Polar Alignment without Polaris (workflow)

Discussions on extending SharpCap using the built in Python scripting functionality
fly_n
Posts: 26
Joined: Wed Oct 02, 2019 7:54 pm

Re: Polar Alignment without Polaris (workflow)

Post by fly_n »

Hello Jean-Francois,
thank you for the log files.
I see the error you are describing. Somehow the Sidereal Time is set to 0.000 for the EQ8 with SynScan and EQMOD, which I don't quite understand.

To make the pointing calculations I need the Sidereal Time, which is a function of the date, time and location.
Unfortunately the Sidereal Time cannot be queried with the SharpCap.Mounts.SelectedMount. function.
I saw three options:
a) have the user enter the ASCOM name of the mount and go through ASCOM to get the number.
I opted against this, as I feared I would not be able to provide a complete list of mount driver names and the selection would take up too much real estate on the GUI.
b) calculate the Sidereal Time with an implementation from a book. I also opted against this idea, as I was not certain it would be equal within a few seconds to the mount and be another error source.
c) lastly I had the idea to use the ASCOM simulator as a source of the information. I thought the ASCOM time would be synchronized with the mount and the location is from the GUI and every ASCOM install also has the simulator... so what could possibly go wrong.
It functions well with the TeenAstro driver, but from your experience does not seem to work with EQMOD / SynScan.

I don't understand why the ASCOM simulator should not work with EQMOD, but if it is so...
Maybe Robin would be so kind to implement a SharpCap.Mounts.SelectedMount.SiderealTime function to make it easier to calculate things or I could try to go with SharpCap.Mounts.SelectedMount.Name and ASCOM to do the query on the actual mount and not go through the simulator.

Kind regards,
Norman
User avatar
admin
Site Admin
Posts: 5725
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Polar Alignment without Polaris (workflow)

Post by admin »

Hi Norman,

I will add the sidereal time property to SharpCap 3.3 – watch out for the next update.

Cheers, Robin
fly_n
Posts: 26
Joined: Wed Oct 02, 2019 7:54 pm

Re: Polar Alignment without Polaris (workflow)

Post by fly_n »

Thank you Robin, that will be helpful.
I just noted that the SharpCap.Mount.SelectedMount.Name route would not work, as ASCOM would like the ProgID and not the name.

Jean-Francois, I was playing with the refraction on actual images and trying to see if I could use an optimizer to find a temperature and pressure combination, that would give the 'perfect' refraction answer for that image series.

What I noted however was the following, the optimizer was chasing it's own tail.
The reason for that is twofold.
A) above about 15° ALT, the refraction is close enough to a linear function, considering other uncertainties.
B) in order for the math to work, I need to set the DEC values such that they are set onto a plane that is rotated about the ALT axis.
I use a constant offset for that. If I subtract any refraction that is quasi linear, it looks just like a small additional rotation.
But as the real answer is unknown, the optimizer cannot figure out what part of the steepness or offset of the refraction is real and what is not.

I guess a derivative of this script could determine the actual refraction by slewing in DEC direction, but I would not expect it to be helpful in the end.
Curious and funny, but likely not helpful in getting better pictures.

Kind regards, Norman
fly_n
Posts: 26
Joined: Wed Oct 02, 2019 7:54 pm

Re: Polar Alignment without Polaris (workflow)

Post by fly_n »

Hello Jean-Francois, Robin,

thank you for the implementation of the SiderealTime property in the latest 3.3 Beta.
I've modified the script to use this now instead of going through the ASCOM telescope simulator.
It's working on my end so please give it a try if you like.

Kind regards,
Norman
Attachments
no_polaris_no_problem_v038.zip
(13.11 KiB) Downloaded 9 times
Jean-Francois
Posts: 82
Joined: Sun Oct 13, 2019 10:52 am
Location: Germany

Re: Polar Alignment without Polaris (workflow)

Post by Jean-Francois »

Hello Norman,

Sorry for the delay.
I test yesterday and today the last version of your script.

The automatic movement of the telescope is now better.
It is necessary to set the correct starting/end AZ in the case of the "pier side East".
But ... I test with other Starting AZ ... 60 or 70 ... and It was not running with these other values.

One point with the saved images and the number of pictures ... If in the preceding polar alignment the number of images is larger as during the actual polar alignment, the script performed the plate solving of all the images (past and new) together. So the user needs to take care to have the same number of images (or more) or to delete the images before performing a new polar alignment. The script knows the number of the new images ... it could be possible to limit the number of images for the plate solving with this number.

Now the real problem ... each test I performed was AZ corr = ALT corr = 0.0
I test the "test mode" ... and the same ... each time 0.0


Concerning the mount movement in DEC ... with the SynScan App of my EQ8 mount, I remark that the driver/software shows the mount axis angles.
It is possible to check that the mount is not moving in the DEC axis. That is the case ... the mount stays all the time at the same DEC value.

I test the "Bright Star SE star sync" and the Bright Star S star sync" ... the first does nothing, the second moves to a star near the meridian in the South.
The telescope moves and after a new image appears, SharpCap perform a plat solve and synchronizes the mount on the image.

I try the "Move for ALT corr" and "Move for ALT and AZ corr" ... but with both AZ corr and ALT corr = 0.0 ... it does not too much.

That is a point to discuss ... You use the mount to move the AZ / ALT correction values. It should be OK with a very precise mount without errors.
But, nobody can say that the mechanical errors are corrected for the small movement.
My proposal is to use the live view of the image itself with a mark at the correct calculated position. In this case the polar alignment is optical and independent of the mechanical error of the mount.

I remark that SharpCap or ASTAP gives the field of view after a plate solve (from the menu) with a factor 0.5.


Regards,
Jean-Francois
fly_n
Posts: 26
Joined: Wed Oct 02, 2019 7:54 pm

Re: Polar Alignment without Polaris (workflow)

Post by fly_n »

Hello Jean-Francois,

thank you for taking the time.
If the mount does not want to move, I suspect it may believe it's coming too close to the zenith.
I check with: if (DEC_START_DEG + 5.) > Lat:
(this includes 5° safety margin)

You're right with the files, it's a bit error prone. I'm hesitant to remove anything from someone elses computer.
But I could pass the number of pictures taken to a global variable and use it in the next step.

The test mode should work in any case as it is evaluating the test images in the test folders.
So if you select Test1, run platesolve results and do calculate correction, it should update to an error of 1deg+

I was wondering if you have the ASCOM package installed on your EQ8. I think you said it ran off an Eagle.
I'm using some ASCOM functions and noted that if it's not installed, it gives back funny values instead of an error.
Haven't figured out the SharpCap.Mounts.SelectedMount.J2000ToJNOW() which I had hoped I could use to replace the ASCOM function.


The last part you describe is the way I use it.
I find a bright star, put the reticule in the center of it and then offset the mount one axis at a time through the script and bring it back to the center of the reticule visually with the manual adjustment knobs on the mount. It has worked well for me the entire last year. I have relied on this method for all my images.

I was wondering if there is a way to create a row of test images with the SkySimulator with a known error, to test the robustness... but the first method has failed. I was thinking of changing the location after taking test images, but that does not work. The error actually has to be part of the of the images. Maybe I'll just create a list of predefined Siderealtime and DEC coordinates in R and run to those.

What I'm actually hoping for is some good weather to finally try out all the updates to my setup, this script, SharpCap and PHD2.
Kind regards, Norman
fly_n
Posts: 26
Joined: Wed Oct 02, 2019 7:54 pm

Re: Polar Alignment without Polaris (workflow)

Post by fly_n »

Hello,

finally the skies cleared up and I was able to try out v38 of the script on actual hardware and sky.
All in all I thought it was quite successful.
https://forum.astronomie.de/media/flamm ... 44038/full

I ran into three issues, though.
1) I noticed I forgot to implement a button to point to ALT=0 and AZ=East, which I use to mount and dismount the scope (and tighten / untighten the clutches)
2) The move ALT&AZ button does not know if the ALT button was pressed already, not to apply the ALT corrections twice. Need to insert some global check and ifs.
3) When doing...
SharpCap.Mounts.SelectedMount.Tracking = False
...later ...
SharpCap.Mounts.SelectedMount.Tracking = True

I got the following error msg:
Exception: Property write ASCOM.DeviceHub.Telescope TrackingRate is not implemented in this driver.

I had to disable all SharpCap.Mounts.SelectedMount.Tracking sections and manually set the tracking to either on of off.
It worked fine in all the ASCOM simulations, though.
I'm assuming it has something to do with the ASCOM polling. The second command being sent, before an update has been polled (?) from the mount.
???
Funny that the error is about TrackingRate, when the command was about Tracking on/off.

Hope I have some time on the weekend for an update.

Kind regards, Norman
Jean-Francois
Posts: 82
Joined: Sun Oct 13, 2019 10:52 am
Location: Germany

Re: Polar Alignment without Polaris (workflow)

Post by Jean-Francois »

Hello Norman,

For the problem with AZ corr = ALT corr = 0.000 ... I do the test with the SkyWatcher SynScan software.
I will repeat with the EQMOD driver ... maybe it will work with other mount driver.

For the simulation of a defined (known) error ... it could be possible to generate 2 set of images ... one with location A , one with location B.
Then to exchange the image header.

Regards,
Jean-Francois
Post Reply