Collimation Confusion

[Kevdan]

Hi all,

Ok so I got my new Skywatcher laser collimator today, I made a jig similar to the one in the Astronomyshed video and checked and it was bang on, no need for any adjustment. SO now the story unfolds,

I set the scope up on its mount, checked and re alinged the secondary, and got this pic of the secondary and I think it looks quite good,

So then I put in the laser collimator and after adjustment got this result

So in my opinion it looks good too - but wait, when I look through my cheshire the collimation is a mile out. and with no collimator in the eyepiece this is the view I get

The laser point on the primary mirror is about 3-4 inches from centre of mirror (10" mirror)

so whats wrong?

Kev.

[Naemeth]

You probably need to collimate the primary as well as the secondary. As far as I am aware, through the Laser Collimator you are just seeing the secondary .

[Kevdan]

The first pic is of the secondar with the primary image blocked out with a yellow card behind, the second is with the block out and the primary adjusted to get the laser into centre of cross hairs.

Kev.

[squeaky]

With the laser collimating - did you centre the laser dot in the middle of the primary FIRST? Doing this by tweaking the adjusters on the secondary?

Then, AFTER getting the dot in the middle of your primary you do the adjustments on the primary to get the dot to "disappear" in the collimator as per your second photo.

If this is what you did there should be no great disagreement between the laser and the Cheshire result. If it isn't - that's probably why there's such a difference.

[Kevdan]

But according to posts, thats not how it should be done, I have watched vids from astronomy shed who say that if the secondary is a full round circle wich is centred as mine appears to be in the first pic then it should be ok, I understand what you are saying and will give it a go.

Kev.

[mdhardy01]

As above

The first pic shows a well rounded secondary but this is not colimated to the primary

Once you remove the card and put the laser in the focuser you then need to look down the ota to see the donut on the primary

Then using the three small screws on the back of the secondary tweak them until the point of the laser is in the middle of the donut

Once that is done then move to the back of the ota and adjust the primary to get the reflected laser to disapear into the hole in the colimated as in pic 2

Hope this helps

Matt

Sent from my iPad using Tapatalk HD

[mdhardy01]

But according to posts, thats not how it should be done, I have watched vids from astronomy shed who say that if the secondary is a full round circle wich is centred as mine appears to be in the first pic then it should be ok, I understand what you are saying and will give it a go.

Kev.

A rounded secondary only cures a rotational error but doesn't colimate the secondary to the primary

What astronomy she'd say is that you can't cure rotational error with a laser and that is when you use the card and sight tube to get a rounded mirror but then you have to get the secondary tilt colimated

Matt

Sent from my iPad using Tapatalk HD

[Kevdan]

Thanks Guys,

I did it the way you said and BINGO, however the secondary looks slightlt oval but I can live with that.

Thanks again guys,

Kev.

[squeaky]

In some scopes (mine among them) it is normal for the secondary to look slightly oval.

Glad to see that you got it sorted anyway.

From now on it should be a two minute job to do a quick check before observing

[umadog]

Good stuff. Iterate back and forth between secondary positioning in the focuser and the tilt until both look good. If it looks oval then you probably have a rotation error.

[squeaky]

What sort of effects do rotation errors have, please?

EDIT: Found this: http://starizona.com/acb/basics/using_collimating_newt.aspx

That's tomorrows homework sorted

[squeaky]

Oops - it wasn't oval that my scope is - it's the position of the secondary. And that was not on every collimation page I found.

So anyway - rotation error: According to this page:- http://starizona.com/acb/basics/using_collimating_newt.aspx I had a slight rotation error and the secondary was too close to the primary (longitudinal error).

Sorted rotation after quite a trial. This because the adjusting screws had dimpled the back of the secondary holder and kept pulling my minor adjustment back to where it started every time I tightened things back up again. I had to take the whole lot apart, carefully wrap and protect the mirror and then file the surface back smooth. Well, smoother. Having started at 1pm I was finally done with this bit by 4pm. <sigh>

Then for the longitudinal error - I only had about 1mm of space left available to move the secondary up (away) from the primary. Tried it. Made a heck of a mess of tilt and rotation so had a major job to get back to where I started. Had a lateral thought - in that if I can't move the secondary up - maybe I can move the primary down. As it turned out - pretty much all my previous checks and tweaks using my laser collimator had resulted in a net movement of the primary up the tube. So, using a 5mm allen key body as my gap gauge I adjusted all three points on the primary back until I just nipped the key between the mirror clamp and the base of the tube.

This was based on the premise that the tube mountings would be square to the axis of the OTA.

Wrong! A look through my Cheshire now showed an absolutely diabolical collimation. When I put the laser in - the return beam was hitting the centre of the primary but wasn't even hitting the secondary at all. Missed it by miles! I couldn't find a way to set the primary adjusters to get it any better than "just" hitting the edge of the secondary, so in the end I went back to tilt changes and deliberately offset it so that the return beam (while wrong on the primary) hit the secondary. Then adjusted the primary to centre the beam in the collimator - and then went back to tilt to centre the beam properly in the primary... and back to the primary...

So a mere six hours after I started I "think" I've very slightly improved my collimation. Only a star test, and an image or two are going to tell me.

The "engineering" of secondary holders, it seems to me after all this, (at least, on my Skywatcher), could do with a serious rework; although costs might not be worth it for those who don't have to adjust collimation very often. The whole assembly could be held in a rack and pinion assembly just like a focuser to allow easy longitudinal adjustment. Tilt adjusters - something like Bobs Knobs would be better - but the ends of the screws should be shaped (with a very slight dome?) and smoothed so that they don't dimple their contact points. And as for rotation - a worm drive with an adjusting knob would make it easy... especially if there was also a mechanism to give a set amount of "release" off the tilt adjusters to allow the rotation.

I think I might give this a serious look. My home site is pretty poor for viewing. Lots of high horizons and a fair old amount of light pollution, so I have to load it all in the car and drive to better sites when I want to do any serious imaging. After every such trip I end up having to tweak things. I'd guess not rotation and longitudinal most of the time - I'd have thought that once done they should stay put - but I'm not going to bet on it Food for thought, anyway, because there's every chance that some of these changes would actually introduce an allowance for movement or creep.

[Tinker1947]

Once the Secondary is set-up, it shouldn't need adjusting very often, the Primary is the bad boy and just loves attention, your lazer collimator looks like it is the same width each end make a cradle up with 4 nails and a piece of wood lay the collimator in the "V" made by placing 2 nails in the wood and the other 2 far enough apart that the collimator lays in both pairs, then switch on the laser and roll the collimator the dot on a wall should stay in the same place, i have a Baader which won't allow this but when placed in a lathe with the laser going out the gearbox end the dot at 12' feet only moves its own diameter when the chuck is revolved.....

[squeaky]

I have a spare finderscope mount which my collimator sits in nicely. At nine metres it draws a 1cm circle on the wall.

But that, as it turns out, isn't what's messing me up. There's a LOT of play when I sit it in the focuser tube to do my collimation. If I just tighten the holding screws I'm pushing the thing off to one side. Putting a single layer of tape around the body has helped but, after several trials, I still cannot take it out, put it back in again, and have the laser dot show up in the same place on the primary. And I'm talking two or three primary dot diameters here. <sigh>

So what I get is that I'll do collimation with the Cheshire to really really close, then do the laser, only to find that when I look back in the Cheshire again the collimation has changed and ALWAYS for the worse.

What makes things harder is that I only ever see my Cheshire graticule as a blur. ?

Here's a diagram of what my collimation currently looks like:-

So if my blurred Cheshire graticule looks centred on the main view of the secondary spider, and the main view of the secondary spider and their reflection from the primary are lined up... will I be collimated correctly?

Apart from the focuser reflection misalignment (up and left) which suggests that the secondary is still too close to the primary ??

[Steve Ward]

Try these sir , I found them a great help .

http://www.astronomy....php?f=19&t=102

[squeaky]

Nothing new there, although some of the things I'd learned the hard way - (but I did decide to fit in a swash plate and a friction sheet to stop the adjustment screws from digging in and make rotation correction easier since he clearly showed that it helps) - right up until video nine...

WOW

That Barlow and laser option completely sidestepped my earlier problem with the laser. 30 seconds later it was done! Yay!

Only a very minor tweak to the primary was needed and it hasn't altered the view through the Cheshire at all.

So thanks very much

That'll do for now.

Next rainy time I'm going to have a darn good look at the focusser, and then maybe go OCD on the secondary counting pixels in an image. I don't really know as yet if my focuser is on the skew and it is this that is at fault and that I may not have a longitudinal problem. If it's purely longitudinal then I'm in trouble because the secondary and primary are as far apart as I can get them. So I might need to drill a new set of mounting holes into the tube.

I'm sure that my collimation isn't too far away from being perfect and, under ordinary circumstances and viewing conditions would probably be absolutely fine; but I'm being rather OCD about this because my main aim with my kit is to hunt for Kepler Eclipsing Binaries - and to do that I'll be taking it way beyond normal viewing or imaging levels. Trying to resolve binaries from mag 14 to whatever extra mags I can squeeze from the camera (down to 18 or 20??) with separations of less than one arc sec. In fact, as little as 0.1 arc sec (or less) IF I can do it. At that stage I won't be expecting to see perfect points of light - but as long as I get two fuzzy blobs that over time move relatively to each other then I'll be a very happy bunny.

