Polar Alignment without Polaris (workflow)

Discussions on extending SharpCap using the built in Python scripting functionality
Jean-Francois
Posts: 107
Joined: Sun Oct 13, 2019 10:52 am
Location: Germany

Re: Polar Alignment without Polaris (workflow)

Post by Jean-Francois »

Hello Norman,

I test the script ... it works, but it is necessary that all the path are correct.
A long time, one of the "R" path was not correct ... 4.0.3 in place of 4.0.2 version. I correct it, and now it can calculate the PA errors.

Some points before ...
- it is nice that you modify the number of "dots" for the number of images during the capture. But, the plate solve takes all the images from the directory. Mean, if in the past you do 10 images, and now only 5, then the plate solve will take all. That is not good, because it could be in two different days (locations). If the new number is 5, then the plate solve has to take in account only this "new" 5 images.
- I test only in the east direction, without the air refraction selected.
- I notice during the telescope movement, that sometimes the DEC moves (it is simple to see curved line of the star during the continuous exposure)
I repeat the images capture sequence several time ... with Start Elev = 15, then the DEC moves each time. I change the Start Elev to 18, then it was OK. I think that the SynScan software commands sometimes the DEC to move a little back "v" and then in the position to finish with "^" direction.
Is it really not possible to command a RA movement alone ?
- The mount was already aligned before (the last days) first with the SkyWatcher hand controller 2-Star + Polar alignment, then with PHD2 drift method.

The results ...
First_polar_alignment.png
First_polar_alignment.png (724.61 KiB) Viewed 784 times
AZ = 0.0151° and ALT = 0.0064° ... also AZ = 0.9 arcmin and ALT = 0.4 arcmin
My initial alignment is not so bad.

I click on the "Bright Star S star sync" ... OK, it moves to a star (visible on the image), but the delta are small, the mount does not moves when I click on the "Move for ALT corr" or the other.
I remark that in SharpCap, when the plate solve result is small, the mount does not move to center the target. It is maybe a threshold in the SynScan software.

In addition I do not rely on the precision of the mount movement and/or the amount of the backlash in the axis.
A better way is to do it "on the image" after the movement and plate solve of the bright star (above step).
Like in PHD2 in the drift method ... It could be more precise in your script to have the information on the live image.
In your case, it will be not a circle ... you should calculate the position of the star depending on the camera rotation, pixel scale and pointing direction (also a lot of interesting thing to calculate :D )

Regards,
Jean-Francois
Jean-Francois
Posts: 107
Joined: Sun Oct 13, 2019 10:52 am
Location: Germany

Re: Polar Alignment without Polaris (workflow)

Post by Jean-Francois »

Hello Norman,

I retest yesterday evening the script.
My goal was to perform the alignment on the other mount (on the tripod) with an unknown starting polar alignment.
It was not successful ... I mount my GSO6"RC small telescope with my QHY-174 camera. But with 1370 mm focal and the relativ small chip, the field of view was too small ... no plate solve was working. It was too long to modify the setting, also I test more on the aligned mount.

I repeat several time with different changes ... (the results are in degree)
Note: 1 arcmin = 0.0167 degree

All with Pressure:1013, Temperature: 5
5 images 17 80 175 refraction-on 0.0169 0.0106
5 images 17 80 175 refraction-on 0.0126 0.0134
5 images 17 80 175 refraction-on 0.0166 0.0113
5 images 17 80 175 refraction-on 0.0169 0.0103
5 images 17 80 175 refraction-on 0.0169 0.0112
Average: AZ = 0.0160 ALT = 0.0113


5 images 17 80 175 refraction-off 0.0247 -0.0178
5 images 17 80 175 refraction-off 0.0253 -0.0188
5 images 17 80 175 refraction-off 0.0255 -0.0187
5 images 17 80 175 refraction-off 0.0261 -0.0187
5 images 17 80 175 refraction-off 0.0296 -0.0214
Average: AZ = 0.0262 ALT = -0.0191


