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 to guide by platesolving scientific images

  • Please log in to reply
10 replies to this topic

#1 PeterWar

PeterWar

    Messenger

  • -----
  • topic starter
  • Posts: 464
  • Joined: 25 Aug 2011
  • Loc: Roses, Girona (Spain)

Posted 15 February 2019 - 04:22 PM

Greetings,

 

My 10 Micron GM 3000HPS has encoders and is quite precise, allowing for an unguided 180 second image exposure with an error way below my seeing. I'm using CCD Commander+Maxim DL for image capture with good results, but guiding adds another level of complexity that I would like to avoid.

 

I'm wandering if there's a way to guide by platesolving scientific images during an image session. Ideally I would like to make 300 images of 120 seconds each without guiding, but still maintain adequate centering of the target star. 

 

Is this possible?


Edited by PeterWar, 15 February 2019 - 04:25 PM.


#2 Charlie B

Charlie B

    Apollo

  • -----
  • Posts: 1407
  • Joined: 22 Mar 2008
  • Loc: Sterling, Virginia

Posted 16 February 2019 - 08:24 AM

Greetings,

 

My 10 Micron GM 3000HPS has encoders and is quite precise, allowing for an unguided 180 second image exposure with an error way below my seeing. I'm using CCD Commander+Maxim DL for image capture with good results, but guiding adds another level of complexity that I would like to avoid.

 

I'm wandering if there's a way to guide by platesolving scientific images during an image session. Ideally I would like to make 300 images of 120 seconds each without guiding, but still maintain adequate centering of the target star. 

 

Is this possible?

One way is to stop imaging about every 50 frames and re-center. SGP could automate this.

 

Regards,

 

Charlie B



#3 han59

han59

    Vostok 1

  • *****
  • Posts: 179
  • Joined: 09 Mar 2015
  • Loc: the Netherlands

Posted 16 February 2019 - 06:34 PM

If your mount maintains stability below the seeing, lets say 2 arcsecond each 120 seconds, so 60 arcseconds/hour, why do you need centering?  You could image for hours without any action.

 

Nevertheless, plate solving/Astrometric solving could help. You could either use plate solving for make a perfect model of your setup (mount+telescope) in advance or to center each hour if required.  For Maxim DL your probably are little restricted to PinPoint and scripting is difficult in MaximDL if I remember well.


  • psandelle likes this

#4 psandelle

psandelle

    Vanguard

  • *****
  • Posts: 2301
  • Joined: 18 Jun 2008
  • Loc: Los Angeles

Posted 17 February 2019 - 11:28 AM

Greetings,

 

My 10 Micron GM 3000HPS has encoders and is quite precise, allowing for an unguided 180 second image exposure with an error way below my seeing. I'm using CCD Commander+Maxim DL for image capture with good results, but guiding adds another level of complexity that I would like to avoid.

 

I'm wandering if there's a way to guide by platesolving scientific images during an image session. Ideally I would like to make 300 images of 120 seconds each without guiding, but still maintain adequate centering of the target star. 

 

Is this possible?

Are you platesolve modeling before unguided tracking already? That's one of the strengths of the 10Micron; its modeling is built into the mount. I can do unguided imaging all night (and do; only 10 minute exposures, but I've taken longer subs to check). Takes care of most of the repeatable hysteresis problems as well. Just make sure you have a really solid rig.

 

Paul


  • PeterWar likes this

#5 sdufoer

sdufoer

    Vostok 1

  • -----
  • Posts: 133
  • Joined: 31 Jul 2006
  • Loc: Bruges, Belgium

Posted 18 February 2019 - 05:07 AM

I have pretty much the same issue as you.  I do some kind of "autoguiding" with the platesolving.  I know how much PE my mount has and hence I know my maximum exposure time so the error stays below pixel resolution.  For my EQ6 and VMC200L I have a max exposure time of 30s.

