Collimation; "confirm with a star test at night". Wat!

Hi! I just got myself a RC10 Truss (TSRC10T64), and while waiting for the sky to clear (it’ll be weeks!) I’m trying to prepare as much as I can. This is my first reflector, so collimation is of course a hot topic :)

Every tutorial I have found ends with “and then of course confirm with a star test at night”, but I haven’t been able to find ANY guide to what that actually means - I get that I’ll see if the stars aren’t perfect, but does anyone know of any tutorial or examples of “if stars look like x, do y”.

As it is, “confirm with a star test at night” is sadly not helpful.. I mean, I get that, for example, if a star is oblong then something is wrong and I should fix it. But there must be some method to this, otherwise we’d all just be randomly testing every single screw during night?

How did you guys and gals learn to do the star test?

For an RC I recommend looking up the DSI method. Basically you inspect the field while defocused. You're looking for the on-axis stars to be evenly illuminated. If they aren't your primary needs adjustment. Then you look at your corners to see if they are even. If they aren't your secondary needs adjustment.

There are a lot of good tools for daytime collimation. My favorite was the Tak collimation scope and only ever had to do small secondary adjustments on the sky occasionally.

Very simple general explanation, it might need additional steps for an RC: first let your reflector cool down to the ambient temperature, next aim on a bright star, then defocus until you see a diffraction pattern (my newtonian reflector does not show a good diffraction pattern when it isn't cooled down enough or even when I move my hand in front of the tube opening) and verify the collimation, adjust where necessary using very small adjustments with the adjustment screws.

This works fine with a camera inserted into the focuser tube but can also be done using a high power eyepiece. It isn't difficult to do, but it takes some time to do it right.

Reg Pratt · Feb 12, 2026, 11:03 PM
DSI method

Wow this is great, thanks! I have much reading to do :)

Normally checking collimation is defocusing the star in-focus and out-focus and there you see if all circles are concentric or not… That is what I do with my Dall Kirkham and all telescopes be it refractors or reflectors show this on a defocused star and tell you it is collimated or not

📷 image.png

A useful software tool that may help (with both the doughnut method mentioned above and the tri-Bahtinov mask method) is SkyCal. From what I read it works just as well on RCs as it does with SCTs and Newtonians.

It saved me much time and hair pulling as I fiddled with the small adjustments required on my 8” SCT. I found using a tri-Bahtinov mask was the most precise, and then confirmed with the doughnut.

You can read about it here (which contains a link to the download page): https://www.cloudynights.com/forums/topic/987005-skycal-focus-and-collimation-software-v40-release/

Cheers.

Paul

Paul Larkin · Feb 13, 2026, 01:59 AM
A useful software tool that may help (with both the doughnut method mentioned above and the tri-Bahtinov mask method) is SkyCal. From what I read it works just as well on RCs as it does with SCTs and Newtonians.
It saved me much time and hair pulling as I fiddled with the small adjustments required on my 8” SCT. I found using a tri-Bahtinov mask was the most precise, and then confirmed with the doughnut.
You can read about it here (which contains a link to the download page): https://www.cloudynights.com/forums/topic/987005-skycal-focus-and-collimation-software-v40-release/
Cheers.
Paul

Thanks - but that’s actually one of many places that have confused me - they state “Depending on your optics, if a marker is outside focus then move the associated adjustment inward. If it is inside focus then move the adjustment outward. The goal is to correct the error while keeping the mechanical balance of the system.”

That’s part of what I just don’t get and I see so many similar phrases in other descriptions, how are you supposed to know which adjustment they mean? For Skycal, I get it that if, e.g. red is wrong then i should turn one of the 6 knobs on either primary or secondary, but how do you know which to turn?

bobtobb · Feb 13, 2026, 07:34 PM
Thanks - but that’s actually one of many places that have confused me - they state “Depending on your optics, if a marker is outside focus then move the associated adjustment inward. If it is inside focus then move the adjustment outward. The goal is to correct the error while keeping the mechanical balance of the system.”

