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

AI based wave front sensing and collimation

Collimation
  • Please log in to reply
649 replies to this topic

#426 Corsica

Corsica

    Vendor (Innovationsforesight)

  • *****
  • Vendors
  • topic starter
  • Posts: 1,206
  • Joined: 01 Mar 2010
  • Loc: Columbus Indiana - USA

Posted 29 April 2024 - 10:53 AM

In a few weeks, we'll release a new SKW update that will add some exciting features.

 

-1-    Balanced astigmatism score using a star field.
SKW computes automatically the level of balance even if the stars in the field are not at a constant distance from the chip center, as long as there are enough stars spread well over the field.

 

-2-    Sensor/image tilt score using a star field.
This SKW score is based on the scope's CFZ.

 

This update will be available free of charge for current and future SKW Collimator and Pro users.


  • taraobservatory, mikela and DaveDE like this

#427 Rasfahan

Rasfahan

    Soyuz

  • -----
  • Posts: 3,612
  • Joined: 12 May 2020
  • Loc: Hessen, Germany

Posted 29 April 2024 - 11:05 AM

What an awesome update! I had hoped for something like this to become available.

#428 arbit

arbit

    Apollo

  • -----
  • Posts: 1,407
  • Joined: 19 Feb 2012

Posted 29 April 2024 - 11:59 AM

In a few weeks, we'll release a new SKW update that will add some exciting features.

-1- Balanced astigmatism score using a star field.
SKW computes automatically the level of balance even if the stars in the field are not at a constant distance from the chip center, as long as there are enough stars spread well over the field.

-2- Sensor/image tilt score using a star field.
This SKW score is based on the scope's CFZ.

This update will be available free of charge for current and future SKW Collimator and Pro users.

Great features. Looking forward to them.

Sent from my SM-S908E using Tapatalk

#429 RonaldNC

RonaldNC

    Viking 1

  • *****
  • Posts: 758
  • Joined: 25 Jun 2019
  • Loc: Fuquay Varina, NC

Posted 29 April 2024 - 03:41 PM

Gaston,

 

This is great!

 

So, currently I get a 9.8 score on collimating my Hyperstar on my Celestron HD Edge 8"... but I still get crescent stars on the top part of the sensor/image.  Can I use this new feature to adjust the tilt on my ASI2600MC Pro?

 

Thanks,

Ron



#430 taraobservatory

taraobservatory

    Vostok 1

  • *****
  • Posts: 151
  • Joined: 19 Jul 2020

Posted 04 May 2024 - 02:30 AM

Hi Paul, 

 

 

Thank you :) 



#431 Object_FS

Object_FS

    Vostok 1

  • -----
  • Posts: 100
  • Joined: 16 Nov 2017

Posted 10 May 2024 - 01:46 PM

In a few weeks, we'll release a new SKW update that will add some exciting features.

 

-1-    Balanced astigmatism score using a star field.
SKW computes automatically the level of balance even if the stars in the field are not at a constant distance from the chip center, as long as there are enough stars spread well over the field.

 

-2-    Sensor/image tilt score using a star field.
This SKW score is based on the scope's CFZ.

 

This update will be available free of charge for current and future SKW Collimator and Pro users.

Very interesting features Gaston laugh.gif

 

Will you update the SkyWave documentation as well, and add the detailed information about these new features?


Edited by Object_FS, 10 May 2024 - 01:46 PM.


#432 KTAZ

KTAZ

    Aurora

  • *****
  • Posts: 4,725
  • Joined: 09 Apr 2020
  • Loc: Scottsdale, AZ

Posted 19 May 2024 - 10:06 PM

This weekend I used SkyWave to collimate my 9.25" Edge. I ended up having to do trial and error at first in order to get a defocused star of the size and quality needed, but I was able to get everything dialed in eventually.

 

Very happy with the software. Many would say it is a bit pricey; however, I found it to be the best that I've used from the perspective of providing very clear analysis and tracking of your progress. I think I burned about a dozen credits; but 2/3 of those were burned after I foolishly decided an 8.9 score wasn't good enough. I ping-ponged back and forth for over an hour until I got, you guessed it, an 8.9.

 

I did gain a fresh appreciation for just how sensitive those collimation screws are. Seriously, the last adjustment that took me literally from an orange zone to the green was literally 1/32 turn. I barely knew that I turned it; I could only be sure by the slight movement on my loop.



#433 RonaldNC

RonaldNC

    Viking 1

  • *****
  • Posts: 758
  • Joined: 25 Jun 2019
  • Loc: Fuquay Varina, NC

Posted 20 May 2024 - 01:57 AM

This weekend I used SkyWave to collimate my 9.25" Edge. I ended up having to do trial and error at first in order to get a defocused star of the size and quality needed, but I was able to get everything dialed in eventually.

 

Very happy with the software. Many would say it is a bit pricey; however, I found it to be the best that I've used from the perspective of providing very clear analysis and tracking of your progress. I think I burned about a dozen credits; but 2/3 of those were burned after I foolishly decided an 8.9 score wasn't good enough. I ping-ponged back and forth for over an hour until I got, you guessed it, an 8.9.

 

I did gain a fresh appreciation for just how sensitive those collimation screws are. Seriously, the last adjustment that took me literally from an orange zone to the green was literally 1/32 turn. I barely knew that I turned it; I could only be sure by the slight movement on my loop.

I agree with your experience.  It took a few trials and errors to get the image just right... orientation, exposure, focus setting, etc., but now it's pretty easy.  I was surprised at how a very small change makes such a big difference... particularly with my HyperStar.  That thing is very, very sensitive.

 

Now I'm wrestling with another problem with the HyperStar.  The center stars are very good, but the out-lying stars... not so much.  I'm assuming it's sensor tilt, but who knows.  I'm hoping that Gaston's new version will help me determine the issue.

 

Ron


  • KTAZ likes this

#434 KTAZ

KTAZ

    Aurora

  • *****
  • Posts: 4,725
  • Joined: 09 Apr 2020
  • Loc: Scottsdale, AZ

Posted 20 May 2024 - 10:33 AM

I agree with your experience.  It took a few trials and errors to get the image just right... orientation, exposure, focus setting, etc., but now it's pretty easy.  I was surprised at how a very small change makes such a big difference... particularly with my HyperStar.  That thing is very, very sensitive.

 

Now I'm wrestling with another problem with the HyperStar.  The center stars are very good, but the out-lying stars... not so much.  I'm assuming it's sensor tilt, but who knows.  I'm hoping that Gaston's new version will help me determine the issue.

 

Ron

Yes, tilt will be my next focus. I already have a CTU in the imaging train, and had adjusted it before, but Gaston is right that we should get the collimation nailed before moving on the tilt adjustments.

 

However, backfocus is the second step after the collimation. In particular, the Celestron Edge HD reducers will impart some CA to stars if this isn't dead on. I will then move on to tilt. Then I plan to go in reverse order; verify tilt is good, check the back focus one more time, then check the collimation one more time.


  • RonaldNC likes this

#435 RonaldNC

RonaldNC

    Viking 1

  • *****
  • Posts: 758
  • Joined: 25 Jun 2019
  • Loc: Fuquay Varina, NC

Posted 20 May 2024 - 01:34 PM

Yes, tilt will be my next focus. I already have a CTU in the imaging train, and had adjusted it before, but Gaston is right that we should get the collimation nailed before moving on the tilt adjustments.

 

However, backfocus is the second step after the collimation. In particular, the Celestron Edge HD reducers will impart some CA to stars if this isn't dead on. I will then move on to tilt. Then I plan to go in reverse order; verify tilt is good, check the back focus one more time, then check the collimation one more time.

Not to derail this thread too much... but using "correct only" with BlurXterminator in PixInsight as a first step is a game changer.  I am amazed how it corrects deformed stars.

 

Ron



#436 KTAZ

KTAZ

    Aurora

  • *****
  • Posts: 4,725
  • Joined: 09 Apr 2020
  • Loc: Scottsdale, AZ

Posted 20 May 2024 - 01:39 PM

Not to derail this thread too much... but using "correct only" with BlurXterminator in PixInsight as a first step is a game changer.  I am amazed how it corrects deformed stars.

