Newtonian Secondary Collimation and Eye Placement

[Captain Scarlet]

A week or two ago, I received in the post the latest addition to my growing and wife-alarming telescope collection: a 2^nd hand Skywatcher 200p Newtonian. I now have the whole gamut here: a 105mm pure refractor, a 200mm pure reflector, and an in-between in both senses, a 150mm Mak.

The scope arrived with a single-speed focuser, which I replaced with a Skywatcher dual-speed that a friend very kindly gave me (off his SW 300 Quattro). Incidentally, that SW dual-speed is a very nice unit.

The point of this new scope is to use it as an experimenter, to help me get experience in all the aspects of collimating a reflector, so that I can waste as little time as possible and be thoroughly familiar when I need to do anything to my 300p when I’m in my dark place. It turns out that it’s teaching me quite a lot.

In particular, I was most curious to see precisely how the secondary-mirror attachment arrangement actually works. All the instructions on the net that I’ve seen about collimating the secondary say screw this, turn that, judge roundness by eye, adjust these three screws, all essentially trial-and-error. What none of them do is show an exploded diagram of exactly what the secondary-holder comprises, which should immediately suggest what needs to be done and in what order.

In that spirit, the first thing I did was completely dis-assemble the scope, removing secondary mirror, spider, primary mirror, mirror cell and focuser.

My plan, and the initial experience, having re-assembled some of it, was as follows, as described by many instructions online:

1.       “Square up” and “coincide” (noun-as-verb alert! terrible crime I know, I know, sorry!) the optical axes, i.e. the main tube axis and focus tube axis;

2.       Viewing from a collimation cap, make the secondary appear centred and circular at the bottom of the focus tube, using centre-screw and hand-twist of the holder;

3.       Tweak it with the 3 grubs such that the reflected primary mirror-cell-circle appears centred in the collimation-cap view;

4.       Further micro-tweak it such that the focuser axis extends exactly to the centre of the primary (with a 90deg bounce in between); (Cheshire or laser)

5.       Adjust the primary orientation to direct its own axis back directly to the collimation-hole (with another 90deg bounce); (Cheshire or laser)

I did all this, but having done step 2, I was expecting 3 to be quite a small adjustment. It wasn’t, I had to use the centre-screw and the three grubs and a deal of trial-end-error to get the reflected primary-mirror-cell circle into symmetry whilst trying to keep the secondary itself centred and circular: it was much more than a tweak and took a while.

I got there in the end, and that night the brief view of the stars I got before clouds rolled in was nice, but that “more-than-a-tweak” was bothering me.

I drew myself a little picture of what was going on (see below) and as soon as I’d done that, the problem leapt out at me. I now think that some of the common instructions available about how to collimate the secondary contain a mistake. Specifically about where to put your eye when getting the secondary to appear circular.

Many instructions advise racking the focus-tube (and hence collimation-cap and viewing-hole) out as far as it can go, or even using an extender as I did, to allow the guaranteed-circular far end of the focus-tube and the edge of the secondary to come close together for accurate evaluation of circularity.

The secondary is elliptical, and when viewed from the side can be made to look “more circular”. This is generally explained, even by the more academic instructions, as a “perspective effect”, which it is, but I prefer to think of it as being the situation where the outer edges of the ellipse subtend exactly the same angle from your viewing point, regardless of how far away those edges are. The circle is therefore a projection of the ellipse from a certain viewpoint.

However, with the secondary placed in its perfect position, angled at 45degs and starting and ending precisely at the edges of the converging primary’s light-cone, and with your eye placed precisely perpendicular to the point at which the centre-line bounces from the primary-centre off the secondary, there is only one place where the angles subtended around the edges of the secondary are the same: the point of focus of the primary. That’s where the eye must be placed.

With the eye at any other position along that line, for a properly-placed secondary, the correct view of it is not circular: it necessarily must have different angles between near and far edges.

Which means: when doing step 2, getting the secondary to appear circular at the bottom of the focus-tube, your eye must be at the focal point. Interestingly, the more academic of the instructions that I’ve come across, including Suiter and Nils Olof Carlin, do state this, but don’t say why.

