Click here if you are having trouble logging into the forums
Privacy Policy |
Please read our Terms
of Service | Signup and
Troubleshooting FAQ | Problems? PM a Red or a Green Gu… uh, User
BDS316
Carpal Tunnel
Reged: 09/16/09
Loc: Sol 3
|
Re: autocollimator question
[Re: Jason D]
#5213964 - 05/09/12 02:50 PM
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Quote:
Bryan,
To get the best result with your autocollimator, you want to observe a fixed image that follows the AC rotation. Let me explain:
The first row represents what I understood from your description. You start off with stacked reflections (top/left illustration) then you observed unstacked reflections when rotated the AC by 180 degrees (top/right illustration).
I am suggesting that you follow what in shown on the second row. Start off with unstacked reflections (P & 2) then maintain the distance between both reflections as you rotate the AC.
Jason
Thanks, I get it. I have Bob's knobs and a washer on the secondary. Is there a trick to it or trial and error?
|
Vic Menard
Post Laureate
Reged: 07/21/04
Loc: Bradenton, FL
|
Re: autocollimator question
[Re: BDS316]
#5213981 - 05/09/12 02:59 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Quote:
So my Xt8 is collimated correctly.. Thanks!!
BTW the Moonlite focuser uses an adapter that does not rotate.
The fact that it doesn't rotate doesn't mean the 1.25-inch bore is coaxial with the 2-inch bore, it just means you can't verify it by rotating it. You'll need an alignment tool that allows you to assess both bores, like a combination 2-/1.25-inch laser.
Quote:
Now following up on this question, if I were to use my AC on other scopes owned by my fellow club members with faster scopes, say, f/5ish, would the same hold true?
This is an important question for me since I might in the future get a 12.5 inch f/5.
That depends. A 5 minute focuser axial tilt error will cause a linear deflection of about 0.07-inch at the primary mirror center for a 12.5-inch f/5. That's easily within tolerance if you're not using a Paracorr. With a Paracorr the high performance tolerance is almost equal to the error, which means you would need a near perfect alignment. And at the greater focal length distance, those reflections get harder to see, and harder to correct.
Jason's solution (reflections rotate around the primary mirror center spot) delivers a better focuser axial alignment, but is equally difficult because of the longer focal length.
And we still don't know the 2- to 1.25-inch adapter's contribution to the observed error (or if it has none).
|
Jason D
Postmaster
Reged: 10/21/06
Loc: California
|
Re: autocollimator question
[Re: BDS316]
#5213992 - 05/09/12 03:01 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Quote:
Is there a trick to it or trial and error?
Look at my illustration. Stack the reflections then rotate by 180 degrees (as you have been doing). Adjust the secondary mirror to reduce the distance between reflections P and 2 by 1/2 but maintain direction. That will puts you in the ballpark. You will need to reiterate between the cheshire and the autocollimator.
In a way, it is trial and error.
|
Starman1
Vendor (EyepiecesEtc.com)
Reged: 06/24/03
Loc: Los Angeles
|
Re: autocollimator question
[Re: Jason D]
#5214279 - 05/09/12 06:34 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Back in the '80s I was talking to, IIRC, Tom Clark at RTMC, one year, about the reflections in the Tectron autocollimator.
He asked me to show him the tool.
I handed it to him, he put it in his scope and turned it around, took it out of the focuser and threw it in a box of scrap metal he had. He then gave me a new one.
The issues I had with collimation on my f/5 scope then sort of "went away".
So even then, a lot of the ACs from Tectron didn't pass muster.
Unfortunately, even today there does not appear to be a 1.25" AC that is consistently good from unit to unit.
If you can swing the price, and since you have a nicely-machined Moonlite focuser, I'd suggest replacing the Tectron with the Catseye Infinity XL or XLK autocollimator.
It's like moving up from a plastic Huygens eyepiece to a TeleVue Nagler. [Well, maybe a slight exaggeration, but not much!]
|
1stScope
super member
Reged: 04/22/07
|
Re: autocollimator question
[Re: Jason D]
#5215275 - 05/10/12 11:20 AM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
"I have just one remaining question: Since the field is dark and since I am able to stack the images at one piont in the circular travel....
Am I collimated?"
No, you can only state that you’re collimated within the error of your AC, but as mentioned previously can try to calibrate your bad AC with some alignment marks. Ideally your focuser tube will be one piece so that the end that the eyepiece rests on has a ‘turned’ finish on it, as a decent cut off on a lathe produces a surface that is very perpendicular to the axis of the part, less than a few ten thousandths of an inch. If the focuser tube is assembled then the following might not be worth the effort.
