When the software performs the meridian flip, for whatever reason, the dome takes a long time to react, and get into the right place (DDW). Generally, since the software keeps increasing the exposure looking for a plate solve, the dome arrives in time for the software not to give up.

But I am wondering if there's a way to build in a delay of, say 90 seconds, after the flip, before the software starts the plate solve, to give the dome a chance to catch up.

Please post a log.

John, the same log I posted yesterday on "Too many guide stars" has it, at about 20:33. I watched it, as I have several times, as the dome rotated in time for the second or third attempt.

This was an object that was at an altitude of almost 90 degrees at the flip, so the dome rotating seems to be very sensitive to a very small movement by the scope, especially when there's a "German offset" of 23 inches.

DDW depends on TheSky writing to a text file, from which DDW determines the dome azimuth. TheSky reportedly updates this file every three seconds. Since you have the mount settling time set at 3 sec., you may want to try increasing the mount settling time to 10 sec. to see if that helps DDW get "in sync" with the mount pointing.

Of course, AutomaDome, due to its integration with TheSky, will keep both in sync automatically. I know some users are quite successful using AutomaDome.