So those sets of instructions that recommend racking out as far as possible to get a better circle-comparison I think are wrong, and perhaps this explains why I had such difficulty going from step 2 to step 3. I used an 80mm extender to “rack out” as far as possible.

If you do what I did and “wrestle” it to step 3 after step 2 using all the degrees of freedom available, you’ll likely end up, even after primary-collimating, with a secondary looking circular and bouncing light to and from the right places, but in the wrong place: i.e. not exactly coinciding with the primary’s light cone and not at 45degs (i.e. the wrong offsets). If it’s not at 45degs, the focal plane will be tilted.

Next steps: re-do it over the weekend, and do the maths for all this. I suspect it all doesn’t matter too much for sufficiently slow telescopes, my diagram is that of an f/1.33 scope for easy illustration. But then again I think that AP is quite a harsh judge of “set-up”.

Thanks for getting this far, if you have,

Magnus

Edited October 25, 2019 by Captain Magenta

[johninderby]

The Telescope Express Concentre is a pricy but simple device for getting your seconday mirror perfectly set up. I had thought my secondary was perfectly set up but found it wasn’t after using the Concentre. Made me realise how imprecise a Cheshire or colimation cap is for setting up the seconday position.
 

Interstingly found my laser colimator gave much better results after setting up the secondary using the Concentre.

 

Edited October 25, 2019 by johninderby

[Piero]

If the secondary mirror is slightly oversized, its alignment is far less critical than what one would imagine, for visual work. For imaging, it's more important.

Try to get the three rings as concentric as possible: edge of the sight tube, secondary mirror, and primary mirror.

[markse68]

Intersting theory Magnus- not sure I get why looking at it further away would make it not a circle but I'm going to bear it in mind and try it next time (hopefully soon) I have to remove my secondary and set it back up again.

[jetstream]

2 hours ago, Captain Magenta said:

Many instructions advise racking the focus-tube (and hence collimation-cap and viewing-hole) out as far as it can go, or even using an extender as I did, to allow the guaranteed-circular far end of the focus-tube and the edge of the secondary to come close together for accurate evaluation of circularity.

This adjustable sight tube works very well and very accurately. I removed the 24" spider/sec for a fiber batten tweak last night and re collimated. Sec is accurately centered and collimation is not perfect (no AC) but very very close. You can extend the tube into the focuser to get a really fine view of the sec edges. Fast, accurate, simple and expensive.

[jetstream]

1 hour ago, johninderby said:

The Telescope Express Concentre is a pricy but simple device for getting your seconday mirror perfectly set up. I had thought my secondary was perfectly set up but found it wasn’t after using the Concentre. Made me realise how imprecise a Cheshire or colimation cap is for setting up the seconday position.
 

Interstingly found my laser colimator gave much better results after setting up the secondary using the Concentre.

 

How does the Concentre work?

[johninderby]

Because it uses circles instead of a crosshair it is a lot more accurate. You move the drawtube in or out until one of the circles matches the outside edge of the secondary mirror. Rotate and move the mirror up or down until it is a perfect circle. This also takes care off offset adjustment. They do 1.25” and 2” versions. Takes eye of eye placement as well during setup.

https://www.teleskop-express.de/shop/product_info.php/info/p5506_TS-Optics-Concenter-2--Collimation-Eyepiece-for-Newtonian-Telescopes.html

Edited October 25, 2019 by johninderby

[jetstream]

1 hour ago, johninderby said:

Because it uses circles instead of a crosshair it is a lot more accurate. You move the drawtube in or out until one of the circles matches the outside edge of the secondary mirror. Rotate and move the mirror up or down until it is a perfect circle. This also takes care off offset adjustment. They do 1.25” and 2” versions. Takes eye of eye placement as well during setup.

https://www.teleskop-express.de/shop/product_info.php/info/p5506_TS-Optics-Concenter-2--Collimation-Eyepiece-for-Newtonian-Telescopes.html

Sounds interesting, I might try one. With the Teletube the end can be extended and the focuser racked so it gives a "circle" edge the same as the sec circle picture sight. This will show the position under the focuser very accurately as well.