That’s part of what I just don’t get and I see so many similar phrases in other descriptions, how are you supposed to know which adjustment they mean? For Skycal, I get it that if, e.g. red is wrong then i should turn one of the 6 knobs on either primary or secondary, but how do you know which to turn?

I do not have an RC but the trick in collimating this things is that it is an Trial & Error procedure.
You have to move the knobs (or screws Push Pull) and watch how it moves and you need to remember this. If it gets worse go to the other direction.
When a scope is new IMHO mostly the primary mirror is OK and collimation is only done on the secondary.

And Yes, Ritchey-Chretién are not easy to collimate.

Maybe the PDF below is a help

https://www.deepskyinstruments.com/truerc/docs/DSI_Collimation_Procedure_Ver_1.0.pdf

I hope you did not move the primary.

If you have not tested it under the Sky why are you worried about collimation?

In Germany we say “ Do not eat the eggs which have not been laid” 🤣

I agree - I just want to be prepared for once the sky clears! I havent touched anything yet.

You can check at home if your mirrors are severely out of alignment like this:

📷 RC_alignment.jpgSource: https://interferometrie.blogspot.com/2013/01/10-rc-gso.html

Another example:
📷 DSC08315-scaled.jpgSource: https://astrojolo.com/gears/ritchey-chretien-rc8-telescope-first-light/

The concentric pattern should be similiar if you slew to a bright on-axis star and defocus it:

📷 collimating_a_ritchey_chretien_20230724212507b.jpgSource: https://dehilster.info/astronomy/collimating_a_ritchey_chretien.php

The examples show you how your alignment should look like, if you “verify” your alignment (without any corrective optics in between) on a star and it does not look like this, you have to collimate your scope. You can try it on the stars, but when I did it the first time with my RC I made it worse. If the decollimation is severe, I would try the “on table” collimation at home first and then do the last 5-10% on the stars (mainly primary adjustment and focuser tilt)

Regards
Tobi

Tobiasz · Feb 14, 2026, 09:02 AM
You can check at home if your mirrors are severely out of alignment like this:

📷 RC_alignment.jpgSource: https://interferometrie.blogspot.com/2013/01/10-rc-gso.html
Another example:
📷 DSC08315-scaled.jpgSource: https://astrojolo.com/gears/ritchey-chretien-rc8-telescope-first-light/
The concentric pattern should be similiar if you slew to a bright on-axis star and defocus it:
📷 collimating_a_ritchey_chretien_20230724212507b.jpgSource: https://dehilster.info/astronomy/collimating_a_ritchey_chretien.php
The examples show you how your alignment should look like, if you “verify” your alignment (without any corrective optics in between) on a star and it does not look like this, you have to collimate your scope. You can try it on the stars, but when I did it the first time with my RC I made it worse. If the decollimation is severe, I would try the “on table” collimation at home first and then do the last 5-10% on the stars (mainly primary adjustment and focuser tilt)

Regards
Tobi

Thank you, this is super useful for me. Much appreciated!

After going through the whole RC collimation adventure a few months ago I’m happy to pass on what I learned.

The first thing is don’t over-think it. A lot of guides, especially the ones that require special tools or software can really lead you up the primrose path and down deep into the rabbit hole.

There are also a lot of guides based on getting everything lined up visually by looking up from the focus position. This works great if the scope is machanically perfect and the optical centers of both the primary and secondary are perfectly centered, two big if’s.

There are also guides that make a big deal about aligning the focuser separately from the primary. I would view that as a refinement, if needed.

Use the DSI method Pratt suggested. use the primary adjustments to get the donut as round and evenly illuminated as possible. Note: it must be in the center of the field. Each adjustments will move it off center so bring it back before the next adjustment. This is critical!

Once done Then move on to the secondary. If the little dot in the center of the donut is off center, use the secondary to center it. Once you have all that you’ll have a useable instrument.

