[Kstars-devel] Possible issues with catalog coordinates & determination of apparent coordinates in KStars

[Jasem Mutlaq]

Hello all,

Working on Ekos Alignment module under simulation requires highly accurate
sub arcsecond coordinates. I initially noticed that the while a star is
centered in the telescope simulator, it is not centered in the CCD
simulator generated frame. After some investigation, I found that the J2000
coordinates for some stars do not match catalog coordinates from either
stars.dat file or other sources like Simbad. So it seems the binary star
data got somehow corrupted or incorrectly encoded.

Furthermore, I noticed that even if I have the correct coordinates for the
stars, the process of coordinate transformation from J2000 to JNow results
in the wrong coordinates. Specifically, the SkyPoint apparentCoord function
have no issues correctly transforming the J2000 RA to JNow RA, but the
declination is always wrong, and this is what causes a vertical shift in
the generated image by the CCD simulator since it is being fed incorrect
declination coordinates.

Turns out that only performing precession while turning off both nutation
and aberration results in the correct actual JNow DEC. This is confirmed
from the CCD simulator (star is now fully centered horizontally and
vertically) and from other programs like Stellarium. Here are my
observations for a sample star (Vega):

1. Vega Apparent Coordinates in KStars (JNow):

RA     18:37:24      ====> KStars Detail Dialog
DEC  38:48:11      ====> KStars Detail Dialog

2. Vega J2000 Coordinates:

RA    18:36:56.52   ===> All

DEC   38:46:58      ===> KStars Detail Dialog
DEC   38:47:01      ===> stars.dat
DEC   38:47:03      ===> Stellarium
DEC   38:47:01      ===> Simbad

3. Vega Actual Apparent Coordinates (JNow):

RA    18:37:24.2    ===> All

DEC   38:47:48      ===> Stellarium
DEC   38:47:45      ===> Star centered in CCD Simulator

4. KStars Precession, Nutation, and Aberration:

RA    18:36:56.52
DEC  38:47:01

+ Precession Only:

J2000 DEC 38:47:01 ===> JNow DEC 38:47:45 (Matches actual apparent
coordinates)

+ Precession & Nutation

J2000 DEC 38:47:01 ===> JNow DEC 38:47:52 ( +7 arcseconds more than actual
apparent coordinates)

+ Precession & Nutation & Abberation

J2000 DEC 38:47:01 ===> JNow DEC 38:48:10 ( +25 arcseconds more than actual
apparent coordinates)

Can anyone shed light on what's possibly causing this behavior?

--
Best Regards,
Jasem Mutlaq
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kstars-devel/attachments/20131017/b372cf8d/attachment.html>