409 replies to this topic

[psandelle]

All this is good stuff! I've still wary at the moment. I'd usually jump in on something new like this, but.... I don't want to tinker much. Spacing is enough of a hassle on fast newts. I shall continue to watch with interest!

 

Paul

[andysea]

I will definitely keep everyone posted. At this time the scope seems perfectly usable with an APS-H sensor. Agena sells m48 aluminum spacers and those are probably the ones that I should use. However for now I used my 3d printer to print a few of those. They are probably not perfect but I hope they are close enough to tell me if I'm on the right track.

[Aemaeth]

Thanks a lot for your investigations and help

 

As for myself, I plan to buy a ASI1600mm or QHY163m and filter wheel to use on it.
But as I mentioned earlier, I plan to do this later in the year.

It will give me plenty of time to get accustomed with collimation at first, and maybe try to solve small issues that you mentioned as the not blackened screws, and maybe other things related to the screws to adjust the primary and secondary.
 

[psandelle]

I will definitely keep everyone posted. At this time the scope seems perfectly usable with an APS-H sensor. Agena sells m48 aluminum spacers and those are probably the ones that I should use. However for now I used my 3d printer to print a few of those. They are probably not perfect but I hope they are close enough to tell me if I'm on the right track.

The intrigue to me (I cut my teeth on 6" f2.8 PowerNewt scopes) is if it can be flat across a full-frame chip. APS-H is a good start to that working.

 

Paul

Edited by psandelle, 19 August 2019 - 07:04 PM.

[andysea]

I have a few hours of data with the ASI1600 and the AD 5nm h-alpha filter. I will post the results later but it looks like there is some residual star elongation in one corner. I was also thinking to test the scope with my Canon 5D4, for some reason the Canon to m48 adapter that I have is about 10mm instead of 11mm; So it puts the camera at 54mm instead of 55.

Robofocus installation was a breeze.

 

It looks like tonight we will have some breaks in the clouds....more to come.

[Eddgie]

Managed to do some testing tonight with the new SharpStar.  I put on it my QHY16200A, which has a 21.5x27mm chip, so is a good test for the sharpness to the corners of this scope.  I collimated the scope using my Glatter laser and tublug and found the defocused stars tonight gave good collimation at the center of the frame.  There was quite a bit of clipping as one goes near the edges of the frame in the defocused stars.  Here is a view of that

 

Screen Shot 2019-07-29 at 11.56.49 PM.jpg

 

My question is whether the clipping of the light shown in the defocused stars would manifest as an aberration in the focused star.  It would certainly affect full field illumination, but I'm not too certain whether it'd affect star shapes when in focus.  

Clearly there is something vignetting the field.  The way the patterns are cut is the result of an upstream circular restriction.

 

Rememeber, at f/2.8 your fully illuminated circle grows 1mm in size for every 2.8mm distance from the focal plane.

 

This means that if you have any circular restriction that is not big enough to pass the f/2.8 light cone at the distance it is in front of the focal plane, it vignettes the field.

 

So, you can't use something like a T2 adpapter unless it is pretty close to the focal plane of the camera.

 

Anyway, that is what the image is showing.  Something is vignetting the field. 

[andysea]

If the concern is the asymmetrical image of the highly defocussed stars, that is pretty normal. These fast newtonian have the secondary that is offset from the center of the spider vanes. The epsilon is the same way. 

Attached is a drawing showing the front of the Sharpstar that clearly shows the secondary offset.

Also check out the photo of the front of the Epsilon 130. The secondary offset is clearly visible.

 

 

Attached Thumbnails

• 
• 

[AstroGabe]

Clearly there is something vignetting the field.  The way the patterns are cut is the result of an upstream circular restriction.

 

Rememeber, at f/2.8 your fully illuminated circle grows 1mm in size for every 2.8mm distance from the focal plane.

 

This means that if you have any circular restriction that is not big enough to pass the f/2.8 light cone at the distance it is in front of the focal plane, it vignettes the field.

 

So, you can't use something like a T2 adpapter unless it is pretty close to the focal plane of the camera.

 

Anyway, that is what the image is showing.  Something is vignetting the field. 

Thanks for the comments.  For those tests, I had used an m48 to m54 adapter for the QHY.  I had a 48mm clear aperture to the chip.

 

Gabe

[psandelle]

Isn't there an M54 or M63 way to go on that focuser? I can't find the documentation that said which. I do nothing less than M54 for all my full-frame needs, and that was what was another cool thing is this wasn't just M48 (which I hate).

 

Paul

[AstroGabe]

Isn't there an M54 or M63 way to go on that focuser? I can't find the documentation that said which. I do nothing less than M54 for all my full-frame needs, and that was what was another cool thing is this wasn't just M48 (which I hate).

 

Paul

Yes, other vendors are listing it as 63x0.75mm.  I thought about unthreading the corrector from it, and putting it into a 2.5" Feathertouch, but that spec is 63x1mm, unfortunately.  I'm sure a 63mm to 54mm adapter from precise parts would make the vignetting a lot less of an issue for a sensor of that size.  

 

Gabe

[andysea]

I noticed that in my sample the spacing of the secondary is 8.5mm instead of the 5mm recommended in the collimation manual.

I wonder if that could cut off some of the light and cause the clipping of the image seen in the out of focus stars.

 

 

Attached Thumbnails

• 

[andysea]

I did some extensive collimation work yesterday, including moving the secondary further out toward the front of the tube. 

I was only able to grab a few shots in between the clouds. Here is one 4 second TIFF. 

The obvious vignetting is caused by the DSLR mirror box.

 

https://drive.google...iew?usp=sharing

 

I didn't check the out of focus stars because the camera itself causes obvious vignetting so it wouldn't be a good test of that. I think that the next step to correct the corners would be playing with the spacing, however since I want to use the Nikon, I can only increase and not reduce it, unless I move the corrector out which I'm not sure is possible at all.

[AstroGabe]

I did some extensive collimation work yesterday, including moving the secondary further out toward the front of the tube. 

I was only able to grab a few shots in between the clouds. Here is one 4 second TIFF. 

The obvious vignetting is caused by the DSLR mirror box.

 

https://drive.google...iew?usp=sharing

 

I didn't check the out of focus stars because the camera itself causes obvious vignetting so it wouldn't be a good test of that. I think that the next step to correct the corners would be playing with the spacing, however since I want to use the Nikon, I can only increase and not reduce it, unless I move the corrector out which I'm not sure is possible at all.

I would have hope you have at least a little wiggle room to reduce the spacing between the ccd and corrector.  This vendor says the reducer is removable.  I presume you can thread the corrector a bit closer to the ccd.  This may affect spacing with the primary, so you might need to move the primary up in the cell a bit w/ the collimation screws.

 

https://www.skypoint...-corrector.html

[andysea]

I thought about that but when I unthreaded the corrector, even just a little bit, the stars got really wonky. I think it's because once unthreaded the corrector wobbles a bit. 

[AstroGabe]

I thought there was a 63x0.75mm retaining ring on the ccd side.  Perhaps there's a hard stop on the OTA side.  Maybe a retaining ring acting as a spacer between the corrector and hard stop would hold the corrector more firmly in place?  

[andysea]

On the ccd side there is a m63 to m48 adapter. In the draw tube I didn't see a hard stop but I need to investigate further. If there is, I can use my 3d printer to print a shim of the right diameter. It's raining all day today so that will be the project for the evening. If that works out I can try with 0.5mm and 1mm and go from there.

Even as is, I think I can use it. Another thing that I noticed is that the DSLR adapters introduce an enormous amount of play/tilt. I had to shim my adapter quite a bit to remove the play between it and the camera.

Every time that I added a layer of tape to shim it I had to refocus, that was a tangible demonstration of how tiny the CFZ is. Not that I didn't know it but it's still interesting to see that.

 

 

EDIT: Come to think of it, I think you're right. There must be a hard stop in the draw tube. I just didn't look for it. I'll report back shortly.

Edited by andysea, 21 August 2019 - 09:42 AM.

