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

Building 10Micron models and the things that affect models

  • Please log in to reply
160 replies to this topic

#76 mworion

mworion

    Explorer 1

  • -----
  • Posts: 59
  • Joined: 08 Mar 2016
  • Loc: Hohenbrunn / Munich

Posted 12 February 2021 - 11:16 AM

Arie, Andrew,

I will check this tonight (if weather is OK). Basically the slew position with selected star in MW4 and selected star in Handpad should be exactly the same as the command protocol implemented by 10micron as main goal. So far I had not much trouble in my setup, but we will see.

I've been updating the hipparcos catalogue (which should be not any issue as the catalogue itself was finished in 1991). But there might be some updates in catalogue data in the firmware done. We'll see.

Michel


  • psandelle and Lead_Weight like this

#77 mworion

mworion

    Explorer 1

  • -----
  • Posts: 59
  • Joined: 08 Mar 2016
  • Loc: Hohenbrunn / Munich

Posted 12 February 2021 - 01:20 PM

Hi, 

After moving through all parts, hiparcos catalogue and all the conversions. I found the problem. So shame over me, I do a normal slew to the selected star, not a slew for correction. All numbers were OK, just the command was wrong. This was introduced when adding a feature of keeping tracking state when slewing to a star. This will be corrected in 1.0.2 and gives now correct results.

I add two images where in the first I used the handpad to find Betelgeuse for Polar Alignment and secondly, where I chose the star out of the hemisphere window.

Both slew the mount to the correct and identical position.

Michel 

 

Bildschirmfoto 2021-02-12 um 19.03.22.jpg

 

Bildschirmfoto 2021-02-12 um 19.03.47.jpg


  • Tonk, lambermo, psandelle and 2 others like this

#78 mworion

mworion

    Explorer 1

  • -----
  • Posts: 59
  • Joined: 08 Mar 2016
  • Loc: Hohenbrunn / Munich

Posted 12 February 2021 - 01:59 PM

V1.0.2 Bugfix is now available.


  • psandelle likes this

#79 SimonIRE

SimonIRE

    Surveyor 1

  • -----
  • Posts: 1,558
  • Joined: 16 Dec 2018
  • Loc: South East England

Posted 12 February 2021 - 02:02 PM

V1.0.2 Bugfix is now available.

Now that’s service...:)

Incidentally, does MW4 support QHY cameras (via ASCOM)?

Edited by SimonIRE, 12 February 2021 - 02:03 PM.

  • Tonk likes this

#80 Lead_Weight

Lead_Weight

    Vanguard

  • *****
  • Posts: 2,262
  • Joined: 21 Jul 2016
  • Loc: Houston

Posted 12 February 2021 - 02:06 PM

V1.0.2 Bugfix is now available.

Thanks! I'll be doing PA this way from now on. lol.gif



#81 Tonk

Tonk

    Cosmos

  • *****
  • topic starter
  • Posts: 9,527
  • Joined: 19 Aug 2004
  • Loc: Leeds, UK, 54N

Posted 12 February 2021 - 02:27 PM

I found the problem. So shame over me, I do a normal slew to the selected star, not a slew for correction. All numbers were OK, just the command was wrong. This was introduced when adding a feature of keeping tracking state when slewing to a star.


Michel - do I detect a refactoring gaffe? Do you use unit tests (e.g. PyUnit) and continuous integration (e.g. Jenkins) to automatically check expected software behaviour after a refactor?
  • mworion likes this

#82 mworion

mworion

    Explorer 1

  • -----
  • Posts: 59
  • Joined: 08 Mar 2016
  • Loc: Hohenbrunn / Munich

Posted 12 February 2021 - 03:29 PM

Hi Tonk,

Good question. MW4 has about 3600+ unit tests overall (for sure different quality) and the build is on GitHub automated for all platforms (Windows, macOS and Ubuntu). Still as It is not a full professional work, the integration and automated validation tests are missing. I though about it, but that's too much work to do.

Still the challenge would be to go for new part to a TDD solution, but that's also tough.

I the case I added an additional parameter, which was tested in unit tests correctly. But the check if (depending on the user input: just slew or polar align) this parameter is passed correctly can't be detected in the existing unit test as the relation is from context setting (operation mode) to GUI (user behavior) to the functional part (choosing the right command protocol.

But I was really boogey about this error.

Michel


  • lambermo likes this

#83 Tonk

Tonk

    Cosmos

  • *****
  • topic starter
  • Posts: 9,527
  • Joined: 19 Aug 2004
  • Loc: Leeds, UK, 54N

Posted 12 February 2021 - 05:58 PM

MW4 has about 3600+ unit tests overall (for sure different quality) and the build is on GitHub automated for all platforms (Windows, macOS and Ubuntu)


wow - excellent. I shall be sticking with MW then. I used to teach unit testing and CI 20 years ago when it was all new and it was hard to get development teams to understand the benefits :)
  • psandelle and Raginar like this

#84 SimonIRE

SimonIRE

    Surveyor 1

  • -----
  • Posts: 1,558
  • Joined: 16 Dec 2018
  • Loc: South East England

Posted 18 February 2021 - 05:19 AM

Hello folks,

 

Quick question regarding the use of Voyager and your 10 Micron (with a model built). 

 

I have forgotten the settings I had for the mount driver under synch. 

 

I am interested to know if you have:

 

"enable synch" unchecked

 

but

 

"use synch as refine" checked - and then in Voyager  have

 

"not synch mount on solved point" checked?

 

In other words, do you let Voyager do a synch align or do you prevent Voyager from messing with your model at all? I am pretty sure that in the past I have not let Voyager mess with the model at all. 

 

Best, 

Simon


Edited by SimonIRE, 18 February 2021 - 05:19 AM.


#85 Tonk

Tonk

    Cosmos

  • *****
  • topic starter
  • Posts: 9,527
  • Joined: 19 Aug 2004
  • Loc: Leeds, UK, 54N

Posted 18 February 2021 - 05:44 AM

"use synch as refine"


Quite simple - never ever set sync refine unless you are actually building a model. Outside of a model building context this is the easiest way to degrade and destroy an existing model.

Which is why I strongly recommend that when you have built a model, that you *save* it as a named model. That way if you do accidentally corrupt the current model you can recover by reloading the saved copy of the original.

 

If you have multiple scopes on your mount then saved models are the pre-requisite step to running a session as you have to load the model for the scope you are using


Edited by Tonk, 18 February 2021 - 01:21 PM.

  • EFT and psandelle like this

#86 SimonIRE

SimonIRE

    Surveyor 1

  • -----
  • Posts: 1,558
  • Joined: 16 Dec 2018
  • Loc: South East England

Posted 18 February 2021 - 05:45 AM

Yep - this is what I thought. Cheers. 


  • psandelle likes this

#87 SimonIRE

SimonIRE

    Surveyor 1

  • -----
  • Posts: 1,558
  • Joined: 16 Dec 2018
  • Loc: South East England

Posted 18 February 2021 - 05:59 PM

Another 2 quick questions; I am trying to switch over to MW4.

 

1. I have downloaded the H17 files into my ASTAP folder but I can't get the path for the index files in MW4 to work with this. It does work with the G17 files - what is my mistake here or is this expected?

 

2. Does MW4 work with QHY cameras?

 

Cheers,

Simon



#88 SLarkin

SLarkin

    Explorer 1

  • -----
  • Posts: 59
  • Joined: 03 Dec 2012
  • Loc: AZ & MN

Posted 18 February 2021 - 06:01 PM

Delete the G17 files. In Windows I did it thru uninstall.


Edited by SLarkin, 18 February 2021 - 06:22 PM.


#89 yzhzhang

yzhzhang

    Apollo

  • *****
  • Posts: 1,077
  • Joined: 13 Dec 2012
  • Loc: SoCal

Posted 18 February 2021 - 06:05 PM

Hello folks,

 

Quick question regarding the use of Voyager and your 10 Micron (with a model built). 

 

I have forgotten the settings I had for the mount driver under synch. 

 

I am interested to know if you have:

 

"enable synch" unchecked

 

but

 

"use synch as refine" checked - and then in Voyager  have

 

"not synch mount on solved point" checked?

 

In other words, do you let Voyager do a synch align or do you prevent Voyager from messing with your model at all? I am pretty sure that in the past I have not let Voyager mess with the model at all. 

 

Best, 

Simon

I think I actually have both "enable sync" and "use sync as refine" checked, but "disable sync mount on solved point" in Voyager. It's been working fine for me.



#90 RazvanUnderStars

RazvanUnderStars

    Apollo

  • *****
  • Posts: 1,321
  • Joined: 15 Jul 2014
  • Loc: Toronto, Canada

Posted 18 February 2021 - 06:37 PM

Regarding #1, I reported the issue to Michel (MW4's author), he'll fix it soon. In the meantime, you can fix it locally by editing MountWizard4\venv\Lib\site-packages\mw4\logic\astrometry\astrometryASTAP.py and replacing 290 with 1476 in the 3 instances in the following code block (290 is the file extension for the G-series of ASTAP index files; 1476 is the file extension for the H-series. The number refers to the file count).

 

        if platform.system() == 'Darwin':
            program = self.appPath + '/astap'
            index = self.indexPath + '/*.290'
        elif platform.system() == 'Linux':
            program = self.appPath + '/astap'
            index = self.indexPath + '/*.290'
        elif platform.system() == 'Windows':
            program = self.appPath + '/astap.exe'
            index = self.indexPath + '/*.290'

 

 

Regarding #2, MW4 works with the cameras through ASCOM so it should work as long as you have the drivers installed.

 

 

Another 2 quick questions; I am trying to switch over to MW4.

 

1. I have downloaded the H17 files into my ASTAP folder but I can't get the path for the index files in MW4 to work with this. It does work with the G17 files - what is my mistake here or is this expected?

 

2. Does MW4 work with QHY cameras?

 

Cheers,

Simon



#91 SimonIRE

SimonIRE

    Surveyor 1

  • -----
  • Posts: 1,558
  • Joined: 16 Dec 2018
  • Loc: South East England

Posted 18 February 2021 - 06:53 PM

Regarding #1, I reported the issue to Michel (MW4's author), he'll fix it soon. In the meantime, you can fix it locally by editing MountWizard4\venv\Lib\site-packages\mw4\logic\astrometry\astrometryASTAP.py and replacing 290 with 1476 in the 3 instances in the following code block (290 is the file extension for the G-series of ASTAP index files; 1476 is the file extension for the H-series. The number refers to the file count).

 

        if platform.system() == 'Darwin':
            program = self.appPath + '/astap'
            index = self.indexPath + '/*.290'
        elif platform.system() == 'Linux':
            program = self.appPath + '/astap'
            index = self.indexPath + '/*.290'
        elif platform.system() == 'Windows':
            program = self.appPath + '/astap.exe'
            index = self.indexPath + '/*.290'

 

 

Regarding #2, MW4 works with the cameras through ASCOM so it should work as long as you have the drivers installed.

 

 

This is great; it works. 

 

But for some reason, I can't get any image to solve. It keeps failing. Usually this is because I have entered the incorrect scope parameters but these are all correct. Everything connects without difficulty, and the images expose, but they fail to solve...



#92 RazvanUnderStars

RazvanUnderStars

    Apollo

  • *****
  • Posts: 1,321
  • Joined: 15 Jul 2014
  • Loc: Toronto, Canada

Posted 18 February 2021 - 06:56 PM

Do they solve if you try directly in ASTAP (even as a blind solve)? You may also want to check what gets written in the FITS header of those images. If it still fails, you can ask HAN in the thread on ASTAP at https://www.cloudyni...solver-program/ or in the software forum here; you'll need to post an image that failed to solve.



#93 SimonIRE

SimonIRE

    Surveyor 1

  • -----
  • Posts: 1,558
  • Joined: 16 Dec 2018
  • Loc: South East England

Posted 18 February 2021 - 07:26 PM

Do they solve if you try directly in ASTAP (even as a blind solve)? You may also want to check what gets written in the FITS header of those images. If it still fails, you can ask HAN in the thread on ASTAP at https://www.cloudyni...solver-program/ or in the software forum here; you'll need to post an image that failed to solve.

 

Well it now works. 

 

I have a ASI 174mm mini as a guide camera on one of my scopes. I was using a ASI 174mm (3.0) on the scope I was model building. But I hadn't the USBs for each camera incorrectly ON/OFF - so I was basically trying to model build through one scope using the guide camera on the other scope. It was only when I opened an image in PI I could see it was the guide camera. I couldn't see this in MW4. Ridiculous error to make. 

 

I still can't seem to get the QHY to connect to MW4 though (via ASCOM). 


Edited by SimonIRE, 18 February 2021 - 07:26 PM.


#94 mworion

mworion

    Explorer 1

  • -----
  • Posts: 59
  • Joined: 08 Mar 2016
  • Loc: Hohenbrunn / Munich

Posted 19 February 2021 - 04:27 AM

Hi Simon, Razvan,

 

in v1.0.4 (which should be published this evening), the ATSAP index files is fixed to all (G17, G18, H17, H18) database files.

MW4 in INDI omits the guiding camera, it only looks to the main camera (no setup to select guid camera), but main camera has to be configured correctly.

QHY should work with ASCOM driver, if there is any issue, pleas post the log file to find out.

 

Michel


  • RazvanUnderStars and SimonIRE like this

#95 dan_1984

dan_1984

    Messenger

  • -----
  • Posts: 429
  • Joined: 05 Jun 2016
  • Loc: Europe

Posted 19 February 2021 - 08:02 AM

Quick question guys. Does the seeing influence the model making and tracking? I can see how a dancing guidestar can mess up things with tracking and autoguiding.
My reasoning is, if I have a good RMS model and good polar alignment then the seeing won't influence the tracking at all.

#96 mworion

mworion

    Explorer 1

  • -----
  • Posts: 59
  • Joined: 08 Mar 2016
  • Loc: Hohenbrunn / Munich

Posted 19 February 2021 - 09:33 AM

That would be my view as well as in a model you anyway do regression over multiple points to reduce also statistical errors (which indeed seeing is).  


  • SimonIRE likes this

#97 dhaval

dhaval

    Vanguard

  • *****
  • Posts: 2,253
  • Joined: 21 Jul 2008
  • Loc: Round Rock, TX

Posted 19 February 2021 - 09:38 AM

Is there a step by step guide to set up a new 10Micron mount in a remote observatory setting? That assumes the use of an image acquisition tool like SGP with a plate solving solution already in place. 

 

Thanks and CS!



#98 Raginar

Raginar

    Cosmos

  • *****
  • Posts: 9,796
  • Joined: 19 Oct 2010
  • Loc: Pensacola, FL

Posted 19 February 2021 - 09:54 AM

Is there a step by step guide to set up a new 10Micron mount in a remote observatory setting? That assumes the use of an image acquisition tool like SGP with a plate solving solution already in place. 

 

Thanks and CS!

You set it up normally and use a MPbox to turn it on and off.  It'll save your model so there's no special procedures besides having the power box installed.

 

You don't need plate solving, but it just works.

 

I'd switch to voyager if you're using a remote setup.  It's significantly more powerful with it's scripting and could control turning on/off the mount among other things.


  • psandelle likes this

#99 RazvanUnderStars

RazvanUnderStars

    Apollo

  • *****
  • Posts: 1,321
  • Joined: 15 Jul 2014
  • Loc: Toronto, Canada

Posted 19 February 2021 - 10:19 AM

Seeing is random but the disc we get over exposures of several seconds is still centered at where the point star would have been without an atmosphere, so the plate solve still sees it in the correct place. In other words, seeing shouldn't matter for making the model (again, when not using very short exposures). 

 

For unguided tracking seeing obviously doesn't matter, for guided the best practice is to use exposures of several seconds to average the seeing, much like above. With a 10M you shouldn't need frequent corrections anyway (if at all).

 

 

Quick question guys. Does the seeing influence the model making and tracking? I can see how a dancing guidestar can mess up things with tracking and autoguiding.
My reasoning is, if I have a good RMS model and good polar alignment then the seeing won't influence the tracking at all.


  • EFT and dan_1984 like this

#100 SimonIRE

SimonIRE

    Surveyor 1

  • -----
  • Posts: 1,558
  • Joined: 16 Dec 2018
  • Loc: South East England

Posted 20 February 2021 - 02:14 AM

Hi Simon, Razvan,

 

in v1.0.4 (which should be published this evening), the ATSAP index files is fixed to all (G17, G18, H17, H18) database files.

MW4 in INDI omits the guiding camera, it only looks to the main camera (no setup to select guid camera), but main camera has to be configured correctly.

QHY should work with ASCOM driver, if there is any issue, pleas post the log file to find out.

 

Michel

 

Thanks Michel - my apologise for the slowness in responding. Test things out is hard here because we have so few clear nights. I have everything working now with my ASI cameras - tonight I will work on my QHY. 




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