I looked into the time settings being potentially the problem but I don't think this explains the persisting "wrong pierside" alignment problem at least for my Atlas Pro mount.
That mount has not seen a hand controller in over 18 months, it has been solely controlled by Indi and Ekos, and Ekos is set to update time and date on the mount. Also, my mount is in perfect Park position, with the counterweight shaft straight down, no deviation from vertical that I can detect and the DEC axis is also set straight ahead at 0 degrees. I cannot possibly do any better. This is also the position in which polar alignment is started and thus the initial sync position of the mount. So the mount should know exactly where it is when it starts out slewing towards the first target.
Historically, the first slew of the mount to any target has been approximately 2 degrees off. Recently, however, i.e. sometime during the last 3 months or so, this has changed dramatically and now the mount is more than 11 degrees off target after its first slew and before first sync.
To emphasize again, no changes were made to the mount, the park position is exactly the same as it has been for almost 3 years now that I have had that mount, and there have been no updates of any kind to the mount using the hand controller, that has not been connected for the at least 18 months. So whatever changed there must have come from Indi or Ekos.
I am suspecting that as a result of this mismatch in the mount model, the mount ends up on the wrong side of pier when aligning to a target close to the meridian (M51 in my case for the last few nights), pointing West already with more than -30min of HA remaining before it passes the meridian. In other words, the mount thinks it sits a lot further East than it actually does.
The geographical coordinates for longitude and latitude are set correctly (derived from the location database in Kstars by city name anyway, not set by me), so the only explanation I can see is that the mount receives the wrong time upon which it bases where it thinks the meridian is and points to accordingly.
Local sidereal time in the EQMOD control panel is calculated correctly, however, but I am wondering whether the mount may be basing its HA calculation on [(Greenwich Sidereal Time) - (Time Zone)] instead of [(Greenwich Sidereal Time) - (Longitude W)]. It is too far off target for this to be explained by bad initial positioning of the mount, especially since the mount is sync'd during polar alignment. It seems to me that the mount must be receiving the wrong instructions for performing its initial slew to target.
The problem may not be apparent to observers with locations close to full multiples of 15 degrees (like Chris and Jasem), but for someone like me, who is sitting right in the middle between 90 and 105 degrees West the deviations in alignment and thus the propensity of the mount to collide with the tripod becomes very apparent.
Does this make sense or am I completely off base here?
Jo
PS: Here the relevant log parts: