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

New PHD2 static polar alignment tool

  • Please log in to reply
61 replies to this topic

#26 kingjamez

kingjamez

    Vanguard

  • *****
  • Posts: 2,310
  • Joined: 03 Oct 2006
  • Loc: Fairfax, VA

Posted 03 October 2017 - 10:28 AM

 

 

How can I alter how much it moves the mount? I tried again tonight and every time it moves the mount the star gets lost. I tried upping the search box size to 50, don't know if that would help or not, but it didn't. Even so, the star moved outside of the search box every time anyway. 

 

I know I can do it manually, but just providing feedback on the motorized method.

 

-Jim

As an afterthought, it could be that the focal length or pixel size entered into PHD2 is incorrect. The clue came from Iver's post earlier. Currently I show the pixel scale on the screen but I suspect that showing the focal length and pixel size could be more intuitive.

Tonight was clear enough for testing so I did a number of tests at different focal lengths, different reference stars and at many distances from the pole inclduing one where it was several FOVs away. I've just submitted a bug fix plus a few useability changes which shoudl become available in the next dev release.

 

I was just coming here to say that I thought the problem is a pixel size issue. I remembered that my guide camera's ASCOM driver incorrectly reports the pixel size as twice what it really is. 

 

Last night I used the tool with my main imaging camera (ASI1600mm-c) and the movement piece worked perfectly. The star was continually tracked. 

 

My issue now is that "I think" that the tool was using the wrong time somehow. I took some screenshots, but they are on my PC at home and I'm at work... I'll try to describe.

 

The green circle was very very close to the "correct" circle (I forget what color, but probably purple from the other screenshot in this thread). However the tool was asking me to move a point roughly 60 degrees away (clockwise) from where it currently was. This would have resulted in gross movements. I know I'm not that far off. If it told me to move from my green circle to a point close on the "correct" circle, I'd believe it. 

 

Also, when using my imaging camera, the scale of the star diagram was very small. It was impossible to determine what stars went with their letter label, save for Polaris. So I just used Polaris.

 

I'll attach screen shots this evening. 

-Jim 



#27 KenS

KenS

    Apollo

  • *****
  • topic starter
  • Posts: 1,091
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 03 October 2017 - 03:11 PM

 

 

 

How can I alter how much it moves the mount? I tried again tonight and every time it moves the mount the star gets lost. I tried upping the search box size to 50, don't know if that would help or not, but it didn't. Even so, the star moved outside of the search box every time anyway. 

 

I know I can do it manually, but just providing feedback on the motorized method.

 

-Jim

As an afterthought, it could be that the focal length or pixel size entered into PHD2 is incorrect. The clue came from Iver's post earlier. Currently I show the pixel scale on the screen but I suspect that showing the focal length and pixel size could be more intuitive.

Tonight was clear enough for testing so I did a number of tests at different focal lengths, different reference stars and at many distances from the pole inclduing one where it was several FOVs away. I've just submitted a bug fix plus a few useability changes which shoudl become available in the next dev release.

 

I was just coming here to say that I thought the problem is a pixel size issue. I remembered that my guide camera's ASCOM driver incorrectly reports the pixel size as twice what it really is. 

 

Last night I used the tool with my main imaging camera (ASI1600mm-c) and the movement piece worked perfectly. The star was continually tracked. 

 

My issue now is that "I think" that the tool was using the wrong time somehow. I took some screenshots, but they are on my PC at home and I'm at work... I'll try to describe.

 

The green circle was very very close to the "correct" circle (I forget what color, but probably purple from the other screenshot in this thread). However the tool was asking me to move a point roughly 60 degrees away (clockwise) from where it currently was. This would have resulted in gross movements. I know I'm not that far off. If it told me to move from my green circle to a point close on the "correct" circle, I'd believe it. 

 

Also, when using my imaging camera, the scale of the star diagram was very small. It was impossible to determine what stars went with their letter label, save for Polaris. So I just used Polaris.

 

I'll attach screen shots this evening. 

-Jim 

 

The magenta circle is the calculated arc that passes through the calibration points and the green orbit is where the reference star ought to be. 

The issue with the corrections is hopefully fixed in the dev2 release coming soon. I tested it extensively last night. It came about when one calculation gave a negative radius for the orbit

It sounds like Polaris is the best option in the NH due to the lack of stars near the pole. Once you have a good PA you could use the tool to familiarise yourself with the star patterns around the pole. Then you can use those for a quick check of your PA each session.


  • kingjamez likes this

#28 lucutes

lucutes

    Messenger

  • *****
  • Posts: 465
  • Joined: 17 Oct 2016
  • Loc: Western Canada

Posted 03 October 2017 - 03:21 PM

Maybe a dumb question, but could this replace the "need" for a QHY Pole Master?



#29 Iver

Iver

    Mariner 2

  • *****
  • Posts: 275
  • Joined: 12 Oct 2007
  • Loc: Monterey County,Ca.

Posted 04 October 2017 - 12:50 AM

     Hi Ken, I installed the updated version and plate solved a guider image to get the exact FL of the guide scope. Your app. worked extremely well!

Thanks so much for adding this excellent tool to PHD2!

 

Capture.JPG


  • kingjamez likes this

#30 KenS

KenS

    Apollo

  • *****
  • topic starter
  • Posts: 1,091
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 04 October 2017 - 02:16 AM

Maybe a dumb question, but could this replace the "need" for a QHY Pole Master?

It is certainly an alternative. The main advantages are the ability to use existing guiding equipment and to use shorter slews to align.

But it is hard to replace the need for retail therapy. 


  • psandelle and Merc like this

#31 KenS

KenS

    Apollo

  • *****
  • topic starter
  • Posts: 1,091
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 04 October 2017 - 02:29 AM

     Hi Ken, I installed the updated version and plate solved a guider image to get the exact FL of the guide scope. Your app. worked extremely well!

Thanks so much for adding this excellent tool to PHD2!

 

Thanks Iver, I'm very glad it worked for you. I also did a quick check of your screen shot and enhaced the contrast to double check the alignment. I could see the stars round Polaris and it does indeed look like a good PA. it also prompts me to add a couple of the brighter stars around Polaris to aid triangulation using the manual alignment.

I also thought up a way to avoid having to enter the focal length but it involes altering the calibration process a bit.


  • kingjamez likes this

#32 Iver

Iver

    Mariner 2

  • *****
  • Posts: 275
  • Joined: 12 Oct 2007
  • Loc: Monterey County,Ca.

Posted 04 October 2017 - 09:19 PM

Ken, I should have mentioned that after I polar aligned I did a short test to see how well it worked. I slewed to the East just below the meridian and started guiding with Dec. corrections off. after 20 min. my guide star had only drifted 1/4 of a pixel, that's about one and a half arc sec. with my guide camera. smile.gif 


  • Merc and KenS like this

#33 kingjamez

kingjamez

    Vanguard

  • *****
  • Posts: 2,310
  • Joined: 03 Oct 2006
  • Loc: Fairfax, VA

Posted 05 October 2017 - 12:35 AM

I installed the dev2 version today. Still no luck in getting it to help me align.

 

How is camera angle measured? I'm pretty sure it's roughly 90 degrees off on my system. The graphic showing the stars is never in the orientation that the real image is showing. 

 

I think that's the reason that the mount is telling me to move to the wrong place(time?) on the green circle. 

 

-Jim



#34 kingjamez

kingjamez

    Vanguard

  • *****
  • Posts: 2,310
  • Joined: 03 Oct 2006
  • Loc: Fairfax, VA

Posted 05 October 2017 - 12:50 AM

Never mind. I’m guessing it takes info from the guiding calibration. I’m using an OAG and have been using my main camera for the polar alignment too since it is more sensitive and has a wider FOV. It’s not in the same orientation as the guide camera so that’s probably the source of my problems.

-Jim

Edited by kingjamez, 05 October 2017 - 12:50 AM.


#35 KenS

KenS

    Apollo

  • *****
  • topic starter
  • Posts: 1,091
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 05 October 2017 - 02:51 AM

Never mind. I’m guessing it takes info from the guiding calibration. I’m using an OAG and have been using my main camera for the polar alignment too since it is more sensitive and has a wider FOV. It’s not in the same orientation as the guide camera so that’s probably the source of my problems.

-Jim

Correct. It reads the calibration data. And it does need that to be accurate to show the right corrections. The alternative is to use manual mode and place three of the stars on their correct orbit. I'm going to add an option to rotate the image for those using manual mode



#36 KenS

KenS

    Apollo

  • *****
  • topic starter
  • Posts: 1,091
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 05 October 2017 - 05:53 AM

Ken, I should have mentioned that after I polar aligned I did a short test to see how well it worked. I slewed to the East just below the meridian and started guiding with Dec. corrections off. after 20 min. my guide star had only drifted 1/4 of a pixel, that's about one and a half arc sec. with my guide camera. smile.gif

Thats a great result! I estimate that corresponds to less than an arc minute PA error and is well within what I was hoping for.



#37 George Simon

George Simon

    Viking 1

  • *****
  • Posts: 717
  • Joined: 01 Jul 2008
  • Loc: Chesnee, SC

Posted 05 October 2017 - 07:27 PM

I will be giving this tool a try during my next imaging session, which I currently have scheduled for the middle or end of next week, weather permitting. I guide using an ST-4 cable, so my effort will provide a test of how well the tool's manual mode performs in the Northern Hemisphere.

 

If I remember correctly, Ken said earlier in this thread that in manual mode, the tool estimates local sidereal time on the basis of the computer's clock. I wonder if it would be possible to include the option for the user to enter his/her exact sidereal time. There are numerous sites on the web that can be used to quickly calculate local sidereal time. Having the exact LST would allow the tool to generate a more accurate depiction of where the reference stars should be in PHD's main window. It should also make easier the task of adjusting altitude and azimuth, since the displayed target position of the reference stars could be trusted in a way that it couldn't if only an estimated LST was being used. The tool could continue to estimate the LST if the user doesn't input a value.

 

Just a thought...



#38 KenS

KenS

    Apollo

  • *****
  • topic starter
  • Posts: 1,091
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 06 October 2017 - 06:57 AM

I will be giving this tool a try during my next imaging session, which I currently have scheduled for the middle or end of next week, weather permitting. I guide using an ST-4 cable, so my effort will provide a test of how well the tool's manual mode performs in the Northern Hemisphere.

 

If I remember correctly, Ken said earlier in this thread that in manual mode, the tool estimates local sidereal time on the basis of the computer's clock. I wonder if it would be possible to include the option for the user to enter his/her exact sidereal time. There are numerous sites on the web that can be used to quickly calculate local sidereal time. Having the exact LST would allow the tool to generate a more accurate depiction of where the reference stars should be in PHD's main window. It should also make easier the task of adjusting altitude and azimuth, since the displayed target position of the reference stars could be trusted in a way that it couldn't if only an estimated LST was being used. The tool could continue to estimate the LST if the user doesn't input a value.

 

Just a thought...

Thanks George - I'm looking forward to hearing how it goes. I'd suggest to line up on Polaris and select it a your reference star. With manual mode the most reliable approach is to place three of the reference stars on their respective orbits once you have finished rotating the mount. I'll look into the LST idea. There are tow factors affecting the orientation of the corrections. One is LST and the other is the orientation of the mount. I'm putting in a feature to input the HA of the mount as that has potential to have a larger error than a calclated LST. Im also adding a couple of brighter stars nea Polaris so its easier to find three stars to place on their orbits



#39 George Simon

George Simon

    Viking 1

  • *****
  • Posts: 717
  • Joined: 01 Jul 2008
  • Loc: Chesnee, SC

Posted 06 October 2017 - 11:56 AM

 

I will be giving this tool a try during my next imaging session, which I currently have scheduled for the middle or end of next week, weather permitting. I guide using an ST-4 cable, so my effort will provide a test of how well the tool's manual mode performs in the Northern Hemisphere.

 

If I remember correctly, Ken said earlier in this thread that in manual mode, the tool estimates local sidereal time on the basis of the computer's clock. I wonder if it would be possible to include the option for the user to enter his/her exact sidereal time. There are numerous sites on the web that can be used to quickly calculate local sidereal time. Having the exact LST would allow the tool to generate a more accurate depiction of where the reference stars should be in PHD's main window. It should also make easier the task of adjusting altitude and azimuth, since the displayed target position of the reference stars could be trusted in a way that it couldn't if only an estimated LST was being used. The tool could continue to estimate the LST if the user doesn't input a value.

 

Just a thought...

Thanks George - I'm looking forward to hearing how it goes. I'd suggest to line up on Polaris and select it a your reference star. With manual mode the most reliable approach is to place three of the reference stars on their respective orbits once you have finished rotating the mount. I'll look into the LST idea. There are tow factors affecting the orientation of the corrections. One is LST and the other is the orientation of the mount. I'm putting in a feature to input the HA of the mount as that has potential to have a larger error than a calclated LST. Im also adding a couple of brighter stars nea Polaris so its easier to find three stars to place on their orbits

 

Good ideas, Ken. And in case you add the HA-input feature to manual mode, would you just confirm for me that the Home position--counterweights down--has an HA value of 18 h?



#40 KenS

KenS

    Apollo

  • *****
  • topic starter
  • Posts: 1,091
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 06 October 2017 - 05:11 PM

 

Good ideas, Ken. And in case you add the HA-input feature to manual mode, would you just confirm for me that the Home position--counterweights down--has an HA value of 18 h?

 

 

 

Thats correct. And it is also the presumed position in manual mode at present. The new feature will be coming very soon. I'm just doing dome final testing.


  • George Simon likes this

#41 KenS

KenS

    Apollo

  • *****
  • topic starter
  • Posts: 1,091
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 08 October 2017 - 07:37 AM

Just submitted a set of changes. These allow the helper display to be panned by double clicking on the point to be centred. e.g. if you want Poalris in the centre of the display double click it. A bread crumb shows the direction of the pole in case you get lost. In manual mode you can now adjust your scope Hour Angle manually to get a more accurate display and corrections. Plus there was a little error in the LST calculation that put it about 2 hours out. There is no more need to click Calculate in manual mode - all input changes take effect automatically. And error handling has been improved. Previously when an error occurred the program just kept going. Now it stops processing. The window has also been tidied up and takes up a bit less screen space. 

During testing I found that it is also important to have the scope properly star aligned so it gives the correct RA reading. However, if it is out, the method of placing three stars on their orbits works.


  • George Simon likes this

#42 KenS

KenS

    Apollo

  • *****
  • topic starter
  • Posts: 1,091
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 21 October 2017 - 05:51 AM

You'll see some nice new features in dev4 that help both manual and auto users:

  • The JNOW algorithm will hopefully result in more accurate placement of the reference stars based on their J2000.0 coordinates which in turn should make the corrective moves more accurate
  • You have the option to turn off the reference star orbits to make the corrections easier to see
  • Auto mode users can now adjust the rotation (hour angle)  of the reference stars. It is meant for small adjustments but the full 360 degrees is available. The idea behind this is that you adjust the rotation so that the corrections to move, say, three reference stars to their correct positions is the same for all three. That provides greater accuracy as it accounts for any pointing error in the mount.
  • The calculated PA error in arc minutes is now displayed on the status line for bragging rights and for comparison with drift alignment. In truth, it should be used so you know when you are "near enough"
  • Slightly brighter colors for alt and az corrections should hopefully make it easier to see them in the dark
  • A "Clear" button is provided to clear the display completely so you don't need to close and reopen the tool
  • A "Flip Camera" option is provided for those situations where the displayed star positions are inverted from their true positions. Other changes in the code hopefully make this redundant but for now its worth having


#43 George Simon

George Simon

    Viking 1

  • *****
  • Posts: 717
  • Joined: 01 Jul 2008
  • Loc: Chesnee, SC

Posted 21 October 2017 - 06:32 AM

 

You'll see some nice new features in dev4 that help both manual and auto users:

  • The JNOW algorithm will hopefully result in more accurate placement of the reference stars based on their J2000.0 coordinates which in turn should make the corrective moves more accurate
  • You have the option to turn off the reference star orbits to make the corrections easier to see
  • Auto mode users can now adjust the rotation (hour angle)  of the reference stars. It is meant for small adjustments but the full 360 degrees is available. The idea behind this is that you adjust the rotation so that the corrections to move, say, three reference stars to their correct positions is the same for all three. That provides greater accuracy as it accounts for any pointing error in the mount.
  • The calculated PA error in arc minutes is now displayed on the status line for bragging rights and for comparison with drift alignment. In truth, it should be used so you know when you are "near enough"
  • Slightly brighter colors for alt and az corrections should hopefully make it easier to see them in the dark
  • A "Clear" button is provided to clear the display completely so you don't need to close and reopen the tool
  • A "Flip Camera" option is provided for those situations where the displayed star positions are inverted from their true positions. Other changes in the code hopefully make this redundant but for now its worth having

 

Ken,

 

I'm looking forward to trying out the updated tool. When I attempted to use it last week, I was hampered by my inability to distinguish which reference star was which (other than Polaris, of course, which was obvious). Had I devoted more time to it, I might have been able to figure things out; however, I was also attempting my first mosaic ever that night and was in a hurry to start imaging. When I try the tool again, I will make sure to choose an easy imaging target so that I can devote the requisite time to using the tool correctly.

 

Thanks for all of your work on this!



#44 Iver

Iver

    Mariner 2

  • *****
  • Posts: 275
  • Joined: 12 Oct 2007
  • Loc: Monterey County,Ca.

Posted 23 October 2017 - 11:24 PM

Hi Ken, I used the dev.4 release Sat. No issues, everything worked very well, quick, easy and great results!

Thanks!!


Edited by Iver, 23 October 2017 - 11:24 PM.


#45 entilza

entilza

    Soyuz

  • *****
  • Posts: 3,801
  • Joined: 06 Oct 2014
  • Loc: Oakville, ON, Canada

Posted 27 October 2017 - 07:59 AM

I just wanted to to give a quick review on this.  I tried out the tool and I had difficulties determining which star was one of the ones in the letter range A-G.  Each time I thought I had the correct star it would appear I did not.  It also was never really showing me the correct angle of the stars that's probably why.

 

I had already been aligned with the polemaster so I know I was close.  Finally I decided to use polaris itself for the calculation it worked as expected.  However in order to get Polaris to be chosen as the guide star, I had to do a few tricks such as turning down the exposure time to extremely low.  Like 0.02 seconds and turn off the mean noise reduction.  This finally allowed me get Polaris small enough to be able to be selected.  I probably could have turned down gain but I hate to touch that in case I forget to put it back.

 

When I turn my mount on I don't do anything with it in terms of the date and time, so I am not sure if that has something to do with it.  Really I just used the polemaster then jump right to SGP and do plate solving.  It looks like I would be able to use this tool by using Polaris right now.   I have a fairly long focal length of 280mm and the asi120mm guide camera so there is not a very wide field of view.

 

It is a nice tool though, I like the small rotations and how quickly the center is calculated.  

 

Anyway thanks for the nice addition.



#46 KenS

KenS

    Apollo

  • *****
  • topic starter
  • Posts: 1,091
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 27 October 2017 - 09:16 AM

Hi Martin , thanks for the feedback.

There is a bit of a shortage of good stars in the north so I generally recommend using Polaris or one of the other stars in the FOV when centred on Polaris.

If you are using dev4 you can double click on the helper display to center it where you clicked. I'm just adding another option to center on the selected reference star as Polaris is often outside the FOV when centered on the pole. If you can choose one of the other reference stars you'll have fewer issues with saturation.

Or if you want to use Polaris you could also try stopping down your aperture.

Using EQMOD you should not have to enter the date and time as it should get these from your PC. In dev4 you can adjust the rotation (Hour Angle) of the helper display or even Flip it so that it matches what you see on your display. 



#47 davidgr1976

davidgr1976

    Vostok 1

  • -----
  • Posts: 153
  • Joined: 06 Oct 2013
  • Loc: Madrid - Spain

Posted 11 December 2017 - 02:14 PM

Hello
Is it possible to use it with an OAG?

Thanks!

#48 KenS

KenS

    Apollo

  • *****
  • topic starter
  • Posts: 1,091
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 12 December 2017 - 03:54 AM

In general yes but the FOV might be a limiting factor. You also use your imaging camera if PHD2 recognises it.



#49 ChristopherL

ChristopherL

    Mariner 2

  • -----
  • Posts: 264
  • Joined: 20 Sep 2016
  • Loc: UK

Posted 12 December 2017 - 09:40 AM

First thanks to KenS for putting the time into this new software. That said haven't we already got well-tried solutions to PA'ing when the poles are visible? What I would like to see is a computer-assisted process to help drift alignment when the poles are not visible. This would be so useful as I could then put my telescope at the other end of the garden and open up a new part of the night sky for me. I guess I will get to grips with ASPA & drift alignment eventually anyway.

 

Apologies if I have misunderstood what KenS is doing, perhaps there's a big advantage to his software over Sharpcap which has escaped me.



#50 KenS

KenS

    Apollo

  • *****
  • topic starter
  • Posts: 1,091
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 13 December 2017 - 02:13 AM

You can always use the original PA drift alignment in PHD2 which allows you to align away from the pole. Based on what I learnt by developing the Polar Drift tool I can envisage ways to improve the useability of the original PA drift tool. In particlar, giving a better indication of which direction to adjust. But apart from that the method is rather difficult to automate as it needs the drift to be measured at two locations abbout 90 degrees apart.

The main advantages of Static PA  are that it is bundled into PHD2 along with the other tools and the alogrithm allows a much smaller movement to determine the location of the RA axis than other tools use. This in turn enables the rotation to be automated more readily.

Bundling the various PA tools has the advantage that you can do a quick rough alignment with Polar Drift, switch to SPA for more accuracy and then to regular drift PA for the greatest accuracy.

There are other advantages in being open source so the tools will always be free.


  • ChristopherL likes this


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