[umadog]

Once the secondary positioning in focuser is sorted it won't need doing often. The secondary tilt will need doing as often as the primary tilt.

IIRC the resolution of a 6" is about 1 arc second (give or take) in perfect seeing, so I don't think 0.1 arc second is possible. Ah, this supports that notion:

http://en.wikipedia.org/wiki/File:Diffraction_limit_diameter_vs_angular_resolution.svg

For trouble with secondary adjustment:

[squeaky]

Once the secondary positioning in focuser is sorted it won't need doing often. The secondary tilt will need doing as often as the primary tilt.

IIRC the resolution of a 6" is about 1 arc second (give or take) in perfect seeing, so I don't think 0.1 arc second is possible. Ah, this supports that notion:

http://en.wikipedia...._resolution.svg

Bum! I'll just have to push things as far as I can then, and accept whatever results I get.

For trouble with secondary adjustment: http://stargazerslou...ndary-syndrome/

Yep. A swash plate and a low friction layer - (a washer and a sheet of plastic) have fixed that one - though I'll be ordering a set of Bob's Knobs right after I finish this post.

[Mr Flibble]

Bob's knobs for the secondary are very pricey for what they are. You could always make your own. I used a standard M5 (I think, or could have been M4?) bolt and screwed on a wing nut, secured it in place with some araldite glue and then painted it black. Much cheapness and works a treat

[LukeSkywatcher]

In some scopes (mine among them) it is normal for the secondary to look slightly oval.

Glad to see that you got it sorted anyway.

From now on it should be a two minute job to do a quick check before observing

On my Heritage 130P, the secondary mirror is actually oval shaped. But when viewed during collimation...............its rounded. That last image was way out.

Glad you sorted it.

[squeaky]

Thanks for the tip on the knobs.

As for the collimation - I just UNsorted it. BIG TIME <sigh>

[Naemeth]

Thanks for the tip on the knobs.

As for the collimation - I just UNsorted it. BIG TIME <sigh>

Whatever you do, do it slowly! Make sure your eye is in the centre of the cheshire and be careful. It's very easy to misalign, and then you have another problem. When tightening the primary mirror screws, keep viewing through your cheshire/laser collimator, as over tightening can set it out again.

[squeaky]

I'm so far off now it makes that earlier diagram look good. I'm going to have to start again from scratch.

[squeaky]

Hokay, I dug up an old camera that will focus this bit and reloaded all the software for it - blah...

I've started again completely from scratch and first did the spider arms. East-West was 166.5mm each, and North-South were 165.5mm each. My upper OTA is oval. So I deliberately eased North-South as a pair and tightened East_West as a pair to get an even 166mm all round.

Here's a shot now down my focuser assembly:-

I drew in the green lines to help me line everything else up.

The difference in diameter between N-S and E-W using the first major change in colour working out from the centre is 1 pixel = 0.2% (ish) rotational error.

The two grey witness bars for East and West are straight copies - so I'm out a fair bit there. This I "think" is longitudinal error but I can't get the secondary and primary any further apart and nothing I've been able to do has changed this.

The two grey witness bars for North and South are copies of each other and show that I'm way out. I have no idea at all on how to fix this.

I can't get an image through the Cheshire - but it looks dire right now. I'm shattered and probably not thinking straight from tiredness right now, so I'll start again in the morning. Probably with a diagram of my Cheshire view. That'll help me review what I have and maybe prompt some answers.

Probably starting with removing the secondary and checking the focuser position and alignment. <sigh>

MUST get something at least acceptable by tomorrow night - EEK - TONIGHT - because the forecast is excellent here.

[squeaky]

Morning all

Right... Cheshire graticule and spider arms lined up - though it's a bit of a guess since I only ever see the Cheshire as a blur - but the hard line of the spider is centred in the blur.

In the image below - taken with my phone and using a 2x Barlow to get me in close - for the very first time I can see that the reflections of the spider line up with the actual spider arms. The donut is very slightly off but I haven't touched the primary yet.

I'm assuming that if I adjust the primary to centre the donut while keeping this view (and line up?) that I'll be really close?

Then I can finally finish off with the laser?

Somebody please say yes

I'm going to have brunch. Yes, you read that right, I've been up over three hours, and then will hit B&Q for M4 x40 bolts and M4 wing nuts. My poor old adjusting screws and indeed my allen key are suffering.

Once done I reckon I'll have a very long nap! Interrupted inevitably by Sally, my cat, who will want feeding at five. We have an excellent forecast here and hopefully I'll just need a quick collimation check and star test before getting back to work with trials on my new camera. Gosh, that's where I was days and days ago. Feels like a very long time.