Note: With clever choosing of pointing direction and some spherical geometry, the polar alignment error on both axes can be
calculated from just two frames! We hope to include this feature in a future version of AstroTortilla.
This is indeed what I'm doing by hand. I record two plates with constant declination. Then
using Iris or nova.astrometry.net, I get RA, DEC and orientation. Using an home made excel sheet, I determine the direction of the rotation axis between the two plates i.e. RA axis direction.
Is anyone interested in the math?
Eric
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I looked back at my excel sheet (attached). It consist in three steps:
- determination of the rotation axis between two pictures at J2000.0 epoch (because plate solving was done at J2000.0 in Iris)
- conversion of the rotation axis from J2000.0 to current epoch
- converting from RA/Dec to Az/Alt.
All the calculations are on the second page of my sheet.
The first step mostly involve rotation matrices. (All the 1 are to check that there are no mistakes in formula). It ended with arctan to get RA (line 37) and Dec (line 38) in rad, deg and finally in time/deg (column E) all in J2000.0.
Then, using precession, a convert is done from J2000.0 to current epoch (column F).
Lastly, rotation axis direction is converted from RA/Dec to observer coordinates using true pole direction as reference.
Eric
PS :
Precession is calculated using simplified formula.
Nutation is not took into account.
Atmospheric refraction is not took into account.
The documentation about Polar Alignment tells:
Note: With clever choosing of pointing direction and some spherical geometry, the polar alignment error on both axes can be
calculated from just two frames! We hope to include this feature in a future version of AstroTortilla.
This is indeed what I'm doing by hand. I record two plates with constant declination. Then
using Iris or nova.astrometry.net, I get RA, DEC and orientation. Using an home made excel sheet, I determine the direction of the rotation axis between the two plates i.e. RA axis direction.
Is anyone interested in the math?
Eric
Yes we're interested. We've reviewed a few nasty algorithms but have not implemented anything yet. Any insight would be good.
Lauri
I looked back at my excel sheet (attached). It consist in three steps:
- determination of the rotation axis between two pictures at J2000.0 epoch (because plate solving was done at J2000.0 in Iris)
- conversion of the rotation axis from J2000.0 to current epoch
- converting from RA/Dec to Az/Alt.
All the calculations are on the second page of my sheet.
The first step mostly involve rotation matrices. (All the 1 are to check that there are no mistakes in formula). It ended with arctan to get RA (line 37) and Dec (line 38) in rad, deg and finally in time/deg (column E) all in J2000.0.
Then, using precession, a convert is done from J2000.0 to current epoch (column F).
Lastly, rotation axis direction is converted from RA/Dec to observer coordinates using true pole direction as reference.
Eric
PS :
Precession is calculated using simplified formula.
Nutation is not took into account.
Atmospheric refraction is not took into account.
Attached the two pictures using in the previous table.
Do you know if astrometry.net use current epoch or any other fixed one?