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

How does your automation software handle passing clouds?

  • Please log in to reply
21 replies to this topic

#1 Henry from NZ

Henry from NZ

    Apollo

  • *****
  • topic starter
  • Posts: 1,321
  • Joined: 26 Jan 2014

Posted 29 November 2019 - 06:22 AM

Unfortunately where I live even on a supposedly clear night passing clouds are frequently encountered.

If I have to wait for a night with absolutely no cloud cover, I would not be doing any imaging at all.

As such, I need to have an automation software that will handle passing clouds well, by appropriately and efficiently suspending and restarting the sequence to minimise down time.

I use SGPro (but only 2.x version), and my experience with its error recovery is not great.

When PHD2 loses the guide star the sequence will attempt to restart - but quite often it cannot restart correctly even though the cloud has cleared.

Tonight is one such night - I have to manually restart the sequence like > 10 times in 3 hours.

The worst that can happen is when SGPro try to re-center on resume, but thin clouds screw up the validation frame. In this situation, SGPro would be completely confused where it is. If I try to restart manually at this point, SGPro would not attempt to re-center before resuming.

 

By this point, I am really about to pack up the gear - but it is a shame as the sky had been so bad lately I have not imaged for months and it would seem that a night like this is as good as it is going to get.

 

My question is - how do you optimise SGPro's error recovery feature?

 

How's your experience with other automation software in this regards?

 

 

 

 

 



#2 Bob Koch

Bob Koch

    Lift Off

  • -----
  • Posts: 7
  • Joined: 05 Oct 2013

Posted 29 November 2019 - 08:41 AM

Have you looked at the latest SGP v3.1 beta? They have put a lot of work into recovery features recently.



#3 Salacious B Crumb

Salacious B Crumb

    Viking 1

  • *****
  • Posts: 917
  • Joined: 13 Apr 2017
  • Loc: Greater Atlanta Area, GA

Posted 29 November 2019 - 09:05 AM

It would be helpful to see Henry what your log says when the error happens. If you haven't done already, I would recommend posting that on the SGP forum and see what Ken and those guys say about it. The SGP is not perfect but I'm not sure if there is such a program. I personally have recovery mode set for 10 hours, trying every 10 minutes. This is mostly for nights when there is another batch of clear skies coming later in the morning  so I can utilize those also, even after hours of clouds before that.

 

 

- Mikko



#4 kathyastro

kathyastro

    Vanguard

  • *****
  • Posts: 2,386
  • Joined: 23 Dec 2016
  • Loc: Nova Scotia

Posted 29 November 2019 - 10:01 AM

The most annoying thing about SGP's error recovery is that useless message "Something terrible has happened!"  Apparently, it didn't occur to them that I might want to know what happened. mad.gif

 

It is very rare that there is only one cloud in the sky.  More likely, there is a front moving in and the clouds will stick around until morning.  So a cloud typically means an aborted session.  I have occasionally had it recover, but the weather here makes it unlikely.


  • Salacious B Crumb likes this

#5 jhayes_tucson

jhayes_tucson

    Fly Me to the Moon

  • *****
  • Posts: 7,388
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 29 November 2019 - 11:42 AM

Error tolerance is something that I've been focused on for some time and it's something that every AP software package should include.  Unfortunately, some of the manufacturers are completely clueless about the concept.  SJP at least makes some effort to handle cloud events.  On the other hand, MaximDL just shuts down for the rest of the night if any fault occurs--and it doesn't matter what it is.  Even the error notification system in MaximDL is a joke.  It's a system stuck in the 90's and mostly unusable.  I have lost a lot of expensive clear sky time to this problem and I've spoken to the folks at DL about this but they won't even discuss it.  Their attitude is that if something (actually anything) goes wrong, it's not their problem to solve.  They are so set in their ways and intransigent that I've completely given up trying to help them with anything.  In principle, handling intermittent clouds or dropped communication events should be easy to deal with.

 

John


  • psandelle, Salacious B Crumb, RossW and 1 other like this

#6 Henry from NZ

Henry from NZ

    Apollo

  • *****
  • topic starter
  • Posts: 1,321
  • Joined: 26 Jan 2014

Posted 29 November 2019 - 01:01 PM

Well, tonight was very frustrating. It was not a perfect night but the sky was clear enough that I should at least get something. But it was just not going to play ball. I have therefore made use of the Black Friday special and purchased a license of Voyager. I figure if I am going to pay to upgrade my SGPro to v3 but without guarantee of improvement, I may as well use similar money to try something else that seems to have good reviews.