The last refinement is to take a defocused starfield image. look at the corners, all the donuts should be symmetrically balanced. If they are all round great. If they are slightly oval then the pointy ends or the flat sides should all point towards the center of the field. Again, symmetrically balanced. If not you need to adjust the secondary until you get a balanced pattern.

RC are harder to collimate then Newtonians but once you get it, it should be very stable over time.

Have fun!

I second/third the DSI method. I bought an RC8 mid last year and wanted the best collimated scope possible. I bought an OCAL with it and it was never quite right with the stars. Turns out the donut on the secondary mirror was off by about 5mm - a known problem with GSO scopes discussed on cloudy nights which is why I checked it. Once adjusted for this it did a pretty good job. I also used the Skywave software with which many have great success. I found it confusing as well but the biggest problem was producing the nine star circle for analysis and finding you’d adjusted the wrong way. It’s an accurate but very tedious and time consuming process .

The DSI method is logically explained and gives instant feedback to any adjustments made.

In the end, near enough was good enough as I feel there are other factors with greater impact, some of which are out of ones control such as seeing.

Thanks all, I really appreciate the help! I now feel pretty well armed with good information. Looking forward to first light!

Hi all! Quick update and a question :)

I used a simple collimator to get things in line (just a tiny change to the front mirror was needed (and only changed 2 screws to not mess with distance to main mirror).

Got the following star test tonight: 📷 image.pngWhich I guess looks ok for now. But my problem is that I can’t quite get things in focus. Hocus Focus fails spectacularly, but going manual I can definately make out where the focus point should be. But even that is too blurry for Hocus Focus which is why it completely fails.

Why could this be? The best focus point I can achieve is still far too blurry: 📷 image.png

I think my backfocus is accurate, so I’ really at a loss for what could be wrong. Any tips appreciated!

EDIT - I noticed I’m unable to achieve pin point focus on my OAG too, with the nearest focus with it all the way at the bottom, so I probably do have too large backfocus! More fun!

The out of focus images are pretty rough because of seeing but it looks like you might still be off a bit, the lower left quadrent looks a little brighter. Take a bunch of images and stacking would allow you to see what’s going on a lot better.

Your image from NINA really doesn’t look that bad and your HFR is down around 2 which is great. The diffraction spikes look sharp so if you’re not in focus you’re very close. I wonder if the problem might not be your auto-focusing setup?

Your focus curve looks way too flat. There is not enough difference in your focus points to run the routine. That’s why it fails. You need a steeper curve. To do that, you need a larger step size in your focus routine.

Aaron Hakala · Mar 6, 2026, 05:41 AM
Your focus curve looks way too flat. There is not enough difference in your focus points to run the routine. That’s why it fails. You need a steeper curve. To do that, you need a larger step size in your focus routine.

Yeah! But that is the point, I can manually deduce where the curve is when by manually adjusting focus. Hocus Focus fails to properlyi identify the stars which is not the root cause - even when manually finding “focus” it is far too blurry. I’ll do some experimentation with backfocus tonight and see if I can figure this one out.

So this is my collimation (didn’t change anything, just found a better star ;D ) - looks pretty good doesn’t it?

Still can’t get things in focus though :(

This is without any correctors, just straight into the camera.

📷 RC10-collimation.gif

Much better. It looks like it’s shifted just lightly to the 11 o’clock position but it’s certainly close enough to get focus.

…..and to follow up with your focusing. Looking at your curve I’d guess that the initial offset is to small and the steps are two small. Increase the offset a lot and the step size slightly and try again.

I FIGURED IT OUT!

Thanks to all for your help. The problem was somewhere else entirely!

I got a iOption EAF with the RC10, which supposedly has zero backlash. However, I had to increase backlash compensation to 500 (!), and now everything is working as expected! No wonder my autofocus routines were messed up!

📷 {C08FE1F7-0876-460C-9E58-2CA09AF760D5}.png

That would do it….

Related discussions