 

Ron

I'm waiting on Russ to release "CollimationXterminator"...

 

Then we're all set...lol.gif


  • RonaldNC likes this

#437 KTAZ

KTAZ

    Aurora

  • *****
  • Posts: 4,725
  • Joined: 09 Apr 2020
  • Loc: Scottsdale, AZ

Posted 17 June 2024 - 02:16 PM

I made the investment in Skywave and have some issues ATM. I might be using it incorrectly or folowing the wrong steps.

 

Initially, I spent 2 hours tweaking my 9.25 Edge SCT with .70 reducer. I ended up with a score around 8.9 and said "good enough". Unfortunately, my star shapes are elongated. Double checked my balance, PA, tilt, etc., and all were dialed in very tight. Tracking was around .50 in PHD2.

 

I made a second attempt and cannot even get it close now. I am able to get the coma adjusted to a 10, but that makes the astig drop into the red zone. If I get the astig close, maybe into the yellow zone, the coma goes back out of the green zone. It is like ping pong with adjusting for the astig; it bounces from one side of the red ring to the other with only minor adjustments.

 

HELP. Any tips or suggestions would be appreciated.



#438 KTAZ

KTAZ

    Aurora

  • *****
  • Posts: 4,725
  • Joined: 09 Apr 2020
  • Loc: Scottsdale, AZ

Posted 18 June 2024 - 11:56 AM

I made the investment in Skywave and have some issues ATM. I might be using it incorrectly or folowing the wrong steps.

 

Initially, I spent 2 hours tweaking my 9.25 Edge SCT with .70 reducer. I ended up with a score around 8.9 and said "good enough". Unfortunately, my star shapes are elongated. Double checked my balance, PA, tilt, etc., and all were dialed in very tight. Tracking was around .50 in PHD2.

 

I made a second attempt and cannot even get it close now. I am able to get the coma adjusted to a 10, but that makes the astig drop into the red zone. If I get the astig close, maybe into the yellow zone, the coma goes back out of the green zone. It is like ping pong with adjusting for the astig; it bounces from one side of the red ring to the other with only minor adjustments.

 

HELP. Any tips or suggestions would be appreciated.

Ok, I went back through the training videos and from what I can ascertain ONLY the coma is important when adjusting an SCT. If spherical is also OK, then the astig is just what is is.

 

Am I correct in this?



#439 Corsica

Corsica

    Vendor (Innovationsforesight)

  • *****
  • Vendors
  • topic starter
  • Posts: 1,206
  • Joined: 01 Mar 2010
  • Loc: Columbus Indiana - USA

Posted 19 June 2024 - 02:48 PM

Ok, I went back through the training videos and from what I can ascertain ONLY the coma is important when adjusting an SCT. If spherical is also OK, then the astig is just what is is.

 

Am I correct in this?

You are correct that SCTs EdgeHD have a spherical secondary mirror. For alignment (collimation) of both mirrors, all you need to monitor is coma for an on-axis (at the center of the camera's chip) defocused star.

Here are a few points to consider:

 

1. Ensure that you do not clip (saturate) the image. Any hot/dead pixels are automatically handled by the SKW pre-processing. We recommend aiming for half to 2/3 of the maximum ADU using at least a 16-bit FITS file format (float format is fine). All the images must be monochrome FITS files.

 

2. Due to seeing conditions, you may want to use a long enough exposure time, at least 10 seconds or so. You could use longer exposures at the end of the collimation for higher accuracy. SKW's engine is a wavefront sensor, which means that it will measure the wavefront distortion from the optics and the atmospheric turbulence. The latter contribution decreases as integration time gets longer.

 

3. When coma is canceled on-axis, you should have a minimum of astigmatism, at least inside the corrected circle of the scope (where you are diffraction-limited) and none on-axis. If you see some astigmatism, there are two likely explanations:

 

-a- Tracking drift during the exposure: Keep in mind that the diffraction limit is around 0.075 wave RMS, which is not that much. At that level of wavefront error, you will not stop the defocused star elongation by just human inspection; this is too little to see. This also means that very little drift leading to an elongated star (not from any aberration) may be reported as astigmatism. If exposures of 10 seconds or more are an issue, you could use shorter exposures and either stack time (using an auto-correlation algorithm, not centroid) or just look on SKW at the center of the score cluster (blue dot) to estimate the average astigmatism value (enable the scatter plot option in SKW collimator tool).
  
-b- Astigmatism off-axis is expected but not on-axis: If tracking drift is not the source, you may have some tilt in the optical train, or even at the level of the field lens (corrector). I would assume that the optical surfaces themselves are near perfect.


  • KTAZ likes this

#440 KTAZ

KTAZ

    Aurora

  • *****
  • Posts: 4,725
  • Joined: 09 Apr 2020
  • Loc: Scottsdale, AZ

Posted 19 June 2024 - 03:15 PM

Thanks much, Gaston. I do see some tilt that is showing up after the collimation, so that is very likely the source.

 

I do appreciate the feedback!



#441 RonaldNC

RonaldNC

    Viking 1

  • *****
  • Posts: 758
  • Joined: 25 Jun 2019
  • Loc: Fuquay Varina, NC

Posted 22 June 2024 - 03:16 PM

In a few weeks, we'll release a new SKW update that will add some exciting features.

 

-1-    Balanced astigmatism score using a star field.
SKW computes automatically the level of balance even if the stars in the field are not at a constant distance from the chip center, as long as there are enough stars spread well over the field.

 

-2-    Sensor/image tilt score using a star field.
This SKW score is based on the scope's CFZ.

 

This update will be available free of charge for current and future SKW Collimator and Pro users.

Gaston,

 

Any update on the update?  I'm looking forward to trying the new features!

 

Thanks,

Ron



#442 Corsica

Corsica

    Vendor (Innovationsforesight)

  • *****
  • Vendors
  • topic starter
  • Posts: 1,206
  • Joined: 01 Mar 2010
  • Loc: Columbus Indiana - USA

Posted 29 June 2024 - 04:05 PM

There is a new version of SKW 5.9 available now, featuring new options for astigmatism-balanced and sensor tilt in the collimator tool. However, their help documentation is not yet updated.

 

To update to the new version SKW, you MUST FIRST uninstall the current version using the Windows OS uninstall tool.

The first time you run the new SKW version, it should automatically find your license key. If not, you can provide the license key file.

DO NOT register a new license since this will render your model unusable.

If you use the same working directory folder, all your settings and models will be automatically recovered.

 

In order to use these new features, you need several defocused stars in the image. For optimal results in correcting astigmatism and tilt, I recommend having 20 to 30 stars scattered across the field.

There is a star distribution quality flag that should ideally be green. It automatically assesses how well the stars are spread across the field and how many are usable for the calculation. Below the first viewport is the raw image, the second is the astigmatism-balanced collimator tool viewport, and the last one (bottom) is the tilt tool.

NewSKWCollimatorTool.jpg

They all provide similar collimation tool aspects already in SKW, with the same user interface and a target.

The astigmatism-balanced tool shows a rotating blue cursor at the location where there is the maximum departure of the absolute total astigmatism from the center of the field. You want to bring the % balance (which replaces the score value) as close to zero as possible, at least into the green zone.

The tilt tool does the same kind of thing. The blue cursor is located where there is a maximum of positive tilt. The deviation shown at the bottom (replacing the score value) is related to the scope CFZ. You want to be below 100%, also in the green zone.



#443 Meies

Meies

    Viking 1

  • *****
  • Posts: 674
  • Joined: 05 Jun 2016
  • Loc: Central Virginia

Posted 29 June 2024 - 06:12 PM

Do the stars need to be equally bright? Should they be defocussed by the same amount as the star used for coma?



#444 Corsica

Corsica

    Vendor (Innovationsforesight)

  • *****
  • Vendors
  • topic starter
  • Posts: 1,206
  • Joined: 01 Mar 2010
  • Loc: Columbus Indiana - USA

Posted 01 July 2024 - 12:27 PM

Do the stars need to be equally bright? Should they be defocussed by the same amount as the star used for coma?

The stars don't need to have the same signal level. SKW automatically selects stars that meet basic criteria, such as no saturation and sufficient SNR. It also performs a statistical analysis of the wavefronts of the selected stars and rejects outliers.