 

To summarise SGPro (v2.x) problem:

 

1. After PHD2 has lost the guide star for some time, SGPro would detect that, that's great

2. However after the cloud has passed it does not seem to direct PHD2 to reacquire a guide star. Sometimes PHD2 would show up an alert that needs to be dismissed manually. Sometimes PHD2 would be looping but the guide star is no longer in the original position and not re-acquired.

3. Sometimes, a guide star would be reacquired, and then PHD2 would try to recenter. However, if recenter fails for some reason e.g. plate solving problem then SGPro will be stuck. To recover this, I will have to manually exit the centering dialog box, which will then abort the sequence, then I have to manually re-try re-centering, manually re-acquire the guide star and then manually restart the sequence.

4. This problem is particularly bad with my longer focal length set up (EdgeHD 8" reduced to 1400 mm) while using an OAG, such the guide star can be very faint for some targets and is very susceptible to even the slightly wisp of passing clouds, and there may be only one guide star in the field. This is also the situation where re-centering may fail due to plate solving problem.


Edited by Henry from NZ, 29 November 2019 - 01:12 PM.


#7 Astrola72

Astrola72

    Viking 1

  • *****
  • Posts: 905
  • Joined: 16 Feb 2014
  • Loc: Maryland, USA

Posted 29 November 2019 - 02:15 PM

... The SGP is not perfect but I'm not sure if there is such a program. I personally have recovery mode set for 10 hours, trying every 10 minutes. This is mostly for nights when there is another batch of clear skies coming later in the morning  so I can utilize those also, even after hours of clouds before that.

 

 

- Mikko

Mikko - Has SGP successfully recovered for you, after hours of clouds? In my experience SGP has never successfully recovered a session. I'd love to know that it's at least possible. - Joe



#8 BenKolt

BenKolt

    Vanguard

  • *****
  • Posts: 2,168
  • Joined: 13 Mar 2013

Posted 29 November 2019 - 02:58 PM

Henry:

 

Yes, this is certainly frustrating, and in the past I have lost precious clear imaging time due to one little stray cloud!

 

However, I have had some successful recoveries with SGP, although note that I am using the current or nearly-current SGP beta version.  There is an option to set the recovery time, and I have set mine for very long (many hours), particularly on nights where I expect stray clouds will be passing by on an otherwise clear night.

 

There are also lots of options available for when to autofocus, plate solve, etc, and it may be that you can adjust to do one or both following a recovery or perhaps after a slew to a new target, which ought to accomplish the same thing.

 

As was advised before, you ought to post this to the SGP forum.  The beta versions are coming out rapidly as there is a welcome flurry of activity going on right now.  One of the topics being addressed is improvement of the recovery algorithm.

 

Good luck!

 

Best Regards,

Ben


  • RossW likes this

#9 rgsalinger

rgsalinger

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 7,332
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 29 November 2019 - 03:12 PM

I use CCD autopilot on top of either MaximDL or the SKYX. It checks my weather system and if it's cloudy / windy / wet close the roof and parks the telescope and just waits for an hour. I have a small helper program provided by another CCDAP user that tries at intervals to resume the session after the wait has expired. I lose little or no clear sky time at all these days thanks to these two programs. Of course just because it's clear doesn't mean that you're going to get fabulous results.

 

I've also seen that Voyager has a similar facility. I had trouble using it because my weather system is on a different (networked) computer at my observatory but I'm sure it works for some people.

 

 

Rgrds-Ross


  • psandelle likes this

#10 Salacious B Crumb

Salacious B Crumb

    Viking 1

  • *****
  • Posts: 917
  • Joined: 13 Apr 2017
  • Loc: Greater Atlanta Area, GA

Posted 29 November 2019 - 07:00 PM

Error tolerance is something that I've been focused on for some time and it's something that every AP software package should include.  Unfortunately, some of the manufacturers are completely clueless about the concept.  SJP at least makes some effort to handle cloud events.  On the other hand, MaximDL just shuts down for the rest of the night if any fault occurs--and it doesn't matter what it is.  Even the error notification system in MaximDL is a joke.  It's a system stuck in the 90's and mostly unusable.  I have lost a lot of expensive clear sky time to this problem and I've spoken to the folks at DL about this but they won't even discuss it.  Their attitude is that if something (actually anything) goes wrong, it's not their problem to solve.  They are so set in their ways and intransigent that I've completely given up trying to help them with anything.  In principle, handling intermittent clouds or dropped communication events should be easy to deal with.