How do you find the accuracy of sec alignment to the primary and vice versa using this tool?

 

[johninderby]

APM has quite a few variations of the Concentre.

https://www.apm-telescopes.de/en/othernbspaccessories/artificial-stars

Accuracy is spot on. It compensates for any other mechanical missalignments and ends up with perfect optical alignment. Also can be used as a regular collimation tool. 
 

Some generic instructions. You can tell it was translated from German. 🙂

Collimating with the Newton Concenter Eyepiece
 This manual should be a small help to align (collimate) your Newton Telescope in a simple and practical way.
When you hold the Newton Aligning-set in your hand, you will notice at the bottom side a look-through plastic-disc with a central hole and concentrically engraved rings.
At the other end you will find a small single hole , that was on purpose made so small that the eye will not have a chance to be in a wrong position and as such cause a faulty alignment. After we inserted the Concenter Eyepiece, in the Focuser, you will see matt half transparent rings in front of the optical components of the telescope.

1. Align the Focuser
With a covered primary mirror !(maybe a piece of paper or main mirror cover)
When using this alignment set, we should pay attention to work in an environment with sufficient light.
Assuming that the focuser was fitted on the telescope tube in a straight angle (90deg), we can now align (shift or rotate) the secondary mirror in that way that he becomes concentric with the rings from the Newton Alignment Set.
With the focus-adjuster of the telescope we can bring one of the concentric rings into that position so that the outer side of the secondary mirror aligns with the fitting ring of the Concenter Eyepiece. By doing so, we can detect already the slightest change from the ideal position.
Attention !
Your align includes the Offset of your Telescope so if you have a real middle mark on your secondary ,it is not in the middle of the concentric rings. Only if you have an Offset mark on your secondary its in the middle

2. Align the secondary mirror with the prime mirror
When you look into the eye-piece you should change the angle of the secondary mirror in that way so that the bright reflection of the prime mirror in the secondary mirror is concentric with the rings of the alignment set.
!

3. Align the prime mirror to the optical axis
When working in darkness you can enhance the visibility of the ring-disc by working with ,for example, with a LED-Lamp.
Now the prime-mirror middle-marking can be aligned with the bright rings that are now lit up. In the beginning the reflections from concentrically rings on the reflection mirror can cause some confusion.
Verify and train the mirror manipulations before you start using the Newton Concenter Eyepiece.
When aligning the mirrors, we should only use the main circles from the Alignment-ocular. When you followed these instructions carefully, you will not need a final align using real stars.
 

Edited October 25, 2019 by johninderby

[Captain Scarlet]

5 hours ago, Piero said:

Try to get the three rings as concentric as possible: edge of the sight tube, secondary mirror, and primary mirror.

Agreed. But only looking from the focal point, is my main point.

By the way, your laser collimator is doing great service here...

[Captain Scarlet]

On 25/10/2019 at 19:28, johninderby said:

Because it uses circles instead of a crosshair it is a lot more accurate. You move the drawtube in or out until one of the circles matches the outside edge of the secondary mirror. Rotate and move the mirror up or down until it is a perfect circle. This also takes care off offset adjustment. They do 1.25” and 2” versions. Takes eye of eye placement as well during setup.

https://www.teleskop-express.de/shop/product_info.php/info/p5506_TS-Optics-Concenter-2--Collimation-Eyepiece-for-Newtonian-Telescopes.html

I now have one of these. As you say John, it's expensive for what it is, but so specialized that it probably couldn't be any other way. It's also well made. In addition I have a very high quality (Glatter) laser which I've verified for collimation over a 25+m distance. I'm very much looking forward to the weekend to using this tool to see how my 2nd attempt "from the focal point but eyeballing a circle" worked out, which I did this week; and of course I'll be disassembling and starting again and comparing results with the laser. I'm one of those people who enjoys the thinking, setting up and tuning as much as the using.