 

The amount of defocus is identical to the measurement of coma or any other aberration in SKW; there is no difference. SKW performs a full wavefront measurement of each usable star and processes the data to build field-dependent aberration maps, which are used to display balanced astigmatism and sensor tilt.

It's important to understand that the tilt is computed using the defocus (the linear component of the field curvature) term of the Zernike polynomial wavefront decomposition, not the size of the star (FWHM or HF), as these values are affected by all aberrations, not just defocus.

 

Adjustments to the image tilt at the optical train level, after the scope's visual back, should be made only when the scope is optically collimated. A miscollimated scope exhibits a tilted focal plane, which can be mistaken for actual sensor mechanical tilt.


Edited by Corsica, 01 July 2024 - 12:37 PM.


#445 Dark_Knight18

Dark_Knight18

    Vostok 1

  • *****
  • Posts: 124
  • Joined: 31 Jan 2020
  • Loc: Baker City, Oregon 97814

Posted 03 July 2024 - 09:29 AM

Hi Gaston-

 

I had sent a personal email regarding credits I had transferred to a different computer. I was wondering if you got my email?

 

Also, as a suggestion, is it possible that you and your team could develop a training video for your software?

 

I think it would be very helpful with learning how to use your excellent program. 

 

Kind regards, 

 

Brian



#446 Corsica

Corsica

    Vendor (Innovationsforesight)

  • *****
  • Vendors
  • topic starter
  • Posts: 1,206
  • Joined: 01 Mar 2010
  • Loc: Columbus Indiana - USA

Posted 03 July 2024 - 11:16 AM

Hi Gaston-

 

I had sent a personal email regarding credits I had transferred to a different computer. I was wondering if you got my email?

 

Also, as a suggestion, is it possible that you and your team could develop a training video for your software?

 

I think it would be very helpful with learning how to use your excellent program. 

 

Kind regards, 

 

Brian

Dear Brian,

 

I replied to your email. Perhaps it went to your SPAM folder.

 

P.S. I am traveling overseas. I monitor my emails, but there may be some delays before I can process and reply to them.



#447 DaveDE

DaveDE

    Mariner 2

  • -----
  • Posts: 206
  • Joined: 08 Oct 2021
  • Loc: Longmont Colorado

Posted 04 July 2024 - 09:25 AM

I would like to perform  the version 5.9 upgrade but want to ensure that the license key file is saved and safe in case something goes wrong. Where is this license key file? What is its name and file extension. I've searched and cannot seem to find anything that looks like a license file.  Thanks - Dave



#448 Meies

Meies

    Viking 1

  • *****
  • Posts: 674
  • Joined: 05 Jun 2016
  • Loc: Central Virginia

Posted 04 July 2024 - 12:42 PM

I would like to perform  the version 5.9 upgrade but want to ensure that the license key file is saved and safe in case something goes wrong. Where is this license key file?

I just updated with no problem. Your file locations may be different from mine. I have a folder called Skywave64 with subfolders for each version, where the executable and instruction files live. A new folder gets created in Skywave64 when you uninstall the previous version and install the new one. In addition, there is a Skywave folder on my desktop where the working files, including my paid model, are located. In my case, this folder was unchanged with the new install. Skywave started up without a problem and located my license and model. I haven't tried it out on a new star image file, but it works on my old ones.



#449 DaveDE

DaveDE

    Mariner 2

  • -----
  • Posts: 206
  • Joined: 08 Oct 2021
  • Loc: Longmont Colorado

Posted 04 July 2024 - 04:58 PM

Thanks, got it working.



#450 Eyecon

Eyecon

    Explorer 1

  • -----
  • Posts: 63
  • Joined: 12 Jul 2023
  • Loc: Ontario, Canada

Posted 09 July 2024 - 01:04 PM

Just installed SKW 5.9 and looking forward to a clear night to finally fix my field aberration issues (I hope!). I'm using SKW on an F4 newtonian that has been challenging to get tuned for good star shapes in the corners of an APS-C sensor. Is anyone else using SKW with a newt?




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





Also tagged with one or more of these keywords: Collimation



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics