Interesting collimation technique

[BrendanC]

Hi all,

I recently watched this interesting video about collimation: https://www.youtube.com/watch?v=8ROvNH5uwDo

I really liked the way it used the overlay to achieve a perfectly round secondary mirror. This has always been a difficulty for me. I just could not get the hang of looking through the collimation cap while twiddling the tilt screws.

However, I don't have the webcam used in that video, nor would I want to modify one. So, I then realised I do actually have a camera: in my mobile phone!

So I've done something similar to that video. You need a laptop or PC, a collimation cap, a mounting bracket for your phone, TeamViewer (the free version is fine), and the image overlay utility called Mire de Collimation which you can get here: http://sweiller.free.fr/collimation.html.

You also need a means of blanking out the primary, and some coloured card to put behind the secondary (which is quite a common technique for helping discern the secondary).

First, blank out the primary (I use some card) and put the coloured card behind the secondary, as explained in that video.

Then put the collimation cap on the eyepiece, and position your mobile phone over it that so you can see the secondary clearly on the phone's camera. Expand the secondary image as large as it will go on the phone's screen. Then log in remotely from a laptop or PC to that camera using TeamViewer, and then run Mire de Collimation and position the overlay circle over that. Now you can play around with the secondary tilt screws and central screw to your heart's content without peering into the cap, guessing at whether it's a circle or central or not, and (if you're anything like me) getting more and more frustrated with the whole process.

I was able to get pretty much a perfect circle for the secondary, and whereas it didn't work quite so well when looking for the primary clips because the camera's auto focus kept changing, I'm sure I could find a way around that (the FV-5 app looks promising in that regard). In the event I just did that by looking through the collimation cap in the 'traditional' way! I also tried putting a pinhole in some tape over the collimation cap to REALLY get the camera central, but it was very difficult to get a good picture, but there might be ways to do that too.

Anyway, this method worked for me. It was so very much easier doing this while looking at the laptop screen (I also got some replacement tilt thumbscrews which really helped too: https://www.365astronomy.com/set-of-m4-thumbscrews-for-secondary-mirror-collimation-45mm-long-set-of-3-screws.html). A star test later showed me perfectly concentric circles as a result.

If it helps, here are some photos showing how I did this.

Phone secured in bracket against collimation cap (you need to be careful and make sure the phone is as parallel to the cap and tube as possible):

OTA with phone in place showing photo of secondary with coloured card behind it:

TeamViewer screenshot showing remote session with Mire de Collimation circle overlay:

I'd be very interested to know if anyone uses any similar technique, or whether anyone tries this method out and with what degree of success.

Cheers, Brendan

Edited August 9, 2020 by BrendanC

[andrew s]

Nice work Brendan the 21 C sight tube. Regards Andrew  

[johninderby]

Basicly the same idea as the Concentre collimator but using a smartphone instead of a dedicated collimator that has an etched screen with circles on it. 

Edited August 9, 2020 by johninderby

[CraigT82]

Its a good technique, I've done the same in the past using a planetary camera with a meteor lens. Haven't seen it done with a phone though, didn't know about linking phone screen to team viewer so will look into that... nice one 👍

[markse68]

Why is it so important to get it perfectly circular? I’ve often wondered this. What if your secondary isn’t a perfect ellipse? If its “in the middle” then when you tilt it to get the whole primary in view you’re going to get a more sensitive and accurate 45degrees aren’t you? Not knocking the technique btw- it looks excellent and useful- just wondering what’s so magical about it being perfectly circular 🤷‍♂️

[johninderby]

If you get the secondary perfectly positioned then the rest of the collimation is easy. Just level the primary mirror by measuring the distance between the mirror and the cell at the collimating screw points and the primary is virtualy collimated without actually collimating the primary and the primary will only need a minor tweek. 

[andrew s]

20 minutes ago, markse68 said:

Why is it so important to get it perfectly circular? I’ve often wondered this. What if your secondary isn’t a perfect ellipse? If its “in the middle” then when you tilt it to get the whole primary in view you’re going to get a more sensitive and accurate 45degrees aren’t you? Not knocking the technique btw- it looks excellent and useful- just wondering what’s so magical about it being perfectly circular 🤷‍♂️

If you don't center the secondary then you won't get even illumination even if the rest is collimated  perfectly. The secondary doesn't have a unique optic axis it's the only way to place it correctly under the focuser. 

Regards Andrew 

Edited August 9, 2020 by andrew s

[BrendanC]

2 hours ago, andrew s said:

Nice work Brendan the 21 C sight tube. Regards Andrew  

Thanks! Although I don't know what you mean about the 21C but I'll just go with it.

2 hours ago, CraigT82 said:

Its a good technique, I've done the same in the past using a planetary camera with a meteor lens. Haven't seen it done with a phone though, didn't know about linking phone screen to team viewer so will look into that... nice one 👍

Yes, it was the sudden realisation that the phone screen could then link to the laptop screen, and Mire de Collimation looks after getting the circle right, which made this whole thing viable.

2 hours ago, johninderby said:

Basicly the same idea as the Concentre collimator but using a smartphone instead of a dedicated collimator that has an etched screen with circles on it. 

I have considered getting the Concentre for exactly that reason - the etched circles - but figured spending £10 on the bracket was less expensive! Also, I'm hoping I can get my camera app to focus better on the primary clips in future, so this method should eventually help with that too. Also, you're still kind of bent over using the Concentre, and believe me, using the laptop next to the scope was so very much less bothersome. 

[andrew s]

21st  century  Regards Andrew 

[BrendanC]

Ah, not 21 degrees centigrade then!

[Peter Drew]

It would never be any use if it had to be at 21c here.     🙂

[Craig a]

It’s a great idea this, I’ve downloaded an app on iPhone and corresponding app on my laptop that turns iPhone into webcam using usb from iPhone to lappy no WiFi connection needed I’m gonna give this ago once my telescope phone mount arrives 

[Avocette]

I’ve been using this technique recently. I tried an old webcam with poor results, mostly failing to physically lock the unit into an old eyepiece tube or similar 1.25” adapter. However I subsequently found that I could combine the 6mm CS-mount lens sold for use with the Raspberry Pi HQ camera, with my ASI120 guide camera. These cameras come packed with a CS-mount 150° ‘all-sky’ lens in an adapter plate that screws into the camera. With this in place this prevents using the 1.25” eyepiece focus tube sleeve, but the camera body already features a 2” focus tube sleeve. The 6mm lens fits in place of the 150° lens and has typical photographic aperture and focus adjustments.

This is the resulting image from SharpCap using the built in collimation circles when I had completed adjustments with the focus set to the centre ring on the primary.

One thing I learned in this process was that the ASI120 sensor chip is not perfectly centred in the 2” mount flange. I adjusted for these offsets using SharpCap’s Region of Interest settings.

[Craig a]

I’ve just had ago at doing this the phones focus wasn’t as sharp as I would of liked but I managed, just need a clear night now to check the stars in the corners 🤞

Edited August 17, 2020 by Craig a

[Captain Scarlet]

On 09/08/2020 at 20:51, markse68 said:

Why is it so important to get it perfectly circular? I’ve often wondered this. What if your secondary isn’t a perfect ellipse? If its “in the middle” then when you tilt it to get the whole primary in view you’re going to get a more sensitive and accurate 45degrees aren’t you? Not knocking the technique btw- it looks excellent and useful- just wondering what’s so magical about it being perfectly circular 🤷‍♂️

If the secondary’s edge, from your eye’s place at the focal point, sits on the edge of a cone (centered on the focal axis) emanating from your eye, then and only then should it appear perfectly circular. Assuming it is a “root-two” ellipse and the “bounce” is 45degs.

Subsequently, if the edge of the primary is concentric to that, you’ve “coincided” the primary cone with your “eye-emanating” cone.

That circularity and concentricity are beautifully convenient optical tricks making Newtonian collimation so much easier.

Edited August 19, 2020 by Captain Magenta

[markse68]

10 minutes ago, Captain Magenta said:

If the secondary’s edge, from your eye’s place at the focal point, sits on the edge of a cone (centered on the focal axis) emanating from your eye, then and only then should it appear perfectly circular. Assuming it is a “root-two” ellipse and the “bounce” is 45degs.

Subsequently, if the edge of the primary is concentric to that, you’ve “coincided” the primary cone with your “eye-emanating” cone.

That circularity and concentricity are beautifully convenient optical tricks making Newtonian collimation so much easier.

I think I understand the geometry it just seems odd to me to stress getting the outline of the secondary perfectly circular when that seems a very imprecise thing to try to attain? What's important surely is that the secondary is positioned so that its opposite edges are symmetrically spaced within the focuser tube circular outline- that is that the minor axis is central and the major axis too, getting it roughly circular, then the precise alignment happens when you tilt it around the minor axis to see the primary evenly spaced within its outline. Isn't that far more sensitive than trying to eyeball the perfect roundness of the secondary? Getting the secondary roughly circular and therefore roughly to 45deg helps to judge the centrality of the secondary though, barring offsets

[johninderby]

Using a Concentre is more precise as it allows for perfectly centering the secondary under the focuser and making sure it appears a perfect circle as well. If the secondary is perfectly positioned then simply adjusting the primary so it is dead level with the mirror support will mean the scope is very close to being collimated and only a minor tweek is needed to reach perfect collimation.

The seconday is not supposed to be physically centered in the tube but rather optically aligned with the primary mirror. 

[AstroNebulee]

Looks a great idea 👍

[Captain Scarlet]

9 hours ago, markse68 said:

I think I understand the geometry it just seems odd to me to stress getting the outline of the secondary perfectly circular when that seems a very imprecise thing to try to attain? What's important surely is that the secondary is positioned so that its opposite edges are symmetrically spaced within the focuser tube circular outline- that is that the minor axis is central and the major axis too, getting it roughly circular, then the precise alignment happens when you tilt it around the minor axis to see the primary evenly spaced within its outline. Isn't that far more sensitive than trying to eyeball the perfect roundness of the secondary? Getting the secondary roughly circular and therefore roughly to 45deg helps to judge the centrality of the secondary though, barring offsets

The circularity is about secondary positioning: if perfectly circular (from the focal point), and if the offset has been set correctly, that circularity is pretty much the only way you can be sure the secondary is in the right place to intercept the primary's light cone optimally, before you move on to the axial alignment. As you say, axial alignment is different: the axes can still be (in extreme case) aligned even close to the edge of the secondary, but you'd then only be intercepting less than half the primary's light-cone. And as you say if you can see the primary's edges concentrically then you're there anyway: the initial circularity helps get you there much more quickly and allows you to "get there" without further having to touch the central screw, only the tilt adjusters.

[BrendanC]

Hi,

I just used this technique again, and realised that @markse68 has a very valid point. Previously, I'd concentrated on getting the secondary perfectly round, as per the instructions on the classic Astrobaby guide.

However, after some issues, I decided to take another look, and realised that the secondary wasn't aligned in relation to the focuser tube.

Thing is, this technique helps with that too: if you pinch/zoom out from your mobile device, you can get both the secondary mirror and the dark edge of the focuser in the image. Then, you can use the Mire de Collimation app to get two circles, one mapping onto the focuser's edge, and one onto the secondary's edge. That way, you can make sure that a) the phone is properly flat against the collimation cap because the circle maps onto the focuser's edge, b) the secondary mirror is round, and c) the secondary mirror is centrally aligned in the focuser tube.

And apart from anything, I still come back to the fact that this method is so much easier than peering into a collimation cap while fiddling with central screws and collimation screws. I would also very, very, very strongly recommend investing in some of these, which have also really helped with all this (I got mine from 365Astronomy but their page has gone - also unavailable at Amazon but at least you can see what I mean): Set of M4 Thumbscrews for Secondary Mirror Collimation https://www.amazon.co.uk/Set-Thumbscrews-Secondary-Mirror-Collimation/dp/B00UJUOXA4

Edited November 25, 2020 by BrendanC

[BrendanC]

Just to add: if you can cast your mobile device's screen to your TV, for example via Chromecast, then that could also work, BUT you won't get the option to superimpose Mire de Collimation on top of the image, which is kind of the point. Still, you'd benefit from it being a lot easier than peering into the eyepiece while making adjustments.

Edited December 2, 2020 by BrendanC

[Starflyer]

On 10/08/2020 at 19:55, Avocette said:

I’ve been using this technique recently. I tried an old webcam with poor results, mostly failing to physically lock the unit into an old eyepiece tube or similar 1.25” adapter. However I subsequently found that I could combine the 6mm CS-mount lens sold for use with the Raspberry Pi HQ camera, with my ASI120 guide camera. These cameras come packed with a CS-mount 150° ‘all-sky’ lens in an adapter plate that screws into the camera. With this in place this prevents using the 1.25” eyepiece focus tube sleeve, but the camera body already features a 2” focus tube sleeve. The 6mm lens fits in place of the 150° lens and has typical photographic aperture and focus adjustments.

This is the resulting image from SharpCap using the built in collimation circles when I had completed adjustments with the focus set to the centre ring on the primary.

One thing I learned in this process was that the ASI120 sensor chip is not perfectly centred in the 2” mount flange. I adjusted for these offsets using SharpCap’s Region of Interest settings.

@Avocette I know this is an old post but I'm interested in trying your method with an ASI120 MC-S.  I'd have to buy the camera and the CS mount lens, can you show me how it fits together with a 2" nosepiece please, and is the 2" nosepiece supplied with this camera, from FLO's page I can't see that it is?

Pictures would be great if you can, your description is good but I can't get my head around how I can have a CS lens and a 2" nosepiece attached at the same time.

 

Cheers,

Ian

[Avocette]

The ASI120 cameras have a built in flange which nicely fits a 2” eyepiece tube. This photo (from the ASI camera packaging) shows an ASI120 with the supplied 150° lens in the CS mount spacer. I just removed this lens and replaced it with the RPi 6mm lens, and the camera body itself fits into the 2” eyepiece tube.  

[Starflyer]

1 hour ago, Avocette said:

The ASI120 cameras have a built in flange which nicely fits a 2” eyepiece tube. This photo (from the ASI camera packaging) shows an ASI120 with the supplied 150° lens in the CS mount spacer. I just removed this lens and replaced it with the RPi 6mm lens, and the camera body itself fits into the 2” eyepiece tube.  

Thank you so much Ed, I had no idea that flange was 2" fit, great stuff.

 

Cheers,

Ian

[BrendanC]

@Starflyer I'm very interested in this bit: SharpCap using the built in collimation circles. 

Where is this in Sharpcap?