To calibrate it find a small mirror that you can lay on top of the focuser tube, ideally one that allows you to judge how well centered it is. I used a mirror from one of my wife’s makeup compacts and just scratched a small hole thru the silvering. Dental mirrors, mechanics mirrors, etc., are other sources. By being able to slide the mirror around on the end of the focuser you’ll get a good feeling for how well you need to center the hole by seeing the images ‘unstack’, which is a handy ability to have anyway.
The critical parameter of an AC is that the mirror is perpendicular to the axis of the focuser tube, and a less critical parameter is how well centered the hole in the AC is. Ideally one would have a longer tube that fit inside the focuser, an end with a nice cut off finish on it, a first surface mirror justified on the end of the tube, and a centered peep hole. The problem that we see with a lot of commercial ACs is that they glue a mirror to a surface which may be turned and thus perpendicular but the gluing adds a tolerance to the assembly, or worse yet use an assembly where the assembly tolerance adds to the gluing tolerance. The closer that one gets to the ideal the better. The error from using a second surface mirror is small, and if you’re worried about it you can check the parallelism of the surfaces with a micrometer (looked good with my older B&S mics with a vernier).
To calibrate your commercial AC just collimate using the mirror placed on the end of the focuser, if you have a helical focuser (like I do) you can rotate it to double check mirror alignment (not a problem). Once collimated place your commercial AC in the focuser and rotate it until it appears collimated, although it might not be possible to do so. Once you find a good spot add some alignment marks to the focuser and the commercial AC and only use it when aligned. If you don’t find a spot where it is collimated you can use your home made AC until you get around to buying a better commercial AC. Just rotate the commercial AC in the focuser to check for alignment of the AC mirror.
|
Vic Menard
Post Laureate
Reged: 07/21/04
Loc: Bradenton, FL
|
Re: autocollimator question
[Re: 1stScope]
#5215647 - 05/10/12 02:23 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Quote:
Am I collimated?"
No, you can only state that you’re collimated within the error of your AC...
In this case, the AC error has been interpreted to be 5 arc minutes. This tilt error will produce a linear focuser axial error measured from the center of an 8-inch f/6 primary mirror of approximately 0.06-inch--better precision than a sight tube and almost certainly better precision than most economy lasers. The "high-performance" tolerance is approximately 0.25-inch (without coma correction). When used in tandem with a Cheshire eyepiece, both axial alignments will be well within the high-performance tolerances.
Quote:
...but as mentioned previously can try to calibrate your bad AC with some alignment marks.
You can only do this if the 2- to 1.25-inch adapter (or the focuser drawtube) also has a tilt error, where one tilt error can be canceled by another tilt error.
Quote:
By being able to slide the mirror around on the end of the focuser you’ll get a good feeling for how well you need to center the hole by seeing the images ‘unstack’, which is a handy ability to have anyway.
As the pupil is moved off axis (assuming the top edge of the focuser drawtube is precisely perpendicular to the primary mirror axis), the reflections will move apart into two individually aligned stacks. This is the signature for parallelism and is the function of the second off axis pupil in the Infinity XLK autocollimator.
Quote:
The critical parameter of an AC is that the mirror is perpendicular to the axis of the focuser tube, and a less critical parameter is how well centered the hole in the AC is.
The pupil defines the location of the focuser axis. If the pupil is not precisely centered (and the top edge of the focuser drawtube is precisely perpendicular to the primary mirror axis), the background reflections will only be able to be merged into a single aligned stack when the pupil is centered. The pupil should also be "small" to minimize the rapid disappearance of the two fainter background reflections which happens when the primary mirror center spot is brought into alignment with the first, brightest background reflection. It's common for this to happen when stacking the reflections--the resulting alignment indicating the two axes (focuser and primary mirror) intersect at the center of curvature. It's also the reason for finishing the alignment with a calibrated Cheshire eyepiece.
Quote:
To calibrate your commercial AC just collimate using the mirror placed on the end of the focuser, if you have a helical focuser (like I do) you can rotate it to double check mirror alignment (not a problem). Once collimated place your commercial AC in the focuser and rotate it until it appears collimated, although it might not be possible to do so.
Assuming your second surface mirror has no tilt component (5 arc minutes across 2 inches is a variance of only 0.003-inch) and you get the pupil properly centered and get a perfect axial alignment, the commercial AC with a known 5 arc minute tilt error will never show a perfect stack no matter how it's rotated (assuming a perfect 2- to 1.25-inch adapter). The reflections will instead "orbit" around the aligned primary mirror center spot (Jason's solution).
Quote:
...If you don’t find a spot where it is collimated you can use your home made AC until you get around to buying a better commercial AC. Just rotate the commercial AC in the focuser to check for alignment of the AC mirror.
Or, the OP can reiterate the axial alignment between the AC and the Cheshire, and either stack the reflections to reduce the focuser axial error to 5 arc minutes, or unstack the reflections and verify the narrowed excursions under rotation to reduce the focuser axial error to less than the 5 arc minute tilt error. Either way, the axial alignment will be well within the high performance prescribed tolerance.
FWIW, even when the Tectron autocollimator is perfectly aligned, it's very difficult to see the third reflection of the primary mirror center spot (there are four reflections including the primary mirror center spot, which, like the other reflections, is seen reflected via the secondary mirror). If you can't see the third reflection, you can't easily see the focuser axial alignment (via the CDP procedure). The narrow field of view and the second surface mirror seem to be the limiting factor (although I've also used a 1.25-inch AC with a first surface mirror that was equally troublesome). This means that most 1.25-inch AC users simply reiterate between a "close stack" and a precision Cheshire alignment--which delivers pretty good results. And when they actually want to see, and align, the focuser axis and parallel axes signatures, they move up to the XLK autocollimator.
|
BDS316
Carpal Tunnel
Reged: 09/16/09
Loc: Sol 3
|
Re: autocollimator question
[Re: Vic Menard]
#5215681 - 05/10/12 02:40 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Hi, thanks for all the info.
Up til now I have been using the Tectron Sight tube and Cheshire and getting good results which are confirmed by star collimation. The next time I observe I will see if the AC results also are confirmed by star collimation and go from there. IIRC when the Tectron AC was first introduced, the goal was to get a darkened field to indicate good collimation (no light escaping) with no mention made of stacking images at all.
|
Vic Menard
Post Laureate
Reged: 07/21/04
Loc: Bradenton, FL
|
Re: autocollimator question
[Re: BDS316]
#5215839 - 05/10/12 04:10 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Quote:
...IIRC when the Tectron AC was first introduced, the goal was to get a darkened field to indicate good collimation (no light escaping) with no mention made of stacking images at all.
Back then, we were making our square center spots from black electrician's tape. The corners of a 5/16th inch square would just protrude from the dark center of the Cheshire ring when the primary mirror alignment was correct. But a black center spot against a dark background was almost impossible to see. So instead of looking for reflections of the dark center spot, we used a bright screwdriver blade positioned between the autocollimator pupil and the center of the secondary mirror to see if we could get its inverted reflection to converge in the center of the autocollimator. This is the premise behind the XLK-C autocollimator, which uses convergence to align two ring targets on the opposing edges of the field of view while observing the reflections of the center spot in the two pupils. The goal was to create a single tool to assess and correct the axial alignments--one tool does all. The ring alignment was much more consistent than trying to converge a moving screwdriver blade, but illuminating the ring target was problematic. And there was an awful lot going on in the two pupils!
With reflective white center spots, it's much easier to see the reflections. And now that the various reflection alignment signatures have been defined, there are more significant goals to achieve after the background has been darkened.
|
Vic Menard
Post Laureate
Reged: 07/21/04
Loc: Bradenton, FL
|
Re: autocollimator question
[Re: Vic Menard]
#5215860 - 05/10/12 04:23 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Quote:
Quote:
...The critical parameter of an AC is that the mirror is perpendicular to the axis of the focuser tube, and a less critical parameter is how well centered the hole in the AC is.
The pupil defines the location of the focuser axis. If the pupil is not precisely centered (and the top edge of the focuser drawtube is precisely perpendicular to the primary mirror axis), the background reflections will only be able to be merged into a single aligned stack when the pupil is centered...
Actually, I was wrong here. Just as an offset/tilted laser beam can be redirected by adjusting the secondary mirror tilt to intersect the center of the primary mirror, and the return beam can be sent back to the laser emitter by adjusting the primary mirror tilt--any defective autocollimator can be made to converge the reflections. But like the offset/tilted laser, the defective autocollimator will not survive rotation verification. This is why it's important that the autocollimator mirror should be both perpendicular to the focuser axis and the pupil should coincide with the focuser axis. In this application (as in most collimation procedures), the focuser axis is the reference axis.
