Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Software to collimate an SCT using a camera?

  • Please log in to reply
40 replies to this topic

#1 Stelios

Stelios

    Fly Me to the Moon

  • *****
  • Moderators
  • topic starter
  • Posts: 7357
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 26 April 2019 - 01:31 AM

I have had SCT's for many years, and know how to collimate them--with a high power eyepiece, under the stars. 

 

But now I have put an Moonlite focuser on my Edge 8, and my imaging train is a convoluted setup with multiple threaded connections. Call me lazy, but I would rather not waste the time to collimate it with an eyepiece, then remove the visual back, attach the reducer + Moonlite + OAG + EFW + camera and recheck spacing--if I can do it with software. And it does appear that I need to nail the collimation--what looked well-collimated visually produces weird star shapes (see attached)

 

heartstars.JPG

 

Can I? Which software? I think Metaguide is supposed to do it, but the only time I tried MG with my ASI1600MM I couldn't get it started (maybe I should try again, I didn't give it much time). Any other software? I'm not averse to paying (within reason) for something that *works*. 

 

Thanks in advance for any responses. Extra thanks if the software suggested is known to support the ASI1600MM (or use of SGP as camera). 


  • Ptarmigan, dswtan and ks__observer like this

#2 Marcelofig

Marcelofig

    Messenger

  • -----
  • Posts: 424
  • Joined: 21 Jan 2015

Posted 26 April 2019 - 02:06 AM

I think CCD Inspector 2 do it. Personally I haven't used it, I was looking at it for other reasons (tilt correction) and I remember reading it.


Edited by Marcelofig, 26 April 2019 - 02:07 AM.


#3 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 8686
  • Joined: 12 Oct 2007

Posted 26 April 2019 - 02:19 AM

Hi Stelios-

 

MetaGuide could definitely help here - and there have been problems connecting properly with the ASI1600.  But I have a beta that should work well.  I'm happy to let you try it if you would like.  Please PM me if interested.

 

For EdgeHD scopes with the ASI1600, it is possible to do video collimation and at the same time optimize the overall field because the sensor is large enough to show slight distortion of stars at the edges.  So what I typically do is get the center star as good as possible - and then explore edge stars to make sure they are good all around.  And you need to keep checking focus all the time.  Interactive video with a star cluster makes this all easier.

 

Frank


  • Ptarmigan, dswtan, RedLionNJ and 2 others like this

#4 rgsalinger

rgsalinger

    Aurora

  • *****
  • Moderators
  • Posts: 4861
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 26 April 2019 - 02:33 AM

CCDI only works with MaxmDL, if memory serves. Take a look at Gold Focus and see if that's a better fit. With GF, as I recall, you just tell it where the image directory is and it picks up each image and give you collimation information a few seconds later.  

Rgrds-Ross



#5 Tapio

Tapio

    Vanguard

  • -----
  • Posts: 2094
  • Joined: 24 Sep 2006
  • Loc: Tampere, Finland

Posted 26 April 2019 - 05:12 AM

I used tried ASTAP and found out it has CCDI. Haven't tried it for collimation.

But I think video collimation with ASI1600 sounds good idea.



#6 miwitte

miwitte

    Apollo

  • *****
  • Posts: 1094
  • Joined: 28 Dec 2016
  • Loc: Charlotte NC

Posted 26 April 2019 - 09:22 AM

I'm in a similar situation, I'm super new at a SCT, and  collimation is like voodoo to me right now. My first night out with the new(used) scope the stars were all crescents horribly out of collimation. I did get a Hotect collimator with the scope and used that to get where i throught things should be good and when zooming in and defocusing it was still off but i was able to get decent doughnuts after adjusting the secondary adj screws. However the stars in the OAG looked horrible so assumed that collimation was ok in the mddel but not edges where the OAG picks up from. I obtained some Bobs knobs, went through a second iteration of the Hotech but weather is just terrible plus I've been traveling so haven't been able to get out to verify.

 

 

I came across a video and some instruction on using a Hubble optics artificial star and Mire de Collimation reticle

 

https://www.youtube....h?v=8hbiUKWavV0

 

http://sweiller.free...ollimation.html  Looks like Mire de Collimation reticle is used here.

 

 

This method looks pretty good since i can do it when its cloudy out. I've order the artificial star hopefully it comes today so i can work through the process. Metaguide looks pretty good as well but i think you need to be outside with a real star losing valuable time. So my plan was to use the artificial star initailly inside house and that should be pretty darn close, then use Metaguide outside to verify/tweak since its real world. 



#7 spokeshave

spokeshave

    Surveyor 1

  • *****
  • Posts: 1990
  • Joined: 08 Apr 2015

Posted 26 April 2019 - 10:09 AM

CCDI only works with MaxmDL, if memory serves. Take a look at Gold Focus and see if that's a better fit. With GF, as I recall, you just tell it where the image directory is and it picks up each image and give you collimation information a few seconds later.  

Rgrds-Ross

As i understand it, the CCDI real-time collimation tool works natively with MaximDL, but can work with any generic capture software as long as a capture folder is specified. CCDI just monitors the folder for new images and analyzes them when they show up. The image analysis collimation metric will work on any FITS image.

 

i have struggled with fine collimation methods with my EdgeHD 14. Because of the large aperture, collimation by star test is virtually impossible. My seeing just doesn't support it. I may have 2 nights a year when the Airy disk and first ring might be discernible enough to be able to collimate. I have had similar experiences with Metaguide and the Goldfocus system. I can't recommend either - at least with this much aperture. After a rough collimation by centering the secondary shadow on a defocused image and then making the rings concentric with somewhat closer focus, the red spot in Metaguide just dances around randomly. Similarly, the Goldfocus software's collimation suggestions also bounce around randomly as the star dances with the seeing. 

 

The best collimation method I have come up with is to first do a rough collimation on a defocused star. Then use a Duncan mask. I have found the Duncan mask to be very effective and even though the spikes also dance around with the seeing, it is much easier to see what the collimation looks like during the fleeting moments of clarity. All of this can be done with a camera.

 

I have not had good luck with CCDI's collimation analysis metric. It does not seem to readily distinguish the difference between collimation and tilt. It is not uncommon for CCDI to tell me that my collimation is nearly perfect when a simple defocused star proves that it is not. In short, the collimation metric is not accurate - at least for me. One thing I have not tried is the CCDI real-time collimation tool. This uses a stack of defocused star images and analyzes the concentricity of the secondary shadow. In theory, it should work well, but in theory, all of the other systems I have tried (to no real avail) should work well too.

 

Tim


  • dswtan, rgsalinger and RedLionNJ like this

#8 Stelios

Stelios

    Fly Me to the Moon

  • *****
  • Moderators
  • topic starter
  • Posts: 7357
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 26 April 2019 - 11:35 AM

Hi Stelios-

 

MetaGuide could definitely help here - and there have been problems connecting properly with the ASI1600.  But I have a beta that should work well.  I'm happy to let you try it if you would like.  Please PM me if interested.

 

For EdgeHD scopes with the ASI1600, it is possible to do video collimation and at the same time optimize the overall field because the sensor is large enough to show slight distortion of stars at the edges.  So what I typically do is get the center star as good as possible - and then explore edge stars to make sure they are good all around.  And you need to keep checking focus all the time.  Interactive video with a star cluster makes this all easier.

 

Frank

Thanks, Frank. I'll PM you for the beta--obligations and weather mean that I won't be able to try it out till next Friday. 

 

But I'm trying to understand your second paragraph--will it affect ability to collimate if I don't have my reducer completely dialed in? In addition to collimation issues, edge stars show minor elongation. I believe that I need to increase spacing by a mm or a bit less--is this something I would have to do beforehand? As I'll be shooting galaxies and planetaries with the Edge, I usually am not too concerned with edge of field as plenty of crop is available.

