A Collimation issue

[JohnH]

Having dived into the deep end by buying an f2.8 Newtonian I have been trying to collimate it ever since. With help from Carole I got close but it still wasn't right. One of my problems is astigmatism so that when I try to assess circularity it is mostly guesswork. As a last resort I have splashed out on an Ocal electronic collimator.

After setup, and applying the correction factor as per manufacturer's instructions I fitted it to my 'scope. The result, using as much offset as I can is this:



What is instantly obvious (well, at least I think so) is that the camera is not perpendicular to the end of the focus tube. The whole point of the process is to have concentric rings of focus tube, secondary mirror and primary mirror with the camera lens centered in the donut on the primary lens. ie I fall at the first hurdle.

The only thing I can think of would be to buy a tilt adjuster but before I part with even more money I wondered whether anyone had a better idea?

I should add that this image is without the extension tubes and filter wheel but the error is the same with them in place.

Desperately,

John

[Carole]

Hopefully Mark will see this post and reply, he has a fast newtonian and is good at this sort of stuff.

Carole

[MarkS]

What is instantly obvious (well, at least I think so) is that the camera is not perpendicular to the end of the focus tube.

I'm not familiar with the Ocal electronic collimator but I took a quick look at the website.  I assume you mean that you are unable to align with green circle in the image with the end of the focus tube.  How does the collimator attach?  Is there a possibility of using some packing pieces e.g. baking foil to give the collimator body some tilt?  Or maybe the camera is slightly tilted with the collimator body - is there some adjustment to it?

In any case it's obvious that the scope is a long way out of collimation.

Mark

[JohnH]

How does the collimator attach?

Hi Mark,

The collimator has an M42 thread which attaches either to an M48 to M42 adapter screwed into the top of the focus tube (as in the case of the above image) or to my filter wheel as if it was an imaging camera which has the same offset.

Yes, my first step is to centre the green ring on the end of the focus tube, then the red ring would be centred on the reflective edge of the secondary mirror and the blue ring centred on the edge of the primary mirror. After that the camera lens is centred on the primary ring donut under the cross hairs - in theory all very simple.

As I see it, the angle of the collimator cannot be altered on its thread. The centring of the collimator is controlled by a precise value provided by the manufacturer - this is individually defined for every camera manufactured.

My cunning plan is to use an M48 tilt adjuster to the (camera end of) the focus tube which is an M48 thread. I can then attach my usual extenders (reduced by the thickness of the tilt adjuster) and the electronic filter wheel. I hope that this will allow me to correct the angle of the focus tube and centre the optics and achieve collimation. I do not view this as ideal because every additional adjustable element to the image train must add a new chance for error to creep in.

My first thought was to use the adjustment screws on the focus assembly but then I realised that this is a single unit and therefore would not solve anything.

What I see does not surprise me too much. When I run Pixinsight's FWHMEccentricity Script I can see that the FWHM minimum is offset from the centre and Eccentricity varies from 0.4 upper left to 0.7 bottom right with 0.5 bottom left.

First Light Optics have just confirmed to me that the adjuster is in the post so, fingers very firmly crossed, it should arrive tomorrow and I can have a bash at it.

Regards,

John

[The Thing]

Hi John,
I hate collimation. Two nights ago my scope slewed to a new target while I was in the obsy and for some reason the secondary moved a bit. So I had to dismount the camera, set up my collimation camera and in -3.5C tweak those damnable screws and knobs. Nearly got it right. Then put it all back and start the sequence again.

It's no use adjusting the angle of the Ocal camera, surely it must be perpendicular to the focus tube if it's screwed on (check the threads?). I did a lot of digging on this Ocal thingy and never came across a report of the unit itself being misaligned. It's the focuser tube which is not right as Mark said.

I fixed my V-power focuser by a complete rebuild and setup of the steel rollers to get even centred pressure then shimming the whole thing ever so slightly. I degreased the rolling surfaces while I was at it. I also 3d printed a 2" holder for a QHY5L-II guidecam which allows me to focus an 8mm cs lens while the camera is installed to see the different collimateable elements and I use that with SharpCap to do what your doing with the OCAL. Maybe not quite as accurate...

It also looks to me like your mirror needs to come down the OTA a bit and then be tilted up at the bottom to make it round in the red circle.

[JohnH]

It also looks to me like your mirror needs to come down the OTA a bit and then be tilted up at the bottom to make it round in the red circle.

Thanks. I cannot help feeling that Sir Isaac Newton has not enhanced his reputation by inventing collimation.

I suppose that what I could do is pretend I don't have a problem with the focus tube and then correct my secondary mirror position and orientation and then proceed to adjust the angle of the secondary mirror to centre the primary followed by centring the camera lens. Then I could rotate the focuser and see what happened to the alignments. If it stayed true (unlikely) then my focuser is perpendicular to the OTA. If it wobbles around all over the place (likely) I could insert the tilt adjuster.

I tend not to play with rotating the focuser because that is likely to cause problems and I have approximately 2.5 x 1.5 degrees field of view to help frame the subject.

Regards,

John

[JohnH]

Another day, another attempt to collimate. After all, what could possibly go wrong?

Using a tilt plate I managed to achieve this (Image 1):



The end of the focus tube is concentric to the centre of the image. After a couple of hours fiddling I achieved this (Image 2):



Not absolutely perfect but better. So, what now? What now is that I am going to completely undo what I have done! To achieve Image 1 I had to pack the tilt plate about 3mm and I then had to pack the focus assembly to the same extent on the opposite side to achieve Image 2. If I rotate the focus assembly the "collimation" disappears. Before "correction" the secondary mirror and primary mirror donut remained stationary if I rotated the assembly. Obviously, this is not correct. Somehow the focus tube has an angle and I am going to have to live with it.

I am collimating with the reducer/flattener in place in the focus tube.The telescope manufacture says that it should be removed before collimating but it seems logical that optical elements should be left in place. Other owners appear to have adopted the same approach.

Regards to all,

John

[MarkS]

Wow!

This sounds like a nightmare. 

If the tilt plate needs 3mm to correctly point along the focuser axis, it suggests a serious issue with the collimator.  It's definitely not right and I would contact the manufacturer for support.

What do you mean by packing the focus assembly?  Again this doesn't sound normal.

Did you buy this scope new?

I agree with the manufacturer that the collimation should be done with the reducer/flattener removed. It makes it easier to collimate and the logic is that the reducer/flattener will also be central and will not affect collimation.

By the way, I speak from the experience of collimating a Tak Epsilon 180ED - a very similar design.

Mark

[JohnH]

Recollimated based upon just the centre of the primary mirror. The result last night was unusable comet shaped stars about 5 times larger than before.

Sigh. Back to the bench.

John

[Carole]

 

[MarkS]

Recollimated based upon just the centre of the primary mirror. The result last night was unusable comet shaped stars about 5 times larger than before.

I'm not clear what you mean by "collimation based upon just the centre of the primary mirror" but it doesn't sound like it was a success.

[JohnH]

I'm not clear what you mean by "collimation based upon just the centre of the primary mirror"

I assumed that the primary mirror was central and aligned the camera lens to the donut. It really was not successful. I have had another go with a Hotech Laser and a cats eye focussed having centered the primary mirror reflection in the secondary, in so far as I was able.

I don't think it is going to be perfect but I hope that it will be an improvement.

[The Thing]

As a footnote and FYI: I finally got my 6" newtonian collimated well (not nearly perfect yet) after various mods to the primary cell.

3D printed mirror mask which stabilises the mirror and stops it tipping  (feature of the GSO mirror cell ), mirror shimmed with foil tape to prevent 0.5mm lateral sliding (no pressure applied to edge), nylon mirror back supports 1/3 from the centre rather than the cork edge rectangles provided, 3D printed flanged grommets for the holes the collimation screw threads pass through to make adjustment smoother and stop the mirror shifting laterally as it moves (holes are way too big), bent the collimation screws so they are straight, stiffer longer springs which are also centred by the grommets so they don't foul the threads.

I would replace it but the way the OTA tube is cut the mirror face is only 30mm from the back end, the rubbish GSO mirror cell is very low profile and there is very little side clearance between the mirror holder and the tube, just enough for the springs. I would have to extend the carbon fibre tube somehow to fit a more normal alternative.

Next I have to address the secondary holder which rotates of its own accord when you adjust the screws, it is a Telescope Service one with not cheap parts so I don't want to replace it. I need to work out a way of fixing the rotation once its set while still allowing the tilt adjustments.

Duncan

[JohnH]

Duncan, what fun this hobby of ours is!

I got 60 (180 sec) Ha frames of the Cone Nebula last night with a view to using it as an L layer. I discovered, accidentally, that the ASIAir App now includes Solar System and Deep Space stacking abilities. This is my 60 subframe stack WITHOUT Darks or Flats - absolutely no processing or adjustments applied, just to see what I got:



I can't claim it is perfect but, to my eye, this is the closest collimation I have ever achieved. I think that I have fairly round stars and not too much in the way of horrible diffraction patterns.

I never was a good scientist and so I changed more than one parameter, possibly affecting the shape of my stars. I used a different guidescope/camera combination.

Telescope: Sharpstar 15028 HNT (f2.
Mount: iOptron CEM25P
Camera: ZWO ASI1600MM Pro (Gain 139, Temp -15C)
Filter: ZWO Ha
Guidescope: Starwave 50mm aperture f4
Guide Camera: ZWO ASI224MC uncooled
Control: ZWO ASIAir Pro
Stacking: ZWO ASIAir App (iOS)

I think I should leave the scope alone now, if I try to "improve" the collimation further I will probably just make it worse.

Regards,

John

[Carole]

Looks good, a processed crop wuold be easier to see.

Carole

[JohnH]

Looks good, a processed crop wuold be easier to see.

Hi Carole,

I have now stacked in Pixinsight, used the automatic Screen Transfer Function histogram adjustment and applied it to the image followed by a bit of vignette removal but nothing else to sharpen or denoise etc.. This is the result:



The stars seem reasonably round into the corners (blame my eyesight).
Regards,

John

[JohnH]

Just thought, I should have used the "Aberration Inspector" script in PI. Here it is showing the corners, mid sides and centre:



Actually,I am quite pleased with that, particularly as the scope did a meridian shift part way through the sequence. A bit elongated bottom right and top left (?) but still a lot better than my previous efforts.

John

[The Thing]

Looking good John, better than I've managed with my TS1506UNC. If you look at my latest image you can see the diffraction spikes in one axis have started diverging - the secondary has moved during slewing and acquisition. It also affects focus at the sides of course. I am working on a spider head design to allow me to use a screw to control rotation so it can be set securely but still allow the tilt adjustments to be made. It'll be 3D printed when I get there. I might go for curved spider vanes while I'm at it then the problem won't show (no diffraction spikes!)

Duncan

[Carole]

That's looking so much better John.  But whilst i enjoyed them I am glad my Reflector days are over.  Refractors are much less hassle.

Carole