Python script for polar alignment using TSX
The aim of the script is to get you close enough to polar alignment that after running a Tpoint model, the star you choose for accurate polar alignment remains in the field of view.
IMPORTANT:
For Linux users, the install script will copy the files into /usr/local/bin and the icon files into /usr/share/pixmaps. Before running the install script, you MUST edit the PAUI.py file - otherwise the data for my setup will be stored in the script.
The install script will also create a desktop icon and put the script into the Education menu.
For Mac and Windows users, you can run the tool by typing: python3 PAUI.py
Before use, edit the following pieces of data near the top of the PAUI file:
Image exposure data:
CAM_DURATION: how long you need for each picture. This should be just long enough to plate solve. It will also determine how long between updates of your alignment.
CAM_BINNING: the binning you require for plate solving. You can generally get away with a highish bin, and that speeds up downloads and frequency of upates of your aligment.
CAM_SCALE: the arcsec/pixel of your binned image - used for platesolving
CAM_FILTER: set this to the name of the filter if you want to specify one for platesolving e.g. CAM_FILTER = "Lum". If left set to "", will not attempt to set the filter, so leave this blank if you don't have a filter wheel.
Image location data:
PA_DEC: the script takes two images at the same DEC to work out the polar alignment. Chose a DEC where you can see the sky from your location, but also sufficiently far from the pole that the locations have a good separation on the sky. Choosing e.g. DEC of 80 will mean that the locations are too close to get an accurate estimate of your alignment, though the routine will try!.
HA1 and HA2: the hour angles for where to take images. Make sure these are on the same side as the meridian if your scope does a meridian flip. Make sure that the (HA2, DEC) position is NOT close to the zenith, nor close to the east-west axis. The script tries to evaluate rotations around both these axes and if you are close to the axis, it is hard for the script to work out what the roation should be (if you are exactly on the zenith for example, any azimulthal rotations would not change the position of the image, just its orientation). The (HA1, DEC) may be close to the zenth or east-west axis.
Aim of the script
The aim of the script is to get you close enough to polar alignment that after running a Tpoint model, the star you choose for accurate polar alignment remains in the field of view. Suggested workflow is:
- Disable tpoint pointing corrections
- Run the script
- Re-enable tpoint pointing corrections
- Run tpoint (re)calibration model
- Use accurate polar alignment.
Since the script and Tpoint only need to be able to plate solve, this procedure can be carried out before nautical dusk.
In practice, I have found that starting from a few degrees from polar alignment, I only try to get within about 10 arc mins of the pole. If I start the alignment again at this point, the difference from the previous run is again about 10 arc minutes, but the sign of the change is often different, an indication of the likely size of the error. However, if I then polar align and try to get close to zero starting from from this much closer location, I can then get to within 2 arc minutes of pole as confirmed by Tpoint, so you can use the script to get accurate alignment even without Tpoint.
Running the Script
Before running the script, you must disable the Tpoint pointing corrections. Otherwise, the script will try and get you bac to your previous polar alignment point.
You must also have TSX running and have enabled the TCP server (under the TSX tools menu). The script will tell you if this is not the case.
After running the script, click the Start button. This will first remind you to disable the Tpoint corrections.
Once you have confirmed that the pointing corrections are disabled, the script will then attempt to take two images at the Declination and hour angles as specified above. From this, it can calculate the position of the telescope axis and will indicate how much to raise or lower and rotate the mount. Only use the altitude and azimuthal adjustment mechanisms. The rotation follows the Tpoint convention - it is clockwise or anticlockwise as seen from above your mount.
Do NOT move the telescope position - just let the mount track normally.
The script will then take and platesolve regular images to update the position. If you move adjust the mount, it is likley that one or two images cannot be plate solved - just wait for a couple of seconds and this will be resolved (unless you are unlucky with clouds, but...)
You do not need to try for perfection - anything within 5 arc mins is likely to be good enough for imaging if you are guiding - polar alignment only prevents field rotation for guided images. Unguided imaging may require better alignment, but your highest accuracy will be provided by Tpoint and the standard TSX accurate alignment routines once this script has you close enough.
Once you are close enough, click "Stop" - the script will complete taking its final image and then stop (this prevents TSX from crashing).
You can click Start again if desired - this will create a fresh measurement of your polar alignment, but do not expect this to be exactly the same. The script does not account for e.g. scope flexure and refraction which will limit the accuracy that can be achieved.
Clear will clear the current display text.
Changelog
V 1.0 - Initial release
V 1.1 - Tidied up interface. Added scrollbar and clear button.
V 1.2 - Added the abilty to define a filter to be used with plate solving. Now removes the image and SRC files by default. Binning and filter wheel positions are restored after the run