Edited by Vic Menard (05/10/12 04:30 PM)
|
Jason D
Postmaster
Reged: 10/21/06
Loc: California
|
Re: autocollimator question
[Re: Vic Menard]
#5215886 - 05/10/12 04:35 PM Attachment (9 downloads)
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Quote:
This is the premise behind the XLK-C autocollimator, which uses convergence to align two ring targets on the opposing edges of the field of view while observing the reflections of the center spot in the two pupils.
Still use it today 
|
1stScope
super member
Reged: 04/22/07
|
Re: autocollimator question
[Re: Jason D]
#5216034 - 05/10/12 06:17 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
“If the pupil is not precisely centered (and the top edge of the focuser drawtube is precisely perpendicular to the primary mirror axis), the background reflections will only be able to be merged into a single aligned stack when the pupil is centered.”
Considering that one needs perpendicularity in the range of a few ten thousandths of an inch in order to display no change in the reflections as one rotates the AC, the centering tolerance is not nearly as critical. One reason is that the hole that one looks thru is very large in comparison to the perpendicularity tolerance, and another is that one can move the hole off center a fair amount before observing impact on the reflections. Certainly one would like to center the hole and with a lathe it is easy to obtain tolerances similar to the perpendicularity with the right manufacturing process, but the required tolerances vary by a couple of orders of magnitude.
“If you can't see the third reflection, you can't easily see the focuser axial alignment (via the CDP procedure).”
I point the scope to an evenly illuminated wall and have always been able to see four reflections.
|
Vic Menard
Post Laureate
Reged: 07/21/04
Loc: Bradenton, FL
|
Re: autocollimator question
[Re: 1stScope]
#5216113 - 05/10/12 07:12 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Quote:
...Considering that one needs perpendicularity in the range of a few ten thousandths of an inch in order to display no change in the reflections as one rotates the AC, the centering tolerance is not nearly as critical.
If the AC pupil is offset 0.01-inch, and the reflections are collimated precisely to the pupil's axis, rotation will cause the pupil's axis to move 0.02-inch. Although this sounds insignificant, with good primary mirror alignment rotation will cause the two flanking reflections to move 0.04- and 0.08-inches away from the primary mirror center spot. If the error is two hundredths of an inch, the distances double (0.08- and 0.16-inch).
The 5 arc minute error was only causing an "actual" focuser axial error of 0.06-inch (what a precision laser would show relative to the primary mirror center spot)--one quarter of the allowable high performance tolerance.
But we seem to be arguing at cross purposes. You seem to be more concerned with perpendicularity than centering (either of which will impact the autocollimator read), and I'm trying to explain why stacking the reflections with an imperfect autocollimator can still improve the alignment over a simple sight tube when the error is small enough to fall within an acceptable tolerance...
Quote:
...the required tolerances vary by a couple of orders of magnitude.
0.02-inch centering error--reflections separate as much as 0.16-inch.
0.0015-inch tilt error (across one inch mirror)--reflections separate 0.24-inch.
More than an order of magnitude, but not by much. And the tilt error is much easier to maintain with first surface mirrors held against CNC lathed parts (which your posts seem to rely upon), while centering needs to be addressed with an eye to registration, in order for the tool to be inserted into the drawtube (unless a collet type eyepiece lock is used).
Quote:
...I point the scope to an evenly illuminated wall and have always been able to see four reflections.
I just can't get them to cleanly separate in the narrow field of my Tectron AC, but my visual acuity isn't what it used to be either.  At least with a triangular center spot it's easy to see which reflections are rotated 180-degrees!
Have you tried to visualize the focuser axial alignment by carefully decollimating the primary mirror and assessing the alignment of the remaining reflection behind the primary mirror center spot?
|
Jason D
Postmaster
Reged: 10/21/06
Loc: California
|
Re: autocollimator question
[Re: 1stScope]
#5216244 - 05/10/12 08:44 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Quote:
"But when you move it off center, you will no longer see a single stack."
Yup, the stack 'tilts' like that shown in some of the animations on using the AC, where the AC is moved / translated.
What did you mean my "tilt"?
Which animations are you referring to? Can you provide a link?
|
SKYGZR
sage
Reged: 08/14/09
Loc: Planet Earth, Milky Way Galaxy...
|
Re: autocollimator question
[Re: Jason D]
#5216514 - 05/11/12 01:39 AM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
I've been reading all about this collimation issue, and pretty much understand what stuff should look like when "your there." Yet, I have a bit of an issue with where I'm "at".