 

Here's a super-bright star at the edge of the field, illustrating the combination of reducer and collimation issues:

 

distortstar.JPG



#9 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 8686
  • Joined: 12 Oct 2007

Posted 26 April 2019 - 05:08 PM

Thanks, Frank. I'll PM you for the beta--obligations and weather mean that I won't be able to try it out till next Friday. 

 

But I'm trying to understand your second paragraph--will it affect ability to collimate if I don't have my reducer completely dialed in? In addition to collimation issues, edge stars show minor elongation. I believe that I need to increase spacing by a mm or a bit less--is this something I would have to do beforehand? As I'll be shooting galaxies and planetaries with the Edge, I usually am not too concerned with edge of field as plenty of crop is available.

 

Here's a super-bright star at the edge of the field, illustrating the combination of reducer and collimation issues:

 

 

Hi Stelios-

 

I went ahead and posted the new beta, 5.4.2, on the web site.  You need to unzip it into the full install directory per the instructions - but it should work with the asi1600.  I will add some tips there on how to use it with the 1600.  I'm happy to talk about it here or offline - I defer to your preference.

 

It used to be that most scopes were only well corrected in the center - which means all you needed to do was collimate them so the center was perfect - and you were done.  But with a well corrected scope it's more important that the overall field look as good as possible - and the edge stars may be a more sensitive indicator of alignment.

 

Your initial post goes straight to the point of how different things are today.  For people with cmos cameras that don't need a shutter and can operate at video rates - to me it would be crazy to take down the imaging train in order to use an eyepiece.  Your avoidance of that isn't lazy - it's good sense.

 

Also - with a normal uncorrected SCT, the coma is so severe that a slight miscollimation will show as coma and it is fairly easy to dial things in to make the star round.  But with EdgeHD it is much less sensitive and the edge stars may be a more sensitive indicator of alignment.  This is where it becomes less an issue of "collimation" than overall adjusting things so the field is optimal.

 

Your edge star above definitely has issues - but the key is to compare it to stars on the other side.  If all the edge stars look about the same then you know you are aligned but the spacing may be wrong.  The trick is that you need to make sure you are in focus all the time and the mirror isn't shifting.  If you have a separate focuser then I would make sure the mirror is in a good position by focusing with the primary counter clockwise - then locking it with the mirror locks - all while the scope is pointing up.

 

Doing the alignment with video and a small pixel camera should work well - and MetaGuide should help also.  I would use a red or IR filter if possible - and with decent seeing and a small pixel camera you may see the Airy pattern of the star in the center.  This stuff is much easier nowadays since you can use the actual imaging train - and its filters - for doing everything without tearing it down so you can do things visually.

 

Frank


  • Stelios, dswtan and rgsalinger like this

#10 Stelios

Stelios

    Fly Me to the Moon

  • *****
  • Moderators
  • topic starter
  • Posts: 7357
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 26 April 2019 - 06:56 PM

Thanks, Frank. I'll download the software, and we can continue this offline unless somebody else has questions, of course. 



#11 dswtan

dswtan

    Explorer 1

  • *****
  • Posts: 58
  • Joined: 29 Oct 2006
  • Loc: Morgan Hill, CA

Posted 26 April 2019 - 11:37 PM

@Stelios, @Frank -- we all have questions, and sometimes exactly the same equipment, so please keep going here and not offline. Thanks! :-)


  • SteveInNZ, dmdouglass, darkstar3d and 1 other like this

#12 dmdouglass

dmdouglass

    Surveyor 1

  • *****
  • Posts: 1679
  • Joined: 23 Dec 2007
  • Loc: Tempe, AZ

Posted 27 April 2019 - 12:24 AM

@Stelios, @Frank -- we all have questions, and sometimes exactly the same equipment, so please keep going here and not offline. Thanks! :-)

+1



#13 Stelios

Stelios

    Fly Me to the Moon

  • *****
  • Moderators
  • topic starter
  • Posts: 7357
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 27 April 2019 - 12:32 AM

OK, but as I said, I won't be doing anything with Metaguide (Orthodox Easter, followed by clouds and even rain :() till middle of next week at least. Additional comments most welcome.

