stevesp At the end it tried to move west a bit
Steve (?), this is interesting.
The reason is this: ASIAIR does not really perform Meridian Flip. The mount is the one that does the meridian flip.
With pretty much every German mount that can perform GOTO, the mount has to know which side of the pier the declination axis is located. The mount will try to maintain the state where the declination axis is on the opposite side of the pier from the star it is asked to perform a GOTO.
What ASIAIR does with an "Auto Meridian Flip" is simply to choreograph (I don't know what other word to use...) the meridian flip. I.e., while the mount is tracking a star, and the star has reached the time "Stop tracking x min before Meridian" (a parameter in the Mount Setup), ASIAIR pauses AutoRun, it then stops autoguiding, and also stops tracking.
Meridian is basically the point where the RA of the star is equal to the Local Sidereal Time (LST). In general, ASIAIR stops a little before this time, to make sure the exposure time of a frame does not go past this "Stop tracking" parameter.
After stopping everything, ASIAIR does a countdown until the time when the second parameter in Mount Setup called "Do AMF x min after Meridian" is reached, i.e., ASIAIR twiddle its thumbs until the RA of the star has reached LST plus this second Mount Setup parameter.
When this second time is reached, ASIAIR sends a GOTO command to the mount, with the assumption that the pier side of the star is now also the pier side of the declination axis!
If the mount also thinks that the star and declination axis are both on the same side of the pier, the mount will execute a Meridian Flip -- i.e., it moves to the same RA, declination coordinates, by traversing through the pole, but repositioning the pier side of the declination axis.
After the mount has stopped moving (the state that ASIAIR can get from the mount's communication protocol), ASIAIR takes an image to plate solve. If the ASIAIR senses that mount has meridian flipped (probably by looking at the camera angle from the plate solve, although many mounts also have a command to fetch the pier side), it centers the star on the plate, and then resumes tracking, then flip the auto guiding calibration (that flip switch in the calibration setup info in the autoguiding window). It then chooses a new guide star, then start autoguiding. It finishes the choreograph by resuming the AutoRun sequence. (What ASIAIR has to do is "As Easy as 1,2,3." The Mount is the one that has to compute the actual path to move the RA and declination motors.)
From the above, you can see why I had called this a choreography :-).
I believe what is happening to you is that when ASIAIR issues this GOTO, the mount does not think the star has yet moved to the other side of the meridian. The RA axis moved west by a little bit (good observation!) because tracking has earlier been stopped, so the GOTO would move the mount "normally" to make up for the tracking time loss since the star has drifted by a little bit from where it should be had tracking not been turned off.
All indications is that the LST and/or RA in the ASIAIR is not in sync with the LST and/or RA in the mount.
Unfortunately, while mount's LST is available in hand controllers, you cannot find the LST that the ASIAIR GUI, so we cannot tell if LST is really in sync. However, LST should be directly related to UTC and local Longitude.
The next time the mount refuses to Meridian Flip. Check to see if the UTC time and Longitude agrees with the UTC and Longitude on the hand controller. There are a number of iOS apps that will display the UTC in the operating system, and ASIAIR should be using the same system call to obtain the UTC. I have even found a lot of iOS apps that will display LST.
There are a number of people who refuse to sync their mount to the ASIAIR time and longitude. This typically is the cause for time and longitude between the ASIAIR and the mount not being in sync. But this usually creates only a few minutes of time difference between Meridian crossing, and ASIAIR will retry the GOTO (after a 1 minute timeout) over and over until the time on the mount has caught up and it performs a Meridian flip.
But, if the mount happens to be off by an hour (for example, if the Daylight Saving Time is turned on in the Mount -- recall that ASIAIR does not observe DST) you may have to wait an hour before a Meridian Flip occurs.
A good sanity check is to try to perform a GOTO to a star from ASIAIR (using coordinates obtained from a planetary program like SkySafari) that is as close as possible to Meridian (or just enter the RA of the Meridian in the GOTO panel), then use a bubble level to check if the OTA is pointed towards the Meridian. You can do this in daylight. (I have one of these cheap Amazon digital inclinometers that I often put to use for such purposes -- pointing at Zenith is one way to align a Takahashi mount without needing to use stars).
If you have to wait a long time for a Meridian Flip, it is very likely that there is a longitude or time disparity between ASIAIR and the mount. I actually expect a lot of that happening this week to US hobbyists, since we just went into Daylight Saving Time just a day ago, and then next Sunday night in Europe :-).
Incidentally, a mount slew (the directional keys in ASIAIR and hand controller) will not obey the pier side rule, you can slew a mount right through the meridian (until the optical train hits the pier :-). The mount will only obey the pier side rule when it is commanded to do a GOTO.
Chen
