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

Jumping into Ekos

  • Please log in to reply
8 replies to this topic

#1 const

const

    Sputnik

  • -----
  • topic starter
  • Posts: 45
  • Joined: 11 May 2020
  • Loc: WA

Posted 06 July 2020 - 11:08 PM

Last weekend I tried Ekos.

 

My little observatory is about 250 meters from the closest shelter. Vantage is great, but spending the whole night there is a bit too chilling, especially in winter. The goal is to be able to setup the scope in the evening and let it go for a night unattended.

Saturday night was full moon, as in 100% full. Perfect night to shoot some DSO, not having any narrow band filters. I was not patient enough to remove all the funky color casts from Andromeda. But overall the image is much better than I hoped smile.gif Actual imaging was done unattended. So my bottom line impression from Ekos is very positive.

 

One problem was Ekos guiding. I used PHD2 before. They recommend pulse guiding via the mount, so I did. And I did the same with Ekos without giving it much thought. Perhaps, ST-4 would be better. What is generally recommended? So, via the mount, it often failed on calibration when doing "DEC reverse". Like could not move enough after some iterations. I guess, it can be backlash. What are the knobs out there to play around?

 

When it calibrated, the guiding itself was quite poor, comparing to what I usually had before. I do not make any conclusions just yet, because there was some wind and likely my polar alignment was bad (in 3 hours the frame rotated ~3°).

 

Last but not least. How to make Ekos cloud resistant? I mean, when guiding fails, because of a cloud or a bump or too big tracking error, the frame may drift away quite far. Eventually the guider picks up some new star and continues happily. PHD2 does that too; a couple of times I collected like 3 hours of good data of boring star field instead of a beautiful nebula smile.gif I foresaw this problem and googled out a workaround. Schedule an infinitely repeated job of just a few images. Even if it drifts away once, soon it will re-align again. Is there a more elegant and easy to setup method? I am not even sure what is the right thing for Ekos to do in such case. Perhaps, the guider should remember all the stars from the guide camera, and then after re-locking verify that the star patterns match, otherwise fail.

 

My rig:

  • Celestron EdgeHD 8" on AVX, .7x reducer
  • Lodestar X2 Autoguider through Celestron OAG
  • Unmodified Nikon D7500
  • Laptop with Ekos on Linux (KStars 3.4.1)

Attached Thumbnails

  • M31.jpg


#2 sunnyday

sunnyday

    Voyager 1

  • *****
  • Posts: 11,276
  • Joined: 23 Sep 2019
  • Loc: the Canadian nebula .

Posted 06 July 2020 - 11:13 PM

it is a very beautiful image,love the details and the depth . thanks


Edited by sunnyday, 06 July 2020 - 11:14 PM.


#3 endlessky

endlessky

    Viking 1

  • -----
  • Posts: 666
  • Joined: 24 May 2020
  • Loc: Padova, Italy

Posted 07 July 2020 - 12:39 AM

Hi, I have been using KStars/EKOS for quite sometime, on a Raspberry 4, and I absolutely love the suite. I haven't got into guiding, yet, but I am in the process of upgrading my equipment and soon I will have an autoguiding setup.

You don't necessarily need to use EKOS internal guider and abandon PHD2. If you used to have good results with PHD2, you can keep using PHD2 and the option in EKOS to use an external guider. For the Raspberry there is a premade package called Astroberry, with all the software preinstalled, and it comes with PHD2. Using the Simulator profile, I was able to get guiding working with PHD2. I will definitely give priority to using PHD2, once I get my autoguiding setup, as I feel it might work better. PHD2 has also other nice features, like drift alignment, so I hope I can make it work also in a real situation, not only with the Simulator.

As for losing the guide star, again, I can't help you much, but I am afraid there might not be a practical solution. I don't think there is a plate-solving option in either PHD2 or EKOS internal autoguider. It would definitely be a nice thing to add, though.

Anyway, nice first light with your new setup!

Clear skies!

#4 Deesk06

Deesk06

    Viking 1

  • -----
  • Posts: 633
  • Joined: 25 Jul 2015
  • Loc: New Jersey

Posted 07 July 2020 - 06:25 AM

You can abandon the imaging sequence/guiding if the deviation goes above a certain threshold(that you set). I have not used it yet, however, it should work. 

 

Here it is directly from the indi library website: "Guiding Deviation: If checked, it enforces a limit of maximum allowable guiding deviation for the exposure, if autoguiding is used. If the guiding deviation exceeds this limit in arcseconds, it aborts the exposure sequence. It will automatically resume the exposure sequence again once the guiding deviation goes below this limit."

 

You can view it here as well: https://indilib.org/...ure-module.html


Edited by Deesk06, 07 July 2020 - 06:43 AM.


#5 const

const

    Sputnik

  • -----
  • topic starter
  • Posts: 45
  • Joined: 11 May 2020
  • Loc: WA

Posted 07 July 2020 - 11:40 AM

You can abandon the imaging sequence/guiding if the deviation goes above a certain threshold(that you set). I have not used it yet, however, it should work. 

 

Here it is directly from the indi library website: "Guiding Deviation: If checked, it enforces a limit of maximum allowable guiding deviation for the exposure, if autoguiding is used. If the guiding deviation exceeds this limit in arcseconds, it aborts the exposure sequence. It will automatically resume the exposure sequence again once the guiding deviation goes below this limit."

True, the Guiding Deviation guard is useful, but for a different purpose. I view it as it cancels a long exposure when guiding tells that it has just slipped. So that you won't spend time finishing a five minute exposure  when it's guaranteed to be bad. But it is a failure of the Capture module, not the Guiding. Guiding likely didn't even lose the guide star, so there is no great reason to drop everything and re-align.

 

I suppose, the guiding algorithms are designed with sturdy mounts in mind :) If there is a star in the 16x16 box, it must be the same star as 10 minutes ago, no need to validate it.



#6 Der_Pit

Der_Pit

    Surveyor 1

  • *****
  • Posts: 1,653
  • Joined: 07 Jul 2018
  • Loc: La Palma

Posted 07 July 2020 - 01:24 PM

As mentioned, you can (continue to) use PHD2.  That's what I do, too.  Always, the critical point for a good calibration is having proper parameters (FL, pixel size).   

And 1⁰ rotation per hour is of course massive, so you likely had a huge DEC drift that ruined the calibration...

 

Against clouds - are you using the scheduler?  It is supposed to restart the observation on failures.  This would then also include re-centering the target (in case it has drifted away too much during the outage).

 

There's currently a lot of work put into the (internal) guider algorithms.  It might help using nightly builds (or compile yourself from git) - but then it's almost mandatory to follow the

INDI forums for issues.

 

Yes, EKOS takes it's time to understand, but it's a powerful tool that constantly improves waytogo.gif

I see little reason to look for something else.



#7 Deesk06

Deesk06

    Viking 1

  • -----
  • Posts: 633
  • Joined: 25 Jul 2015
  • Loc: New Jersey

Posted 09 July 2020 - 09:53 PM

As mentioned, you can (continue to) use PHD2. That's what I do, too. Always, the critical point for a good calibration is having proper parameters (FL, pixel size).
And 1⁰ rotation per hour is of course massive, so you likely had a huge DEC drift that ruined the calibration...

Against clouds - are you using the scheduler? It is supposed to restart the observation on failures. This would then also include re-centering the target (in case it has drifted away too much during the outage).

There's currently a lot of work put into the (internal) guider algorithms. It might help using nightly builds (or compile yourself from git) - but then it's almost mandatory to follow the
INDI forums
for issues.

Yes, EKOS takes it's time to understand, but it's a powerful tool that constantly improves waytogo.gif
I see little reason to look for something else.


Maybe a stupid question, what is a Nightly Builder and how is it useful?? I have heard of this before but still have no idea what it is and how it is implemented. Thanks!

#8 Der_Pit

Der_Pit

    Surveyor 1

  • *****
  • Posts: 1,653
  • Joined: 07 Jul 2018
  • Loc: La Palma

Posted 10 July 2020 - 03:41 AM



Maybe a stupid question, what is a Nightly Builder and how is it useful?? I have heard of this before but still have no idea what it is and how it is implemented. Thanks!

Development happens in a GIT repository.  All changes are in there immediately, and once things are settled, a new stable version is released.  That only happens once in a while.  Therefore, for some distributions/OS every night a version from git is compiled.  So it depends on what system you use.  Look on the [url=https://indilib.org]INDI homepage[/url, under 'Get INDI', and select your OS/distribution.  If nightly builds are available for it, is is mentioned there, as well as how to get/install them.

It is 'bleeding edge' though, so you should know how to revert to earlier versions if things go bad....


  • Deesk06 likes this

#9 Deesk06

Deesk06

    Viking 1

  • -----
  • Posts: 633
  • Joined: 25 Jul 2015
  • Loc: New Jersey

Posted 10 July 2020 - 06:08 AM



Development happens in a GIT repository.  All changes are in there immediately, and once things are settled, a new stable version is released.  That only happens once in a while.  Therefore, for some distributions/OS every night a version from git is compiled.  So it depends on what system you use.  Look on the [url=https://indilib.org]INDI homepage[/url, under 'Get INDI', and select your OS/distribution.  If nightly builds are available for it, is is mentioned there, as well as how to get/install them.

It is 'bleeding edge' though, so you should know how to revert to earlier versions if things go bad....

Oof. Probably best if I stay away then. I know how to use a computer, but the second I need to jump into using commands and reverting back to previous versions its all bets off for me. I need to learn how to do that because it seems like it would be really helpful. Anyway, for now I will just steer clear. 

 

Thanks!


Edited by Deesk06, 10 July 2020 - 06:08 AM.



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