 

I do most definitely intend to download the beta and try to collimate my Edge800 with it. I'm also curious to compare Metaguide's performance with PhD2's for guiding (but that's another story).



#14 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 8686
  • Joined: 12 Oct 2007

Posted 04 May 2019 - 10:45 PM

Hi Stelios-

 

If it's ok - and based on interest in this thread - I can talk more about collimation with MetaGuide because I have posted a new version and install on the web site - version 5.4.4.  Free as always.

 

It should work much better with ASI cameras such as the 1600 or 290.

 

Here is an annotated image of what the Airy pattern looks like - in realtime - with EdgeHD11 at f/10 and IR filter - using ASI290.

 

Even using this setup without MetaGuide should allow a reasonably good view of the collimation status - but it would be jumping around and speckily - making it hard to interpret.  But when you can see the first ring clearly - and it is uniform - you know you are collimated.

 

For the ASI1600 it helps to have high video frame rate - and that means using an ROI with a smaller part of the sensor.  This is now possible with MG and is explained in the documentation.

 

For general tips on how to collimate with video - the main thing is simply to note which side of the star looks flared - and simply adjust the collimation screws to move the star in that direction.  Then re-center and check again.  There is no real need to calibrate the view and decide which screw turns which way - because the movement of the star tells you that.

 

But you do need to stay in focus - which may mean focusing periodically - and you need very short exposures with a bright star overhead.  A red or IR filter helps a lot.

 

I'm happy to answer more questions if there is interest - again as long as it's ok with everyone.

 

MGCollimatedWithAnnotationNew.png

 

Thanks,
Frank


  • DesertRat, dswtan, dmdouglass and 4 others like this

#15 dmdouglass

dmdouglass

    Surveyor 1

  • *****
  • Posts: 1679
  • Joined: 23 Dec 2007
  • Loc: Tempe, AZ

Posted 05 May 2019 - 11:29 AM

Well, i can only speak for one person, that being me.  But i sure would like you to continue with your information. You are providing an education in an area of significant interest, evidenced by the number of posts on the subject of collimation of SCTs....  And it is not as if you were going to make money on this deal...  You software is a free download.  



#16 RedLionNJ

RedLionNJ

    Soyuz

  • *****
  • Posts: 3568
  • Joined: 29 Dec 2009
  • Loc: Red Lion, NJ, USA

Posted 05 May 2019 - 11:43 AM

Frank's MetaGuide is an awesome tool for collimation on a real star.  I've had less success using it on an artificial star (likely due to focus changes).

 

David's Hotech SCT collimation tool (the expen$ive one) is also of some utility - that will help you tweak the OTA and focuser to remove any variations from the optical axis due to minor (invisible!) focuser mis-alignment. Unless the focuser is perfectly aligned along the optical axis, and change in focus is essentially also a change in collimation. This is just so not good. You might consider borrowing or renting a Hotech from an other astro-imager or a club. You may only need to use it once or twice to get most of the value out of it.

 

That being said, I'm also with the person above who said they only get a clean Airy ring on maybe one or two nights a year. This undoubtedly makes collimation by ANY means a bit challenging. This is where MetaGuide (awesome job, Frank!) comes into its own with the way it can stack a few frames on the fly. I'm not sure if it can sharpen/deconvolve a bit too (can it, Frank?), but that would likely help in quite a few cases as well. Perhaps a LOT of cases.

 

Finally, with any OTA where the primary mirror is movable, it's helpful to be doing this with the corrector/primary/secondary spacing set to optimal for the particular OTA.  Different OTAs have different optimal back-spacings (distance from rear flange to image plane). The better corrected the overall field is, the less critical the absolute centering of the star image is going to be.  

 

It can be quite tedious to keep precisely re-centering the star every time a minor collimation tweak is made. Using help software to "autoguide" the star back to center can help a lot here, too.



#17 Stelios

Stelios

    Fly Me to the Moon

  • *****
  • Moderators
  • topic starter
  • Posts: 7357
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 05 May 2019 - 11:57 AM

Hi Stelios-

 

If it's ok - and based on interest in this thread - I can talk more about collimation with MetaGuide because I have posted a new version and install on the web site - version 5.4.4.  Free as always.

 

It should work much better with ASI cameras such as the 1600 or 290.

 

Here is an annotated image of what the Airy pattern looks like - in realtime - with EdgeHD11 at f/10 and IR filter - using ASI290.

 

Even using this setup without MetaGuide should allow a reasonably good view of the collimation status - but it would be jumping around and speckily - making it hard to interpret.  But when you can see the first ring clearly - and it is uniform - you know you are collimated.

 

For the ASI1600 it helps to have high video frame rate - and that means using an ROI with a smaller part of the sensor.  This is now possible with MG and is explained in the documentation.

 

For general tips on how to collimate with video - the main thing is simply to note which side of the star looks flared - and simply adjust the collimation screws to move the star in that direction.  Then re-center and check again.  There is no real need to calibrate the view and decide which screw turns which way - because the movement of the star tells you that.

 

But you do need to stay in focus - which may mean focusing periodically - and you need very short exposures with a bright star overhead.  A red or IR filter helps a lot.

 

I'm happy to answer more questions if there is interest - again as long as it's ok with everyone.

 

attachicon.gif MGCollimatedWithAnnotationNew.png

 

Thanks,
Frank

I actually did try this. I didn't find how to do the stacked view. What I had was a small box to the bottom left showing a magnified star, which kept on pulsing and changing shapes like a maniac Transformer on meth. 

Examples (all snips very close to each other, and not even representative of the extremes):
mg1.PNG mg2.PNG mg3.png

 

There was also a red dot running along the periphery of this small image with great rapidity. 

 

I assumed this was due to bad seeing, but I supposed I need to do more of RTFM smile.gif

 

This was slightly off-focus but in-focus was not much different. 
 



#18 jdupton

jdupton

    Surveyor 1

  • *****
  • Posts: 1779
  • Joined: 21 Nov 2010
  • Loc: Central Texas, USA

Posted 05 May 2019 - 01:54 PM

Stelios,

 

   The red dot (Coma Dot) showing the asymmetry direction in the star is only useful at focus. You still have a small donut of a star. Once you focus down to a non-hollow blob or reasonable Airy disk, the dot will help you do the final tweaks to the collimation screws to nail it down.

 

   Frank can expound more on the details.

 

 

John


Edited by jdupton, 05 May 2019 - 01:55 PM.

  • RedLionNJ likes this

#19 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 8686
  • Joined: 12 Oct 2007

Posted 05 May 2019 - 05:00 PM

Hi-

 

Yes the main thing is to make sure the star is in focus and the exposure/gain are way down low so the star isn't saturated.  For a bright star this may mean exposures around 1 ms or less.

 

The images Stelios shows are far from focus - and as a result the stacking won't work at all.  Normal star stacking is based on a center of gravity centroid - but MG uses something completely different to handle speckily star spots - and it won't work at all if the star has a bit of a donut shape.

 

In the latest release you change the amount of star stacking with the "StackTime" setting - which typically is around 1 second.  The video camera may be able to operate at 10-30 frames per second - and the combination of stacktime with frame rate tells you how many frames will be stacked.

 

Once the star is in focus you should get a steady looking star spot in the lower right window - and the coma dot will hang out in one direction and tell you which way to make adjustments.  Usually you want to make adjustments so the star moves in the direction of the dot - but it could be the other way.  Try one way and see if it gets better or worse - after re-centering the star.

 

And for automatic re-centering - connect up the mount for guiding and guide on the star.

 

Make sure the star is in focus, the exposure is short, and there is no red bar appearing over the star image in the lower right - which means some of the pixels are saturated and you need to reduce the exposure.

 

For people with the ASI1600 you can get faster frame rate by choosing a smaller screen size and then make sure the ROI is in the center of the sensor.  See the docs for details.

Frank


Edited by freestar8n, 05 May 2019 - 05:00 PM.


#20 Lorenz0x7BC

Lorenz0x7BC

    Vostok 1

  • -----
  • Posts: 134
  • Joined: 03 Apr 2019
  • Loc: Vienna Austria Europe

Posted 24 July 2019 - 02:13 PM

And for automatic re-centering - connect up the mount for guiding and guide on the star.

Hi Frank,

 

I was trying to do this (connecting to my mount in MetaGuide->Setup->ASCOM Scope Chooser->CPWI so that automatic re-centering is enabled) but every time I choose the ASCOM CPWI driver MetaGuide crashes after a few seconds.

 

CPWI version 2.1.25 connected by USB to the Celestron Evolution mount (hand control). 

MetaGuide version 5.4.4 (downloaded today)

 

In SharpCap mount control through CPWI ASCOM works fine.

 

Thanks for your help,

Lorenz



#21 cfosterstars

cfosterstars

    Vanguard

  • *****
  • Posts: 2443
  • Joined: 05 Sep 2014
  • Loc: Austin, Texas

Posted 24 July 2019 - 04:04 PM

CCDI only works with MaxmDL, if memory serves. Take a look at Gold Focus and see if that's a better fit. With GF, as I recall, you just tell it where the image directory is and it picks up each image and give you collimation information a few seconds later.  

Rgrds-Ross

CCDI is not that easy to use. I have tried it several times. It will work with SGP. I dont recommend it though. I used defocused stars on the camera and it works "OK". I have been wanting to try out Metaguiding to see about an optimization method with software also. I just have not had the time to dig into it. It on my to do list, but dose not seem to bubble to the top. With this thread and the new version, I need to prioritize this task.

 

Frank,

 

Thanks as always. This is really good stuff.


Edited by cfosterstars, 24 July 2019 - 04:15 PM.


#22 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 8686
  • Joined: 12 Oct 2007

Posted 24 July 2019 - 05:56 PM

Hi Frank,

 

I was trying to do this (connecting to my mount in MetaGuide->Setup->ASCOM Scope Chooser->CPWI so that automatic re-centering is enabled) but every time I choose the ASCOM CPWI driver MetaGuide crashes after a few seconds.

 

CPWI version 2.1.25 connected by USB to the Celestron Evolution mount (hand control). 

MetaGuide version 5.4.4 (downloaded today)

 

In SharpCap mount control through CPWI ASCOM works fine.

 

Thanks for your help,

Lorenz

Hi-

 

Sorry about your troubles.  I have been using CPWI for some time and it connects ok with MG 5.4.4.  Do you have the mount connected to cpwi when you start MG?  Are you able to connect other devices via ascom to the mount?

 

If you continue to have troubles please send me the MGDebug.log file after mg crashes.

 

Thanks,
Frank



#23 Stelios

Stelios

    Fly Me to the Moon

  • *****
  • Moderators
  • topic starter
  • Posts: 7357
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 24 July 2019 - 10:11 PM

Mine was also crashing when I tried to guide (immediate crash, repeatable, app just exits). However before then I was able to see that my collimation was almost perfect, and so I went with that (seeing was not good enough to bother adjusting).



#24 Ptarmigan

Ptarmigan

    Lagopus lagopus

  • *****
  • Posts: 4619
  • Joined: 23 Sep 2004
  • Loc: Arctic

Posted 24 July 2019 - 11:09 PM

How often do people collimate their SCT?

 

I have not collimated mines since I got it a couple years ago.



#25 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 8686
  • Joined: 12 Oct 2007

Posted 24 July 2019 - 11:36 PM

Mine was also crashing when I tried to guide (immediate crash, repeatable, app just exits). However before then I was able to see that my collimation was almost perfect, and so I went with that (seeing was not good enough to bother adjusting).

Sorry about the trouble, guys.  Looks like I broke something.  It does work but there must be something specific about different setups that is causing it to fail.  I'll look into it.

 

Frank




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics