Disappointing Collimation with Catseye Infinity XLKP

[Martin Meredith]

Thanks also from me Jason. I did try the 180 degree (I assume) rotation test and sure enough the reflections moved a fair bit! My approach is to get it as good as I can but without letting it eat too much into the observing session. Certainly I appreciate the hotspot design in making it much easier (and dare I say pleasurable) to collimate fast Newts.... Anyway, I spent most of the night looking at faint open clusters and the stars seemed pretty pinpoint to me (I'll post some later in the video forum).

Martin

[richyrich_one]

On 27/02/2017 at 10:00, Jason D said:

By examining your photos, I can visualize the position of the autocollimator with respect to the primary mirror in 3D. See attachments. Left photo from the center pupil and right photo is from the offset pupil.

Please try to the following slightly modified steps:

1- Stack P1 and P2 via the offset pupil by adjusting the secondary mirror. You stated that this step was hard due to the larger P2 size. Do your best. Precise stacking is not needed. If stacking is difficult, at least make sure P2 is shifted little outward -- as opposed to inward. In your photo, P2 was shifted inward. Ignore the remaining reflections when executing this step.

2- The modified step: Align the AC center pupil reflection with P1 by adjusting the primary mirror. Typically, this step is difficult when the AC is close to the focal plane since the background will get darkened  which makes it hard to discern the pupil reflection. However, when the AC is substantially below the focal plane, the background will never be fully darkened and I presume it would be possible to discern the pupil.  Ignore all other reflections. Just focus on P1 and the pupil.

3- Re-iterate between steps 1 and 2

4- This is an optional step that might give you that last little precision. Stack P2 and P1 by adjusting the secondary mirror as seen from the central pupil (step 1 above was from the offset pupil). If discerning P2 is difficult in the clutter, ignore this step. 

Again, I want to remind you that the sensitivity of the AC increases when it is located below the focal plane. After completing the above steps, you might notice that some reflections might not stack perfectly. That is OK. Any stacking errors are highly magnified due to the extra sensitivity. 

Jason

 

 

Jason

I have managed to follow your instructions I think successfully.

P+2 stacked via the offset pupil using secondary adjustment, central pupil moved to centre of P using primary adjustment and reiterated until no further adjustment required. Primary confirmed with telecat.

Trying to stack P+2 via the central pupil is just too confusing though so I missed that step.

What I am struggling to understand is, if the above procedure aligns everything pretty much perfectly, then when I use the telecat to view the concentricity of the primary reflection in the secondary they are not concentric. I would expect them to be. What am I doing wrong?

Unfortunately I'm struggling to get pictures.

I'm being particular about this as I am having focal plane tilt issues and was hoping this was going to cure it.

Any further insight would be very helpful and much appreciated.

 

[Jason D]

11 hours ago, richyrich_one said:

What I am struggling to understand is, if the above procedure aligns everything pretty much perfectly, then when I use the telecat to view the concentricity of the primary reflection in the secondary they are not concentric. I would expect them to be. What am I doing wrong?

I will try to explain though I am afraid I will run the risk of confusing you. Let me try.

The autocollimator performs what is referred to as "axial alignment". After alignment is complete, the primary mirror axis should reflect off the secondary mirror and coincides with the focuser axis. Likewise, the focuser axis will reflect off the secondary mirror and strikes the primary center. In other words, both the focuser and primary axes will be coincident, hence, the term "axial alignment". The two goals for achieving axial alignment is to minimize coma and to eliminate focal plane tilt.

Aligning the secondary mirror edge against the primary mirror reflection is a completely different and independent alignment. The main goal of this alignment is to optimize field illumination. That is, the image will be brightest at the FOV center. Telecat is the right tool to complete this alignment. You need to complete it before you start axial alignment using the autocollimator. Rotate the secondary mirror and use the center bolt if needed to center/round the secondary mirror under the focuser. 

Back to the autocollimator discussion when the it is positioned below the focal plane. I dusted off my old light-ray simulator that I developed to study the autocollimator back in 2009. I ran few simulations at different locations below the focal plane. See first four attachments. Check how reflection 2 starts at the same distance as reflection P when the autocollimator is located at the focal plane. As the autocollimator is lowered, not only reflection 2 will rise above the primary mirror surface but the angle as shown by the green arrows will increase. The focal tilt error is the same for all simulations. The angle between the green arrows relates to the sensitivity of the tilt error. As the autocollimator is lowered, the error sensitivity will increase. Stacking reflections P & 2 as you did via the offset pupil should ensurethe virtual elimination of the focal plane tilt error.

I also tried to run light-ray simulation based on the numbers you have provided (650mm FL with the autocollimator mirror located ~30mm below focal plane). The results (the last and 5th attachment) did not match the information I was able to extract from your photos. Are you sure your 30mm estimate is correct? According to my ray simulation, I am guessing ~55mm is a more accurate number. That is the distance between the autocollimator mirror and your camera sensor. Do not forget that the autocollimator mirror surface is located almost 1/2 way across the height of the autocollimator.

Jason

 

 

 

 

 

[richyrich_one]

21 minutes ago, Jason D said:

I will try to explain though I am afraid I will run the risk of confusing you. Let me try.

The autocollimator performs what is referred to as "axial alignment". After alignment is complete, the primary mirror axis should reflect off the secondary mirror and coincides with the focuser axis. Likewise, the focuser axis will reflect off the secondary mirror and strikes the primary center. In other words, both the focuser and primary axes will be coincident, hence, the term "axial alignment". The two goals for achieving axial alignment is to minimize coma and to eliminate focal plane tilt.

Aligning the secondary mirror edge against the primary mirror reflection is a completely different and independent alignment. The main goal of this alignment is to optimize field illumination. That is, the image will be brightest at the FOV center. Telecat is the right tool to complete this alignment. You need to complete it before you start axial alignment using the autocollimator. Rotate the secondary mirror and use the center bolt if needed to center/round the secondary mirror under the focuser. 

Back to the autocollimator discussion when the it is positioned below the focal plane. I dusted off my old light-ray simulator that I developed to study the autocollimator back in 2009. I ran few simulations at different locations below the focal plane. See first four attachments. Check how reflection 2 starts at the same distance as reflection P when the autocollimator is located at the focal plane. As the autocollimator is lowered, not only reflection 2 will rise above the primary mirror surface but the angle as shown by the green arrows will increase. The focal tilt error is the same for all simulations. The angle between the green arrows relates to the sensitivity of the tilt error. As the autocollimator is lowered, the error sensitivity will increase. Stacking reflections P & 2 as you did via the offset pupil should ensurethe virtual elimination of the focal plane tilt error.

I also tried to run light-ray simulation based on the numbers you have provided (650mm FL with the autocollimator mirror located ~30mm below focal plane). The results (the last and 5th attachment) did not match the information I was able to extract from your photos. Are you sure your 30mm estimate is correct? According to my ray simulation, I am guessing ~55mm is a more accurate number. That is the distance between the autocollimator mirror and your camera sensor. Do not forget that the autocollimator mirror surface is located almost 1/2 way across the height of the autocollimator.

Jason

 

Jason

Thanks for clarifying the difference between axial alignment and mirror concentricity.

As far as focal plane tilt is concerned, it's good to know I have the right tool for the job.

I'm not sure I entirely understand the diagrams but your explaination makes sense. Bottom line, if using the AC below the focal plane and P+2 are stacked then the secodary alignment and therefore the focal plane tilt correction is MORE ACCURATE although it might not be as easy to do so. If the mirrors are not concentric when done then it's highlighting that the secondary was not properly positioned in the first place.

I'll double check the AC placement in realtion to the focal plane.

[Jason D]

 

Over the past 1 month (since the last post in this thread), I have been spending time (on and off) running ray simulation to have better understanding of the autocollimator performance when it is located far away from the focal plane. I have new information to share:

1- The XLKP autocollimator performs its best when the AC mirror is within 1.25% FL (Focal Length) from the focal plane. For a reflector with a 1000mm FL, that translates to ~+/-12.5mm or ~+/-0.5". As long as reflection 2 "appears" to have the same size as reflection P then you must be within that 1.25% FL window. Within this range, reflection P&2 parallax will be hard to detect. Outside this range, the relative size and parallax will gradually become more prominent.  