[psandelle]

That vignetting doesn't look that bad at all (and should be better if I use the M63 attachment). Definitely spacing (maybe a mm or two out more?), but once that's taken care of (if possible; it's always a pain), there might be a little weirdness in the edge stars beyond that; might not be, hard to tell with the spacing issue making them elongate. Still, not bad, generally. Or, maybe I've just seen much worse! 

 

Paul

Edited by psandelle, 21 August 2019 - 10:45 AM.

[andysea]

Quick update. I sent  a test image to Astronomics 6 days ago to get their feedback on the collimation and the star elongation. Since I haven't heard back I emailed Sharpstar directly. 

Michael Fong responded with a few suggestions within a day. Good customer support and responsiveness from Sharpstar!

Michael suggested loosening up the retaining ring of the mirror. If the mirror is clamped too tight it can introduce aberrations which makes perfect sense. Done!

He also noted that I can try to increase the spacing by 0.5mm. That will be much easier than reducing it. So I have all the needed shims and should be able to do another quick test tonight. 

[Aemaeth]

That's definitely a wonderful support from Michael.

 

In fact, I got mine yesterday. After a quick trip down Subway in Paris, and Bullet train to Le Mans, it arrived safely at home.

But...Almost all the screws were loose. The ones retaining the alloy base to the tube, the one retaining the primary clip, and 2 nylon tipped screws for the primary had fallen...

 

So, I screwed everything together, but I think that could be a good idea to put a small amount of blue Loctite to prevent theses screws from getting loose during transport?

And I'm also starting to find black screws and nut to replace all of them in the scope

Edited by Aemaeth, 22 August 2019 - 10:48 AM.

[psandelle]

That's definitely a wonderful support from Michael.

 

In fact, I got mine yesterday. After a quick trip down Subway in Paris, and Bullet train to Le Mans, it arrived safely at home.

But...Almost all the screws were loose. The ones retaining the alloy base to the tube, the one retaining the primary clip, and 2 nylon tipped screws for the primary had fallen...

 

So, I screwed everything together, but I think that could be a good idea to put a small amount of blue Loctite to prevent theses screws from getting loose during transport?

And I'm also starting to find black screws and nut to replace all of them in the scope

This "loose screw" business is really disheartening. It's a simple thing to do right, and there can be consequences of not doing it right. If I do get one, I'm going to ask the vendor to double-check everything and cherry it out for me BEFORE I receive it. I don't want scratched mirrors, lost screws, stuff like that. I have enough to deal with. 

 

Paul

[gionk]

First of all I need to say, that I'm no ways a collimation expert and my knowledge of Newtonian Collimation before buying this scope was limited to single point laser collimation. I want to report my progress so far, and maybe you can give me hints where I misunderstood some concepts, or maybe some of you can learn with me together in this journey.

 

Also, I would like to point out that I'm constantly in contact with Michael from SharpStar. He's super helpful and responsive and he just sent me some replacement hex screws (for the focuser) and one replacement holder ring knob for free! Really happy with their customer service.

 

I was able to capture some data after my recent collimation attempt. I've noticed only later that my focus was far from perfect. Also, since this shot I've also recollimated the scope from scratch, but I will tell you more about my latest collimation attempt and some findings further down.

 

First, here's a shot of NGC1582 open star cluster. It was taken with my ASI1600MM (3.8 micron) and through a blue unmounted 31mm ZWO filter.

 

I was using a variable spacer to create a distance of exactly 55mm between the sensor and the adapter surface as shown in the guide of the scope. I tried many different combinations and so far 55mm worked best for me. When I plate solve this image, the solver tells me that the focal length of my scope is 416.25mm.

 

Overall this was my 4th attempt to collimate the scope and I've used a sight tube, cheshire and autocollimator from CatsEye. My mistake was that I did not correct for any focuser misalignment, which in fact led to a non-concentric and even partially cut of primary reflection within my secondary. I've corrected all focuser axis errors using the secondary adjustments and since my focuser tube is off quite a bit, this lead to a severe secondary misalignment in respect to the tube axis and therefore some light loss of the primary. I took that into account within my latest collimation attempt which I will describe later.

 

Aberrations

 

Mechanically my collimation was horrible in this 4th attempt, as described above. However, optically I think it was bad but not the worst. Here's a prepared mosaic using aberration spotter in Pixinsight:

 

Looking at the aberrations, besides the not so perfect focus, I think I can see some radial elongated star shapes in the corners which indicate too much corrector distance. If it would be too less distance, there would be some coma trails pointing away from the center point. Also, I think the primary collimation is quite okay since the radial star shape aberration seems to be symmetrical in the image.

 

Another aberration is the elongated star shape throughout the whole image, also in the center. I extracted the PSF of the center 26 stars in the center area 1500 x 1500 pixel and here's the zoomed image of the unstretched and stretched version: 

 

I was also running eccentricity analysis on that center area and it calculates a median eccentricity of 0.43 which coincides with the PSF. The aberration could be a medium astigmatism of the primary or secondary. Unfortunately I did not notice the aberration during my capturing session, so I could not verify if it was the primary. I think it would be quite easy to verify. If you move the focus from intra to extra, the elongation should switch direction / should be mirrored. If that's the case the astigmatism is within the primary. It removed my primary to mark it with a Catseye triangular center mark and I verified that the screws on the primary holder are fixed with almost no torque, just the force of my fingertips. If this aberration persists, I will investigate further. The astigmatism could also be a cause of the bad mechanical alignment I guess and the effect of some turned down  edge aberration which is introduced if the secondary edge is returning a lot of the light.

 

This last thought about mechanical misalignment also brings me to the last aberration which I think is caused by that. All the stars on the left edge of the image have kind of a pacman shape. I hope this was just caused due to my bad mechanical alignment.

 

My most recent collimation and some findings

 

I did a complete recollimation of my scope after reading a few times through the excellent book of Vic Menard "New Perspecives on Newtonian Collimation". After reading through it, I got the understanding that there are two common ways to offset the secondary mirror in fast newtonian telescopes. The classical approach includes an offset in both directions (bi-directional), away from the focuser and toward the primary mirror. I also read that this offset is putting more constraints on collimation, since the mechanical alignment gets much more important. Specifically the focuser should be perpendicular to the scope tube. This is not the case for the "New model" of offset or uni-directional way how to offset. Although Vic, who I consider the collimation guru himself, suggest to go with the "New model" (uni-directional) offset, the SharpStar 150 uses the classical way. I think this was one of my issues when collimating previously. I did the whole mechanical part, aligned everything concentric under the focuser, and then collimated optically using a cheshire and the autocollimator. I did so by adjusting the secondary holder adjustment screws. Now, the problem (at least with my scope) was that my focuser draw tube is far from perpendicular to the tube axis. So when I did the focuser axis collimation, I introduced a huge mechanical colimation error in that the secondary was not reflecting the primary mirror and it's holder as a concentric circle.

 

I have now fixed that by following this protocol: I first did a mechanical alignment and aligned all components concentrically and eliminated any eccentricity. Now the primary mirror holder ring was perfectly aligned within the circle of the secondary mirror. But now the problem was that my focuser axis was off optically. Instead of adjusting the secondary adjustment screws, I now started to calibrate the focuser drawtube with the six adjustment screws. This corrected the optical error in the focuser axis but did not introduce much mechanical alignment errors. The primary holder reflection was still concentric within the seondary mirror.

 

As you can see from the above photos, I needed to offset / shim my focuser quite a bit in order to achieve the best possible compromise between optical and mechanical alignment. I guess as a result, and because of the fact that this scope uses "classical" offset design, my focuser should now also be more or less perpendicular to the tube axis.

 

One thing I found with all this experimentation and specifically by using the 6x - 8x error magnification of the autocollimator, is that, at least in my case, the focuser itself, introduces angular axial misalignment when moved. This scares me a bit. Within the autocollimator, I can clearly see an angular change when moving the focuser in the opposite direction. As soon as the angle changes, due to change of focus direction, the focuser is stable. However, this is quite a problem, since this forced me to decide which focuser movement I prefer to do the collimation with. I now collimated everything when the focuser moves inward. However, now I need to make sure, my autofocus routine stops with an inward movement of the focuser. I'm still in exchange with Michael from SharpStar to see if that's an issue with my scope or misconfiguration form my side.