 

John

John with your experience I bet a lot of us would be curious to know if you have thought about giving a chance to one of these newcomers; Voyager, Nina, etc?

 

 

- Mikko



#11 Salacious B Crumb

Salacious B Crumb

    Viking 1

  • *****
  • Posts: 917
  • Joined: 13 Apr 2017
  • Loc: Greater Atlanta Area, GA

Posted 29 November 2019 - 07:05 PM

Mikko - Has SGP successfully recovered for you, after hours of clouds? In my experience SGP has never successfully recovered a session. I'd love to know that it's at least possible. - Joe

It has, not sure about the hours thought, but definitely possible yes. As I have said, SGP works for me nine times out of ten so I really don't have a justified reason to try anything else. But I do like the concept of an open source program, hence the reason I want to try Nina at some point .

 

 

Mikko



#12 Alex McConahay

Alex McConahay

    Cosmos

  • *****
  • Posts: 9,573
  • Joined: 11 Aug 2008
  • Loc: Moreno Valley, CA

Posted 29 November 2019 - 08:19 PM

I have come back in the morning a few times and did not have the expected sequence finish (with SGP). I noted that there was a while where it was in recovery mode. But it restarted itself. However that lost hour or so meant the sequence did not have time to finish.

 

Probably a passing cloud. I think the longest was about an hour before it kicked back in. I have it check for 90 minutes before giving up. However, most of the time once the recovery mode starts, I've pretty much lost the night. I do not think clouds in my part of the world are the puffy pass on through type. As with Kathyastro earlier, once they come, they are here for a while.

 

And I have had even more mornings when I went out, and it was cloudy, and the rig had shut down hours ago when the clouds had first arrived.  

Alex



#13 jhayes_tucson

jhayes_tucson

    Fly Me to the Moon

  • *****
  • Posts: 7,388
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 01 December 2019 - 12:59 AM

John with your experience I bet a lot of us would be curious to know if you have thought about giving a chance to one of these newcomers; Voyager, Nina, etc?

 

 

- Mikko

 

The main reason that I'm still using MaximDL is that it works seamlessly with FocusLock.  I'd love to get away from MaximDL but whatever I consider has to work seamlessly with FocusLock or I won't go there.  So, how well do Voyager, Nina, etc. work with FocusLock?

 

John


  • psandelle likes this

#14 rockstarbill

rockstarbill

    Cosmos

  • *****
  • Posts: 7,604
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 01 December 2019 - 01:03 AM

The main reason that I'm still using MaximDL is that it works seamlessly with FocusLock.  I'd love to get away from MaximDL but whatever I consider has to work seamlessly with FocusLock or I won't go there.  So, how well do Voyager, Nina, etc. work with FocusLock?

 

John

In the case of Voyager - it does not specifically work with FocusLock. You could disable focusing entirely though, and just let FocusLock do its thing. 



#15 Henry from NZ

Henry from NZ

    Apollo

  • *****
  • topic starter
  • Posts: 1,321
  • Joined: 26 Jan 2014

Posted 01 December 2019 - 05:58 AM

I am happy to report that Voyager has solved the problem. Last night I had a whole sequence run from beginning to end with no interventions from me despite PHD2 losing the guide stars more than a few times, in weather condition worse than the night before with SGPro. I did have problems getting Voyager to do a few tasks needed for me to get the rig up and running at the beginning, but that's probably because I do not know where to look. I have posted on the voyager forum to see if there are some suggestions.


  • psandelle and AstroGabe like this

#16 Charlie B

Charlie B

    Apollo

  • -----
  • Posts: 1,498
  • Joined: 22 Mar 2008
  • Loc: Sterling, Virginia

Posted 01 December 2019 - 07:54 AM

I use SGP3 and have both recovered and non-recovered events.  Recovered events were almost always passing clouds.  The unrecovered events were: 1 clouds just prior to a meridian flip which caused the mount to hit the limit, 2 A meridian flip where the counterweight also slipped and the mount could not track and 3 dew heater failure on the guide camera.  Otherwise, SGP seems to work as advertised.  

 

Regards,

 

Charlie B



#17 BenKolt

BenKolt

    Vanguard

  • *****
  • Posts: 2,168
  • Joined: 13 Mar 2013

Posted 01 December 2019 - 11:25 AM

The main reason that I'm still using MaximDL is that it works seamlessly with FocusLock.  I'd love to get away from MaximDL but whatever I consider has to work seamlessly with FocusLock or I won't go there.  So, how well do Voyager, Nina, etc. work with FocusLock?

 

John

John:

 

SGP has an interface now with FocusLock, but I have no experience using it yet, so I can't say how "seamlessly" it works with it.

 

BEn



#18 psandelle

psandelle

    Soyuz

  • *****
  • Posts: 3,520
  • Joined: 18 Jun 2008
  • Loc: Los Angeles

Posted 01 December 2019 - 11:40 AM

I am happy to report that Voyager has solved the problem. Last night I had a whole sequence run from beginning to end with no interventions from me despite PHD2 losing the guide stars more than a few times, in weather condition worse than the night before with SGPro. I did have problems getting Voyager to do a few tasks needed for me to get the rig up and running at the beginning, but that's probably because I do not know where to look. I have posted on the voyager forum to see if there are some suggestions.

Yup, Voyager is bulletproof when it comes to this sort of thing. Seriously cool software. Once you get the hang of it, where things are makes a lot of sense. And then just get the web interface and check on it all wirelessly (in yet another well-thought placement of things).

 

Paul



#19 Fernando134

Fernando134

    Messenger

  • -----
  • Posts: 421
  • Joined: 02 Feb 2013
  • Loc: Brasil

Posted 01 December 2019 - 12:08 PM

The most annoying thing about SGP's error recovery is that useless message "Something terrible has happened!"  Apparently, it didn't occur to them that I might want to know what happened. mad.gif

 

It is very rare that there is only one cloud in the sky.  More likely, there is a front moving in and the clouds will stick around until morning.  So a cloud typically means an aborted session.  I have occasionally had it recover, but the weather here makes it unlikely.

"The most annoying thing about SGP's error recovery is that useless message "Something terrible has happened!"  Apparently, it didn't occur to them that I might want to know what happened. mad.gif"

 

+1!!!!



#20 jhayes_tucson

jhayes_tucson

    Fly Me to the Moon

  • *****
  • Posts: 7,388
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 01 December 2019 - 12:29 PM

In the case of Voyager - it does not specifically work with FocusLock. You could disable focusing entirely though, and just let FocusLock do its thing. 

 

Bill,

I wish that it were that simple.  FocusLock does it’s thing using the guide camera image so unless the guiding software can pass that image to FL in real time, it won’t work.  The guiding software also has to pass filter information to FL so that the offset parameters don’t get lost.

 

John


Edited by jhayes_tucson, 01 December 2019 - 12:39 PM.


#21 jhayes_tucson

jhayes_tucson

    Fly Me to the Moon

  • *****
  • Posts: 7,388
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 01 December 2019 - 12:34 PM

John:

 

SGP has an interface now with FocusLock, but I have no experience using it yet, so I can't say how "seamlessly" it works with it.

 

BEn

 

Hi Ben,

That’s right.  SGP is indeed a viable option but since I have a system that mostly works, I’ve been lazy about jumping ship.  Getting SGP, PHD2, and FL all talking to each other, learning SGP, and shaking out the bugs isn’t going to be a one night affair.  That whole project currently sits on my “I gotta get this done one of these days” shelf.  It’s a good project for some full-moon nights.

 

John



#22 BenKolt

BenKolt

    Vanguard

  • *****
  • Posts: 2,168
  • Joined: 13 Mar 2013

Posted 01 December 2019 - 05:36 PM

Hi Ben,

That’s right.  SGP is indeed a viable option but since I have a system that mostly works, I’ve been lazy about jumping ship.  Getting SGP, PHD2, and FL all talking to each other, learning SGP, and shaking out the bugs isn’t going to be a one night affair.  That whole project currently sits on my “I gotta get this done one of these days” shelf.  It’s a good project for some full-moon nights.

 

John

This is exactly the state that I'm in with adding FocusLock or trying out Innovation Foresight's new software options for guiding and focusing.  I had a few clear nights over Thanksgiving, but since the moon was new, I didn't want to expend any of the precious time on adding new software.  I know from experience that trying out new software will almost guarantee sacrificing a night of imaging, so I just plan for it.

 

Ben




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