posted 29 January 2011 - 04:35 PM

But the question is how accurate is that 'X'? Did OMI use some sort of high-precision apparatus to place the mark? Was the mark made by hand?
I am afraid without verification; the X location should not be accepted as the unquestionable center location.

#369 johnnyha

johnnyha

    Fly Me to the Moon

  • *****
  • Posts: 6,564
  • Joined: 12 Nov 2006

Posted 29 January 2011 - 06:22 PM

Well when I say "engraved" I guess that's a little misleading, the "x" or cross in the center is actually two very fine, small straight lines in the coating that intersect, they were clearly scratched in precisely with a blade or tool. It's not a perfect x or + and it's not stamped. I have no idea what method they used to determine the center but the results speak for themselves. On a side note, when the laser hits the center of the "x" it's apparent because the light "sparkles".

#370 Vic Menard

Vic Menard

    Cosmos

  • *****
  • Posts: 7,631
  • Joined: 21 Jul 2004

Posted 29 January 2011 - 07:02 PM

Jason,
The marks I've seen are diamond etched into the glass before the mirror is aluminized. I'm guessing it's done on a measuring table of some sort. If you were to ask OMI, I'm guessing they would put the accuracy somewhere around +/- one or two hundredths of an inch...

#371 Jason D

Jason D

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 7,418
  • Joined: 21 Oct 2006

Posted 30 January 2011 - 01:04 AM

With all this talk about how some of you went back to assess the centerness of Catseye center spot and found out it is few millimeters off-center concerned me.
So, I took out my Orion stock primary mirror to assess the placement of my yellow HotSpot. I found the placement to be really good. See attachment.

Attached Thumbnails

  • 4349858-centerness1.jpg


#372 Jason D

Jason D

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 7,418
  • Joined: 21 Oct 2006

Posted 30 January 2011 - 01:11 AM

Same setup but with super close-up. I see a slight shift which I assessed to be around 0.2mm. Well, I noticed that the discrepancy between the template and the HotSpot was within 0.2mm after several retries. That tells me that 0.2 seems to be the human-error factor when placing the center spot. But that will translate to 0.1mm error at the eyepiece.
But 2mm and 3mm errors sound excessive and should not be expected when Catseye template is used correctly. Again, I suggest using blocks and weighing down one side of the template.

Attached Thumbnails

  • 4349862-centerness2.jpg


#373 Jason D

Jason D

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 7,418
  • Joined: 21 Oct 2006

Posted 30 January 2011 - 01:12 AM

I redid the earlier experiment but this time I used the real mirror instead of a plate. See attachment. I estimate the error due to sagitta to be around 0.1mm. I do not believe it is worth doing anything about.
(In the photo, the ruler spacing is 0.5mm)

Attached Thumbnails

  • 4349864-centerness3.jpg


#374 Starman1

Starman1

    Vendor (EyepiecesEtc.com)

  • *****
  • Vendors
  • Posts: 48,151
  • Joined: 23 Jun 2003

Posted 30 January 2011 - 01:48 AM

On my mirror, the engraved X at the center is over 3mm off-center. I wouldn't trust the X unless you've measured it.
Obviously, the technique used to place the X wasn't very accurate, in my case.
Yours may not be any better.

#375 Jason D

Jason D

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 7,418
  • Joined: 21 Oct 2006

Posted 05 September 2011 - 05:07 PM

The subject of center spot placement accuracy comes up in this forum periodically. The typical recommendation is to use the Catseye template. But some are concerned that the template might not provide the desired accuracy mainly because of possible template shift when it is pressed.
Recently, I pulled out my Stevens premium mirror to install and evaluate a Zambuto loaner mirror.
Before I returned my Stevens mirror back, I re-evaluated the centerness of my Hotspot which was originally placed using Catseye template. See attached photos in this post and the following one.
I am happy with the accuracy of Catseye template.
The reason I chose to add my new comments to this old thread because the last few pages were dedicated to this subject. I did the same evaluation with respect to my stock mirror Hotspot. Two separate mirrors were spotted using Catseye template and both mirrors were re-evaluated for placement accuracy and both have passed the evaluation with flying colors.
Bottom line: If it is done right, Catseye template is a cheap, easy, and an accurate method to center spot primary mirrors.

Jason

Attached Thumbnails

  • 4790961-cat1.jpg



CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics