Is the sky map set for equatorial coordinates instead of azimuthal? To try, go west or east (press w or e) and verify the orientation of the horizon. If it is horizontal and close to the bottom of the sky map then you are in azimuthal coordinates. If not then you're in equatorial coordinates. You can switch between the two by pressing the space bar.
Wouter van Reeven
ASI6200MM and 7 slot 2" filter wheel with a SkyWatcher Esprit 80 ED on a SkyWatcher HEQ5-Pro
ASI1600MM-Pro Cooled and 5 slot 1.25" filter wheel with an 8" TS Ritchey-Chrétien on a SkyWatcher EQ6-R
Did You create this yellow frame with the "FOV"-Symbols (-> Settings/FOV Symbols/Edit)? Then it has nothing to do with the real rotation angle of the sensor, e.g. camera! This you will get by taking a shot in the "align"-modul in EKOS doing a "Capture&Solve" with the option "Slew to Target". With all the correct configuration details regarding the plate solving (astrometry.net) You will receive a so called "Solver FOV" (white frame!) reflecting the real rotation of the sensor.
Od course You can rotate the mentioned FOV-Symbol by entering an new rotation angle in the "Edit"-window of the symbol.
I'm writing this reply, because I myself stumbled upon a inconsistency in the usage ot the rotation angle. As You see in the attached screen shots, I got a rotation angle of ~ -180° in the Align-module with a "Load an Slew" whereas I had to enter a value of 0° for the FOV-Symbo to get the same rotationl!! The figures are distorted by exactly 180°! Seems to me like a wicked bug in the programming.
AOK Skywalker DDM
Celestron Edge HD 14"
Starlight Trius SX35
Starlight Lodestar X2
Starlight SX Maxi Filterwheel
Focus Boss II Focuser
KStars Ubuntu 20.04 LTS on NUC
Thanks for the replies, escriban and bart.
I did not create that reticle, nor the fov. That's the automatic reticle (the yellow one, I mean) that shows up when the camera is connected and detected. I supposed that always would show the real and supposed camera rotation.
I noticed about this rotation issue (or misunderstanding from my side) last night. My batteries ran out and I had no more chances to try with the real cam; only with the sim.
The solver fov (when done the plate solving), as escriban says, actually reflects the current camera rotation. That's ok. But my doubt is why (without plate solving) kstars apparently does not show a proper camera rotation: if I'm pointing at the ncp and move only declination axis to south (that is, going to east), the camera rotation should move along with this movement with the proper rotation, and not always making the sensor pointing (hence, rotating) to the ncp: