stevesp restarted autorun with targe RA 5h40m51s DEC -2°52’33”
It is important to keep in mind that when you autoguide, the field rotation will be around the guide star, the field does not rotate around the center of the FOV of the guide camera, nor rotate around the center of the FOV of the main camera. With multiple star guiding, the field will rotate around the weighted centroid of the stars that are selected for guiding.
Once you understand this last paragraph, you can look for why autoguiding had dragged the main camera's plate center away further once you started to autoguide.
That being said: notice a couple of other things:
(1) after a meridian flip, your RA is off by about half a minute of arc.
Remember what I said earlier about needing to have the mount perfectly level to ground, or doing a multi-star alignment, to tell the mount where it really is pointed? If you want this error to be smaller, your mount's hour angle must be aligned to the sky's right ascension to at least this accuracy.
As I also mentioned in the earlier post, this can also be caused by bad polar alignment. If the target is close to the pole, for example, each arc second that the polar alignment is off by will also cause the meridian flip to err by the same amount.
Just visualize the problem in your head, or draw it out on paper -- imagine that you have two spherical coordinate systems whose poles are not coincident; one spherical system is the sky and the other spherical system is you mount's axes.
Even if you are perfectly polar aligned, but the east-west level is incorrect, the two spherical coordinate systems will be rotated from one another around the coincident pole.
If you see an RA error after the flip that is larger than you can tolerate, you need to fix this by better polar alignment and making sure the mounts zero hour corresponds exactly to the Meridian of the sky. With mounts with good encoders, you just need to level the mount (east-west). With cheaper mounts, you need to do multiple star alignment.
(2) notice that autoguiding further pulled the center of the FOV of the camera away? See the first paragraph above. If you want this error to be small, do a plate solve of the FOV center of your main camera/main OTA. Then another plate solve for the FOV center of your guide camera/guide OTA. If there is a difference that is too large for your tolerance, you need to get this number down by trying to point the guide camera better to where the main camera is pointed at.
If the two cameras are not pointed at the identical spot, over time, the field will be rotated around some location in your guide camera's FOV. If the two cameras are not pointed to the same place, the field rotation can be quite large. Your guide camera may be rotated by a small amount, but because the main camera is pointed elsewhere, the field rotation is execrated. Just calculate what the field rotation looks like over time for the difference of FOV centers that you measure. The larger the misalignment between the guide scope and the main OTA (even including any tilt in the guide camera), the faster the main camera's image will drift away, even though your guide star is perfectly steady -- the guide star may not even be in the FOV of the main camera if you are off by a lot. Over a few hours of imaging, that long term drift will cause your plates to not be aligned, even though the effect on each plate is small enough not the be noticeable with shorter exposures (i.e., it may not be visible with 2 minute exposures, but two plates taken 2 hours apart will show a dislocation that is 60 times greater).
(3) depending on "third axis balance," if you don't have a sturdy tripod or a sturdy platform that the tripod sits on, your mount could have also physically moved as the center of gravity moved from the right side of the pier to the left side of the pier when you perform a meridian flip. But I assume you have already made sure there is no third axis imbalance, for other reasons. It is insufficient to attain balance with the OTA parallel to the ground -- that is insufficient to assure balance for all positions of the OTA over the night -- or, in this case, before and after a Meridian flip.
Chen