M

[johninderby]

A simple device but I now consider it the only way to set up a secondary. You might be surprised at how far out your secondary is when you check it with the Concentre.

Edited October 31, 2019 by johninderby

[niallk]

Very interesting post Magnus!  I like your analytical approach 

I took the controversial step of centre dotting my secondary per Dave Kriege's suggestion, and I use an Astrosystems laser for collimating in the dark.

The 'concenter' tool looks interesting.

-Niall

[NGC 1502]

Lots going on here.   I’d just like to comment on one thing that’s been mentioned......

Having a slightly oversized secondary mirror is a good idea.  It does give more latitude to get collimation sorted. Also it’s often said that secondary mirrors can have issues at the edges of the reflective surface, so an oversized secondary avoids using those outer edges.

Also a slower focal ratio makes things less critical.  That’s why I prefer my 8” f6 Dob to my 10” f4.8, easier to collimate and with good eyepieces coma is a non issue for me.

I’m often at my club’s dark site, I take my 10” for the extra light grasp, but at home my 8” gets used........I could borrow a larger club scope, but the instant set up of a solid tube is just so convenient 👍

Of course as always.....your mileage may vary.......🙂

Ed.

[Captain Scarlet]

With the Concenter concentric-circles sighting tube in my possession, I used it to judge my prior efforts at centering the secondary at the bottom of the focus tube, and then to try to correct it. It worked well, and didn't take long.

1. My last week's effort, having centered it "by eye" and viewing from the wrong place (10cm further out from the focal plane). Clearly not centred, but perhaps not too bad ... ?

2. The same view through the Concenter. Looks a lot worse when compared to some actual circles...

3. After re-centering, much better:

4. With the primary mirror exposed, and after having done the final tweak with laser and also checked with a Cheshire, all of which were consistent:

... the only thing that remains now is to actually look at some stars!

Edited November 4, 2019 by Captain Magenta

[johninderby]

Such an easy tool to use and now your secondary is spot on. 👍🏻

 

[Captain Scarlet]

... fiddling around with this again over the weekend I put the laser in the eyepiece-holder with the laser dot on the primary centre-spot. As I moved the scope around on its mount, the spot moved around On the primary too! Up to 5mm of movement at the primary, and definitely not the primary sliding around, I shimmed that in.

The focuser is a newish SW dual-speed one as supplied with their Quattro series which a friend gave me. It FEELS secure enough, there’s no *feeling* of movement when I stress it manually, but the spot also moves when I apply finger pressure to the eyepiece holder.

I’m not sure if it’s tube flexure or focuser flexure, ie whether it would be sorted by getting, say, a Moonlite...

 

[markse68]

Could it just be the laser collimator shifting in the ep tube?

[Captain Scarlet]

No, first thing I checked. And the spot, wherever it was, stayed stable if I rotated the laser. Laser is a Glatter, which I confirmed is itself collimated a couple of weeks ago.

[jetstream]

On 18/11/2019 at 14:39, Captain Magenta said:

Up to 5mm of movement at the primary

Not good, but nice catch. I found out playing around that unless there is a bunch of focuser slop, that drawtube flexure only affects sec collimation when racked out and with a hvy weight.

I cannot use binoviewers in my VX10 because of tube flexure and in the 15" with a Moonlight collimation stays steady until over 1/2 racked out with the Binotron 27's.

Obviously your spider vanes are tight? Sometimes they can loosen up from pulling the tube in.

[Captain Scarlet]

16 hours ago, jetstream said:

... when racked out and with a hvy weight. ...

The laser I’m using is quite heavy, the dual-size Glatter, and the Delos eyepieces I have are heavier still and taller.

I’m going to play around further this weekend and try to isolate what’s moving. The vanes and spider all seem good, as the spot doesn’t shift at all when I grab the secondary and apply torque with my hand.

My best bet is it’s draw-tube flex, but I need to pin it down before I go Moonlite-ing.

Edited November 21, 2019 by Captain Magenta