2- The requirement described above is quite generous for "visual" reflectors. Any visual reflector that can't meet the above requirement will most likely not be able to bring some if not all the eyepieces to focus. 

3- "Imaging" reflector will most likely not meet the above requirement. To be fair, the focal plane requirement is mentioned in the XLKP procedure from Catseye. Below are your options:

A) If you are considering the XLKP autocollimator for an imaging scope and you have not bought one, consider contacting your local vendor (for this forum it is FLO) or Catseye to inquire about a special order XLKP AC with an extended barrel. I am not affiliated with both businesses and I do not know if your request can be accommodated but I am aware of few requests that were accommodated by Catseye in the past as special orders for an additional cost.

B ) If you already own an XLKP autocollimator (the production/common version without an extended barrel), consider purchasing (if you do not already own one) a quality 2" extension to elevate the XLKP AC to bring the AC mirror within 1.25% of the focal plane. Use relative size and parallax between reflections P&2 as your guide to know if the AC mirror is within the desired range. Do your math before purchasing to ensure the extension has about the right height. I want to emphasize the word "QUALITY" extension. Do NOT use or purchase a poor quality 2" extension. 

C) In case you do not have a quality 2" extension and do not want to spend the money to purchase one, I recommend the following modified step: Do not stack reflections P&2 from the offset pupil. Instead, align their inner edges. Below are detailed steps that are only applicable to the Hotspot (refer to the first attachment to clarify the steps):

i) Rotate the AC until one of reflection P Hotspot edges is aligned with the central pupil. No need to be precise - close enough is good enough.

ii) Adjust the secondary mirror to align the edge of reflection 2 with the edge of reflection P.

iii) Use Catseye Blackcat XL to align the Hotspot reflection with the Blackcat XL ring reflection by adjusting the primary mirror. If you do not have the Blackcat XL, align reflection P with the AC central pupil reflection which should be visible when the AC mirror is substantially below the focal plane.

iv) Re-iteration between steps ii and iii. NOTE: Always re-insert the XLKP AC in the focuser with the same orientation. Use the position of the offset pupil as your guide.

As I have repeatedly stated in this thread, once the AC mirror is below the focal plane, its sensitivity increases. The difference between stacking P&2 versus aligning the inner edges in terms of error is small; however, inner edge alignment will decrease the already small error. If someone were to ask me whether the inner-edge-alignment procedure without an extension tube is as good as the stacking procedure with extension, all I can say is that the difference between them is very small "theoretically"; HOWEVER, I am not an imager expert and I do NOT know if the small error difference will be visible at the sensor. Someone with an imaging expertise will need to try both procedures (stacking with extension) versus (edge-alignment without extension) and share their expert and honest opinion with other imagers.

The second attachment includes a table showing reflection 2 size change with respect to reflection P as the AC mirror moves away from the focal plane. When the AC moves above the focal plane, reflection 2 will decrease in relative size. When it moves below the focal plane, reflection 2 will increase in relative size. For example, based on this table I have estimated Rich's AC mirror to be located ~42mm below the focal plane. The ~42mm estimates is based on the scopes's focal length of 650mm (provided by Rich), reflection 2 being ~17% larger than reflection P (based on the photo provided by Rich), and the second attachment table that suggests when the AC mirror is at ~6.5% FL below the focal plane reflection 2 will be about 17% larger than reflection P.

The third and last attachment gives an idea about the relative size difference and parallax for different AC mirror locations with respect to the focal plane. Note how the size difference and parallax is hard to detect with the AC is at +-1.25% FL from the focal plane.

One last word: If you have been satisfied with the way you have been using your XLKP for imaging reflectors, my all means continue to use your same procedure. If you would like to follow some of the suggestions described in this post, do so and see if you can see any difference in the final images. If you see a positive difference then switch to the updated procedure. If you don't then it is up to you whether to keep and original (stacking) procedure or switch to the (edge-alignment) modified procedure.

Jason

 

[richyrich_one]

On 28/03/2017 at 08:42, Jason D said:

Over the past 1 month (since the last post in this thread), I have been spending time (on and off) running ray simulation to have better understanding of the autocollimator performance when it is located far away from the focal plane. I have new information to share:

Thanks Jason, you've gone the extra mile again!