 

Here's a video of the angular change of the focuser tube, when you change focus and direction: https://photos.app.g...f6RGg4WioQL1r18

 

I yet have to test my latest collimation, in which I put a lot of hope. I'm a pixel peeper, so for me I can't live with eccentric stars or non flat fields and such, so I still hope I will find a way to get what I need. Let's see, weather's not too good here in Switzerland at the moment.

 

Looking forward for your comments!

 

Cheers

Gion

[andysea]

Thanks for your report.

I had to learn how to collimate the scope which is just a newtonian and now I have a pretty clear understanding of the process.

 

My understanding is that there are three main scopes:

 

1 Center the secondary under the focuser.

2 Center the secondary on the center spot of the primary

3 Center the primary.

 

The laser that I used (Hotech SCA) Is next to useless due to the non repeatability of the tilt of the beam when inserted in the eyepiece holder. Even a 2" holder does not guarantee perfect squareness.

 

My procedure is quite simple and only uses sight tubes with crosshairs.

Here is a summary

 

- I removed the corrector and installed a 2" eyepiece holder.

- I put a white sheet of paper behind the secondary to remove the confusing reflections of the primary.
- I inserted the silver 2" Catseye Telecat in the eyepiece holder, without tightening it down, and centered the secondary within the sight tube using the reflection of the white paper.
- In order to do that I had to adjust the secondary's rotation and shift it along the axis of the telescope.
- I think that the sight tube that I used can be replaced by a simple 2" cheshire with no crosshair and obtain the same result.
- I removed the white paper.
- Purposely de collimated the primary to get the reflection of the center spot out of the way.
- Replaced the 2" eyepiece holder with the 1 1/4" that came with the scope.
- Used a 1 1/4" reticle eyepiece/cheshire to center the center spot of the primary under the crosshair, using the collimation screws on the secondary. This can also be done with the 2" crosshair tube that I have but my eyes worked better with the 1 1/4" to see the crosshair.
- Center the reflection of the center spot of the primary by using the primary collimation screws in the back.
- Once that is done everything should be nice and concentric, the secondary should show the offset and all is perfectly symmetrical.
- After this process I replaced the corrector.

 

 

Here is a link of the test image after collimation taken with my G3 16200 camera (through the clouds). The top left corner isn't good and I think that was due to focuser sag. 

 

https://www.dropbox....t/test.fit?dl=0

 

I removed the pinion assembly from the focuser. There are two tension screws under it that can be used to tighten down the draw tube and make it more stable. They only need a slight tweak. After that the focuser seemed a lot firmer with less slop. I will have to re-test at the first clear-ish night.

 

After reading your post I went back and checked the reflections with the auto collimator. There was a slight misalignment so I tweaked the tilt screws of the focuser just a couple of tenths of a mm and everything went together and all the reflections overlapped. I don't see the shift that you observe when racking the focuser in and out with the auto collimator installed so I think everything should be OK.

Edited by andysea, 08 September 2019 - 10:36 AM.

[andysea]

One more thing, 

In order to center the secondary under the focuser I had to use the 2" eyepiece holder and 2" tool because the 1 1/4" tool would crop the edge of the secondary making correct centering impossible.

[andysea]

Ugh I just realized that I posted the link to an image binned 2x2. That's not very indicative of defects. Sorry about that. I will have to take some 1x1 test shots when we get a clear night.

[andysea]

Here is the result from the Pixinsight aberration inspector. This is a single 4 second shot with the Moravian G3 16200 binned at 1x1. The FWHM is about 1.9 pixels which translates to about 5.5". Still pretty high but I suspect that it's due to the 6 micron pixel size of the sensor. I haven't been able to improve on this yet. 

 

The uneven illumination is due to thin high clouds....as usual. The black borders are the overscan areas.

Attached Thumbnails

• 

Edited by andysea, 11 September 2019 - 09:28 AM.