10 images 17 80 175 refraction-on 0.0125 0.0137 (9 images plate solved)
10 images 17 80 175 refraction-on 0.0117 0.0143
10 images 17 80 175 refraction-on 0.0156 0.0120 (8 images plate solved)
10 images 17 80 175 refraction-on 0.0192 0.0089
10 images 17 80 175 refraction-on 0.0154 0.0114
Average: AZ = 0.0149 ALT = 0.0121


All with Pressure:970, Temperature: 5
10 images 17 80 175 refraction-on 0.0162 0.0103

All with Pressure:970, Temperature: 0
10 images 17 80 175 refraction-on 0.0166 0.0099


Difference between 5 and 10 images: AZ = 0.0011, ALT = 0.0008 ... AZ = 0.066 arcmin = 4 arcsec, ALT = 0.05 arcmin = 3 arcsec
So, in average it is not really necessary to have a large number of images.

My setup ... Newton 10" / 2.8 (with corrector-reducer), Atik Horizon-C, but 2x2 binning.
The field of view was relative large and the stars nice small. It could be different, if the images source is a 50/200 finder scope.


Regards,
Jean-Francois
fly_n
Posts: 31
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 feedback and the in-depth testing.
As the weather was good the last few days, I was glad to get some imaging time in.

Yes, if one forgets that there are already pictures in the folder from the last session and you choose a lesser number, it will process all.
I've changed this now for v40 to only use the amount specified in the box.

Concerning the DEC movement, I think this is something each mount handles by itself.
In this case the slew command looks like this:
SharpCap.Mounts.SelectedMount.SlewTo(RA, DEC) and I only change the RA value.
It sounds like your mount tries to take out backlash in its movement, but it is entirely up to the mount.
From SharpCap or ASCOM I see no possibility to change the movement strategy.
Here is the link to the variables and commands that are implemented in ASCOM:
https://ascom-standards.org/Help/Develo ... copeV3.htm

Working with the live image could work. ASTAP is very fast, if the deviation is small.
One could specify the target RA, DEC and check the deviation for every picture, until the process is stopped.
... or
a poor mans approach could be to offset the mount as is implemented right now and verfy the move with a platesolve and sync.


Concerning your other scope, I hear you.
In areas with few stars, I've had trouble taking pictures that can be platesolved. Upping the exposure time helps somewhat.
I've been using my main scope and as it has a similar fov it did not cause an issue.
In your case I would try the guide scope, typically your fov is around 1deg and you can play with the exposure time.

I'll freely admit that I have three guidescopes ranging from 240mm - 420mm because picking up enough stars with short exposures is always an issue.

And for the massive test data, I thank you.
Not sure yet if that is pointing to the accuracy of the goto of the offset or just the limitation of the method.

I'll attach the v40 here which contains
- the update to only solve the specified amount of pictures in the folder
- adds a SolveAndSync step after the offset to verify the correct new position (hope this helps the Skywatcher with the small movement)

Kind regards, Norman
Attachments
no_polaris_no_problem_v040.zip
(13.7 KiB) Downloaded 32 times
Jean-Francois
Posts: 107
Joined: Sun Oct 13, 2019 10:52 am
Location: Germany

Re: Polar Alignment without Polaris (workflow)

Post by Jean-Francois »

Hello Norman,

Thank you for the next version.
I can test it with the SkySimulator with the "mobile" telescope in my kitchen (I move is today while the bad weather is back).

Concerning the movement of the telescope ... it exist the "Move Axis" method and it is possible to specify the axis and rate with "IRate" and "AxisRates".
But it could be more tricky to implement ... the movement starts and moves indefinitely ... until a command stops the movement.
It is OK for small movement at low rate (satellite tracking), but maybe less safe with the mount at maximum speed.

Concerning the additional SolveAndSync step ... is that without the telescope movement after the sync ?

One alternative proposal is to implement a "manual" movement with the hand controller (or per software with the "<" or ">" button).
The method could be useful for people with a mount without computer control. A manual movement around the RA axis can be performed at any time. The "all sky" plate solving will find the pointed position. In this case, the polar adjustment must be "live" on the capture image.

You have only 3 guidescopes ... I have more :-) ... 2 Meade, 1 GSO, 1 TS and 1 Williams Optics but several guidscopes with fix eyepieces.

Regards,
Jean-Francois
cefabien
Posts: 2
Joined: Sun Apr 04, 2021 3:48 pm

Re: Polar Alignment without Polaris (workflow)

Post by cefabien »

Great job Norman,
yesterday i thought why Sharcap can't do a polar alignment without Polaris, because it is possible per exemple with Kstars et EKOS using plate solve...
And i found your post here !! :D :D
So, i get your script and i'm trying your wonderful solution, all is ok with the configuration but i have just one question :
- I don't found clearly a tutorial on "how to use your script" among your differents posts. I'm french, i don't understand all functions. Per exemple, in which order you need to click on button in the interface. I understand, in first you click "Readyness" but after i don't know the order...

Thank you for all
Fabien
fly_n
Posts: 31
Joined: Wed Oct 02, 2019 7:54 pm

Re: Polar Alignment without Polaris (workflow)

Post by fly_n »

Hello Fabien,
thank you.
Here is how I do it....
a) the readyness.. once it's all running you can typically skip this one
a+) (optional) I slew the mount East with clutches open and visually check all is pointing in the right direction. (then tighten clutches when ok)
b) I go to a bright SE star, do the plate solve and check the focus with a Bahtinov (or vice versa).
c) then I do the slew with sensible starting values.
d) the platesolve will show how many good pics you got. 7+ then proceed
e) calculate correction angles. this will show by how much you need to correct the mount.
f) I then go to a bright S star. South is best for aligning. ALT and AZ are perpendicular.
g) check that the star is close to the center of the camera view and activate crosshair and put them on the star.
h) move ALT button and manually turn the ALT knob on the mount until the star is back in the center of the cross hair.
i) move ALT/AZ button and manually turn the AZ knob on the mount until the star is back in the center of the cross hair.
(Pro tip) watch the in which direction the star is moving as you hit the move button. This makes it easier to find it again.
j) platesolve and sync
--> done

If you are unsure, you can re-run slew/ platesolve and calculate correction and check how well it went.
Should be 0.1deg or better.

Kind regards, Norman
PS: Let me know how it works out for you.
cefabien
Posts: 2
Joined: Sun Apr 04, 2021 3:48 pm

Re: Polar Alignment without Polaris (workflow)

Post by cefabien »

Thank you so much Norman,

i will let you now if i do something. at this time, the weather is bad. I have a balcony with some trees...
Hope test it soon :)
Fabien
zerolatitude
Posts: 51
Joined: Mon Mar 01, 2021 5:24 am

Re: Polar Alignment without Polaris (workflow)

Post by zerolatitude »

This looks very useful.

How do us non-software types get it installed and running :-)?

Thanks
fly_n
Posts: 31
Joined: Wed Oct 02, 2019 7:54 pm

Re: Polar Alignment without Polaris (workflow)

Post by fly_n »

Hi Zerolatitude,

I put a readme in the zip file to assist with the installation.

You need the following basic ingredients for it to work:
Sharpcap Pro, ASTAP (plate solver), & R (r-project.org)

Then install the practical math library (pracma) in R. This is likely the worst step for technically non-inclined people.
You open the R console by running R.exe and install the library by typing: install.packages("pracma") (and confirm with Enter)

Last but not least you need to unzip my script to a convenient location and adapt the paths of the no_polaris_no_problem.py python script.

You can then run the script in Sharpcap. The "readyness" button will tell you if everything can be found aka if the paths are correct.

Kind regards,
Norman
zerolatitude
Posts: 51
Joined: Mon Mar 01, 2021 5:24 am

Re: Polar Alignment without Polaris (workflow)

Post by zerolatitude »

Thanks. Will try it out.

Edit. The zip in your post above v040 only has the python file. No readme. Am I looking at the right version?
Post Reply