[richyrich_one]

I've gone the 2" extension route with this now and it seems to be giving the most repeatable results. It's a 30mm extension and with that, coupled with moving my mirror up the tube by about 15mm to lose the focuser intrusion, using the above comparison images puts me at about 2.5% below the focal plane.

At this position I can almost completely stack all the reflections via the central pupil. There is still some very minor ghosting, almost non-existant. The problem is when I do this the telecat does not agree showing a slight error, no matter how many iterations I go through. I have to consistantly move the primary exactly the same amount and same direction to get the hoptspot central with the telecat.

Which should i trust? At 2.5% below the focal plane are the central pupil reflections not accurate enough?

 

[Jason D]

Tell me more about your iteration steps? Do you adjust the secondary mirror to stack P&2 via offset pupil followed by adjusting the primary mirror to align the HotSpot with the Telecat ring -- back and forth?

Jason

 

[Jason D]

First attachment explains the concept of axial alignment 

Second attachment explains how the P&2 stack via the offset pupil of the XLKP is only sensitive to the tilt between the AC mirror plane and focal plane

Third attachment explains how the HotSpot alignment with the Telecat or Blackcat is only sensitive of the displacement between the Telecat/Blackcat center and the focal point

Both alignments are "independent" and should be able to achieve

[richyrich_one]

Thanks Jason they are really useful animations.

My steps:

1) Stack P+2 via offset by adjusting secondary

2) Stack all via central by adjusting primary

Repeat 1 and 2 until they agree.

3) Check telecat and adjust primary

and repeat.

I would be expecting a refinement of the adjustment needed each time but this doesn't appear to be whats happening. After steps 1 and 2 the primary is out again according to the telecat by the same amount and the same direction.

 

 

[Jason D]

Can you re-iterate between P&2 via offset (secondary) and Hotspot&ring via Telecat (primary)?

when done, check the central pupil of the XLKP

It is late here in California -- almost 1:00AM... I will continue this exchange in the morning.

Jason

[richyrich_one]

Just now, Jason D said:

Can you re-iterate between P&2 via offset (secondary) and Hotspot&ring via Telecat (primary)?

when done, check the central pupil of the XLKP

It is late here in California -- almost 1:00AM... I will continue this exchange in the morning.

Jason

No worries

[richyrich_one]

On 21/04/2017 at 08:49, Jason D said:

Can you re-iterate between P&2 via offset (secondary) and Hotspot&ring via Telecat (primary)?

when done, check the central pupil of the XLKP

Hi Jason

Sorry it's been a while.

I did as you asked and via the central pupil the hotspot is not stacked. When I adjust the primary to refine the collimation and stack the reflections (pretty good but some very minor ghosting) rechecking with the telecat shows a very slight error.

The only difference is the method of mounting the telecat and the XLKP. I have to use a self-centering extension for the XLKP but the telecat fits snugly in the focuser unaided. I tighten 1 thumbscrew on both so as to best simulate the mounting of my camera. Unfortunately this time of year testing opportunities are rare, very late nights are rarely convenient.

So I was at least hoping to get everything agreeing nicely on the bench before testing.

As a side note, collimating my 250P-DS, with the same extension, seemed much easier and everything agrees although everything is a lot smaller. A pretty much textbook process. I am presuming the larger focal length(1200mm) is having a much smaller effect on the correct position of the XLKP as the percentage of error will be smaller. I hope that makes sense.

I need to test. If at the end of the day my images are free of tilt I don't care whether they agree.

[Jason D]

On 4/27/2017 at 01:11, richyrich_one said:

s a side note, collimating my 250P-DS, with the same extension, seemed much easier and everything agrees although everything is a lot smaller. A pretty much textbook process. I am presuming the larger focal length(1200mm) is having a much smaller effect on the correct position of the XLKP as the percentage of error will be smaller. I hope that makes sense.

That makes sense. XLKP is easier to deal with for telescope with longer focal lengths.

I am glad to hear that the results are looking better -- though not perfectly -- for your other scope with shorter focal length.

Report back once you get a chance to use the scope for astrophotography

Jason

 

[richyrich_one]

Using a threaded means of fitting the XLKP I have now got it to agree with the telecat.

Now for a clear sky...

[Jason D]

[Arc minute]

Hi Rich and Jason! This has been a fantastic Post, a true learning journey! I have been thinking on purchasing catseye system and stumbled to this Post by accident. I have been using only precision  cheshire up until this point.

It would be great to hear from Rich what was the end result when imaging?! 

I have f4 8" imaging rig so this discussion has been highly relevant for me. I also sent an email to Jim of Catseye If he already has some remedies brewing.

 

Thanks guys!!

[richyrich_one]

Hi Arc Minute

I think I have corrected the tilt. I have some coma to deal with but it looks pretty even.

You probably gathered the biggest issue for me was repeatability. Using a threaded means of attaching the self-centering extension /catseye assembly and then using the same threaded connection to attach the camera has made all the difference. 

[Jason D]

On 9/3/2017 at 00:45, Arc minute said:

I also sent an email to Jim of Catseye If he already has some remedies brewing

Hello,

I am not sure if you heard back from Jim yet -- it was a national Holiday here in the USA yesterday (Labor Day). A while back Jim stated that he is planning to offer an autocollimator with an extended barrel specifically for imagers . 

Check out this page from Catseye official website:

http://www.catseyecollimation.com/autocol.html#xlkfeatures

Refer to the info under the section titled "ATTENTION IMAGERS WITH ASTROGRAPHS"

Jason

 

[Arc minute]

19 hours ago, Jason D said:

Hello,

I am not sure if you heard back from Jim yet -- it was a national Holiday here in the USA yesterday (Labor Day). A while back Jim stated that he is planning to offer an autocollimator with an extended barrel specifically for imagers .

Jason

 

Hi

Jim kindly replied to me and we exchanged few emails for dimension measurements.

I measured that my "native" focal point is about 52mm outwards from the focuser lip when it is fully extended. Jim messaged me that the AC would require 2,5" longer tube than normal and he would have those available shortly. He already had 1" extended versions but that wasn't enough in my case. 

I'll have to manage with my previous means in the meanwhile.  The imaging season is just starting here after summer's nightless nights. I am eager to get all the updates for my rig ready and operational for the autumn sessions :).

[Jason D]

Yesterday I found out that Catseye is now officially offering new versions of the XLKP autocollimator with longer barrels for astro-imagers. I could not help but to recall this interesting thread which is quite relevant to the new offering.

All the details are on Catseye website https://www.catseyecollimation.com/

Jason

[richyrich_one]

You forgot to mention the free "evaluation" stock available for those involved in this thread. 

[25585]

On 25 February 2017 at 22:51, richyrich_one said:

I've not had a chance to get out and get that info, should be able to tomorrow.

I've had a thought, it shouldn't be too difficult to make up an inexpensive adaptor from 2 pieces of alluminium tube from here:

https://www.forwardmetals.co.uk/

A piece 50.8mm OD with wall of 4.8mm slid inside another piece 63.5mm OD with wall of 6.4mm making an ID of 50.7mm, might have to cool one and heat the other to fit them?

I've used them before, they do 2 free cuts so just need to calculate a rough length to bring the AC within the focal plane. They should cut them nice and square. At £13 delivered it might be worth a try. They would be a much better fit than the usual extension tubes and pretty substantial.

Oh I wish I had a lathe!

 

Would the Howie Glatter parallizer,  an exact centering tube, be what you need? FLO sell those & other HG kit. 

[Jason D]

On 4/17/2018 at 07:43, richyrich_one said:

You forgot to mention the free "evaluation" stock available for those involved in this thread. 

Just saw your response -- two weeks late

Hmmm, I am involved in this thread and I did not get any free "evaluation" stock       Actually, since I do not own an imaging reflector, the new product is not for me.

Rich, did you get one of the new XLKP products? If yes, which one did you get (barrel size)? How well did it work for you? I am very much interested in your feedback -- assuming you got one.

Jason

 

[richyrich_one]

Hi Jason

As far as I can see they aren't available in the UK. I'm sure FLO could get one if I asked them. I just didn't want to shell out for a new improved version. They are quality but they aren't cheap either. Wish I'd sent it back for a refund when I had the chance, then I'd have been in a better position. 

I'm curious... What made them make an extended version for imaging scopes in the first place?