Everything looks fine, yet when I drop the autocollimator in the focuser (don't tighten the set screw), I can't get the reflected hot spots to show up unless I rock the collimator back and forth. Trying to tighten the compression ring screw doesn't help, it's not visible in the center hole, or the offset.
So, I'm wondering if perhaps the secondary isn't quite perfectly pointing back to the focuser? It looks correct using the sight tube, nice and round, and centered under the focuser.
So the question(s) are, should I drop the AC in and then rotate the secondary till the hot spot reflections are visible, then re tilt using the laser to hit the center of the primary spot, and then see if I can merge them?
Don't know if this tweek has been tried by others, as I haven't read any thing about finalizing the secondary's position using this method.
Other then this little hicup, things "look" really good.
|
Jason D
Postmaster
Reged: 10/21/06
Loc: California
|
Re: autocollimator question
[Re: SKYGZR]
#5216540 - 05/11/12 02:34 AM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Did you watch the following video?
http://www.youtube.com/watch?v=Vj12cx3tnsM
Jason
|
SKYGZR
sage
Reged: 08/14/09
Loc: Planet Earth, Milky Way Galaxy...
|
Re: autocollimator question
[Re: Jason D]
#5217561 - 05/11/12 05:47 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
I did awhile ago, prior to getting the catseye tools. Thanks for reposting. Will revisit the alignment(s) again.
|
1stScope
super member
Reged: 04/22/07
|
Re: autocollimator question
[Re: SKYGZR]
#5231191 - 05/20/12 04:20 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
"What did you mean my "tilt"?
Which animations are you referring to? Can you provide a link?"
All four reflections are not shown…
http://www.cloudynights.com/ubbthreads/showflat.php/Cat/0/Number/3532750/page/0/view/collapsed/sb/5/o/all/fpart/1
#3532769 - 01/01/10 12:29 AM
#3901098 - 07/03/10 05:18 PM
|
Jason D
Postmaster
Reged: 10/21/06
Loc: California
|
Re: autocollimator question
[Re: 1stScope]
#5231303 - 05/20/12 05:44 PM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
Quote:
"What did you mean my "tilt"?
Which animations are you referring to? Can you provide a link?"
All four reflections are not shown…
http://www.cloudynights.com/ubbthreads/showflat.php/Cat/0/Number/3532750/page/0/view/collapsed/sb/5/o/all/fpart/1
#3532769 - 01/01/10 12:29 AM
#3901098 - 07/03/10 05:18 PM
You have referred me to the following two posts:
http://www.cloudynights.com/ubbthreads/showflat.php/Cat/0/Number/3532750/page/0/view/collapsed/sb/5/o/all/fpart/1/vc/1#3532769
http://www.cloudynights.com/ubbthreads/showflat.php/Cat/0/Number/3532750/page/0/view/collapsed/sb/5/o/all/fpart/11#3901098
Both posts are mine and the included animations in both posts are also mine.
None of the animations show clearly what happens when you slide a mirror with the pupil across the focuser opening. Assuming a quality mirror and a perfect focuser then you should see two stacks (reflections P&2 stack versus reflections 1&3 stacks) moving about with respect to each other. If you see 4 reflections unstack then your setup is not accurate enough for an autocollimator.
Jason
|
1stScope
super member
Reged: 04/22/07
|
Re: autocollimator question
[Re: Jason D]
#5231986 - 05/21/12 06:45 AM
|
|
Edit
|
Reply
|
Quote |
Quick Reply
|
|
|
My scope is not available at my current location, so I cannpt verify the difference between two sets of two reflections unstcaking vs 4 reflection unstacking. But, if I'm using a circle for the center spot on the mirror what you seem to be saying is that I should only see two reflections, not four.
The reflections don't unstack when the focuser is rotated, but I can unstack when moving the pupil from center in any direction.
As far as accuracy goes I don't what size pupil you use in your AC, but if it is something like 1/8 in or even 1/16 it will be difficult to claim something like 0.1mm of centering accuracy.
|
|
7 registered and 20 anonymous users are browsing this forum.
Moderator: Jarad, ausastronomer, okieav8r
Print Thread
|
Forum Permissions
You cannot start new topics
You cannot reply to topics
HTML is disabled
UBBCode is enabled
|
Thread views: 1797
|
|
|
|
|
|
|
Previous
Index
Next
Threaded