In CCDcommander I have a subroutine for every object that loops like 300 times and the actions are: 

-move to RA/DEC

-Platesolve 20s with sync/reslew

-take 10x 30s images

 

Then after that I have a skip-to at time/altitude/or RA angle to break the subroutine.  I do this for every object (variable stars mostly).  The platesolves barrely show up on the end results.

 

With my big setup (ASA DDM85 with 16" cass), the mount model is enough to keep an object centered for several hours.  I do here the same strategry for all-nighter objects: I take 2 hours worth of images, then a quick platesolve to recenter the OTA, and so on the whole night on the same object.  Works flawless.

Here it is with simple copy/paste of 3 actions:

-move to RA/DEC

-take image 60x120s
-platesolve 10s, sync reslew

-move to RA/DEC

-take image 60x120s
-platesolve 10s, sync reslew

-move to RA/DEC

-take image 60x120s
-platesolve 10s, sync reslew

-move to RA/DEC

-take image 60x120s
-platesolve 10s, sync reslew

-move to RA/DEC

-take image 60x120s
-platesolve 10s, sync reslew

-move to RA/DEC

-take image 60x120s
-platesolve 10s, sync reslew

 

This is 12 hours worth of observing time !!


  • PeterWar likes this

#6 daw316

daw316

    Vostok 1

  • -----
  • Posts: 122
  • Joined: 10 May 2012

Posted 18 February 2019 - 10:22 AM

Solve/sych/slew is a great way to stay centered!  Do have/care about field rotation issues?



#7 PeterWar

PeterWar

    Messenger

  • -----
  • topic starter
  • Posts: 464
  • Joined: 25 Aug 2011
  • Loc: Roses, Girona (Spain)

Posted 20 February 2019 - 02:01 PM

I have pretty much the same issue as you.  I do some kind of "autoguiding" with the platesolving.  I know how much PE my mount has and hence I know my maximum exposure time so the error stays below pixel resolution.  For my EQ6 and VMC200L I have a max exposure time of 30s.

In CCDcommander I have a subroutine for every object that loops like 300 times and the actions are: 

-move to RA/DEC

-Platesolve 20s with sync/reslew

-take 10x 30s images

 

Then after that I have a skip-to at time/altitude/or RA angle to break the subroutine.  I do this for every object (variable stars mostly).  The platesolves barrely show up on the end results.

 

With my big setup (ASA DDM85 with 16" cass), the mount model is enough to keep an object centered for several hours.  I do here the same strategry for all-nighter objects: I take 2 hours worth of images, then a quick platesolve to recenter the OTA, and so on the whole night on the same object.  Works flawless.

Here it is with simple copy/paste of 3 actions:

-move to RA/DEC

-take image 60x120s
-platesolve 10s, sync reslew

-move to RA/DEC

-take image 60x120s
-platesolve 10s, sync reslew

-move to RA/DEC

-take image 60x120s
-platesolve 10s, sync reslew

-move to RA/DEC

-take image 60x120s
-platesolve 10s, sync reslew

-move to RA/DEC

-take image 60x120s
-platesolve 10s, sync reslew

-move to RA/DEC

-take image 60x120s
-platesolve 10s, sync reslew

 

This is 12 hours worth of observing time !!

Thanks, that could work as a partial solution. I have a 60 star model in my 10 Micron with virtualy no polar aligment error, so field rotation is not an issue. I want to use this feature for exoplanet photometry and would be great to have a tool to do this without loosing 10-20 seconds platesolving as they could be important to ascertain the correct shape of the transit.



#8 StarmanDan

StarmanDan

    Aurora

  • *****
  • Posts: 4509
  • Joined: 27 Aug 2007
  • Loc: Deep in the heart of Texas

Posted 20 February 2019 - 04:18 PM

You don't need super precise tracking to do exoplanet photometry.  My club has a 24" f/9 scope and camera that produces a 10x10 arc min FOV.  We have no guiding on our scope and don't have the fancy absolute encoders like on your mount which allows it to do very accurate unguided tracking.  As long as the stars are reasonably round in each exposure you're good.  Our tracking is good enough that we simply monitor the exposures every 5-10 min and mark where our target star is, and if we see it starting to drift, we manually adjust the tracking rates to keep the target reasonably centered in the camera.  But with your mount, the encoders should get you very good tracking without needing to guide.



#9 PeterWar

PeterWar

    Messenger

  • -----
  • topic starter
  • Posts: 464
  • Joined: 25 Aug 2011
  • Loc: Roses, Girona (Spain)

Posted 21 February 2019 - 03:50 PM

You don't need super precise tracking to do exoplanet photometry.  My club has a 24" f/9 scope and camera that produces a 10x10 arc min FOV.  We have no guiding on our scope and don't have the fancy absolute encoders like on your mount which allows it to do very accurate unguided tracking.  As long as the stars are reasonably round in each exposure you're good.  Our tracking is good enough that we simply monitor the exposures every 5-10 min and mark where our target star is, and if we see it starting to drift, we manually adjust the tracking rates to keep the target reasonably centered in the camera.  But with your mount, the encoders should get you very good tracking without needing to guide.

StarmanDan I highly recommend the book Exoplanet observing for amateurs, it's freely avaliable at Bruce Gary website, here's an abstract of what the book says on this topic:

 

 

When autoguiding is not used the star field will drift across the pixel field and cause systematic errors
that vary during the observing session; all star flux ratios throughout the image will vary. These
systematics won’t be completely removed by the flat field calibration because no flat field is perfect.

 

I found this to be true, Dennis Conti's excelent guide also states that Autoguiding is essential for research-grade exoplanet observing in order to minimize the drift of

the target and comparison stars.

I think with the help of encoders in my mount I could accomplish the same thing, sending guiding pulses every 120s by platesolving the scientific image could suffice.


Edited by PeterWar, 21 February 2019 - 03:53 PM.


#10 sdufoer

sdufoer

    Vostok 1

  • -----
  • Posts: 133
  • Joined: 31 Jul 2006
  • Loc: Bruges, Belgium

Posted 22 February 2019 - 02:12 PM

I think with the help of encoders in my mount I could accomplish the same thing, sending guiding pulses every 120s by platesolving the scientific image could suffice.

My mount needs at least 5s to settle after a correction, so you will lose this time as well.  You could write a program that opens pinpoint+mount driver and reads the most recent fits in the CCDcommander folder and do a platesolve + sync.  You could trigger this program as "run external program" from CCDcommander.  I have a similar self-written program that does allsky blindsolving since it's not a feature in CCDcommander.

 

Another (maybe OT) question: how do you process the exoplanet transit data?  I load the data into TRESCA.  http://var2.astro.cz...resca/index.php.  Their software makes the curve itself.

This is one of my observations: http://var2.astro.cz...0392783&lang=en


  • PeterWar likes this

#11 PeterWar

PeterWar

    Messenger

  • -----
  • topic starter
  • Posts: 464
  • Joined: 25 Aug 2011
  • Loc: Roses, Girona (Spain)

Posted 22 February 2019 - 05:22 PM

My mount needs at least 5s to settle after a correction, so you will lose this time as well.  You could write a program that opens pinpoint+mount driver and reads the most recent fits in the CCDcommander folder and do a platesolve + sync.  You could trigger this program as "run external program" from CCDcommander.  I have a similar self-written program that does allsky blindsolving since it's not a feature in CCDcommander.

 

Another (maybe OT) question: how do you process the exoplanet transit data?  I load the data into TRESCA.  http://var2.astro.cz...resca/index.php.  Their software makes the curve itself.

This is one of my observations: http://var2.astro.cz...0392783&lang=en

Thats very interesting feedback, thank you Sjoerd. I used to model-fit the data using TRESCA, but I have a better option now which is AstroimageJ. I strongly recommend Dennis Conti's guide to help you understand how to use it. 


Edited by PeterWar, 22 February 2019 - 05:26 PM.



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