Use target coordinates for auto-center after pier flip


#1

I thought someone already requested this but I can’t find it.
It would be very helpful to have the ability to skip the initial, pre-flip plate solve and use a target’s coordinates to re-center after a meridian flip, rather than first doing a plate solve to get the current coordinates before a flip, then flipping and centering on the solved coordinates.
If dithering in the current way, the solved coordinates will be slightly different from the target’s coordinates so the re-center will always be off some. Also, for mounts like AP’s that can be set to stop tracking when they hit a meridian limit, the plate solve always fails since the mount isn’t tracking anymore. Subsequent plate solves in Recovery Mode will also fail since the mount is no longer tracking.
Without the initial plate solve it wouldn’t matter if the mount has stopped tracking.


#2

I guess I don’t understand. I use an AP 900. I am fairly new to SGP and do not pretend to fully understand all functions, but so far as I know, the initial plate solve (PS) before a meridian flip is simply to assure PlateSolve2 (PS2) and the mount have the best position information before flipping. After the flip it does a PS to find out where it is after the flip and then does subsequent PSs to center the target. So far as I know, centering is based on the target position you input and is not affected by dithering at all. Of course, how closely this matches the actual position will depend on the tolerance limits you input. For example, my tolerance is 25 pixels, so once it gets within 25 pixels, SPG declares victory and resumes the sequence. With my mount this takes one additional PS. So worse case, I suppose, is that I might be as much as 50 pixels off of my pre-flip position on the chip. I would also guess that the tighter the tolerance, the more PSing might be required to satisfy PS2.

BTW: I am fairly sure that SGP always does an initial PS before centering a target, meridian flip or not. I am guessing that PS2 need this information before the initial slew. Otherwise, how would PS2 know where to slew? Am I wrong? if so, someone should correct me.

Question: why would you wish to stop tracking at the meridian if you are still working on a sequence?

Ed


#3

Ed,
I believe the way SGP currently works is that the plate solve before the flip is to determine the coordinates to slew to, instead of slewing to the target coordinates defined in the sequence. To be honest, I’m not sure why it does this, i.e., why anyone would want to slew to something other than the pre-defined coordinates, but I’m sure there is a reason. I’m asking for the ability to skip the initial plate solve before the flip, and then use the target coordinates to do the after-flip centering.
AP mounts have the ability to track past the actual meridian, but you need to tell them how far to go. APCC allows setting that limit for multiple DEC values since the mount can go further past the meridian at some DEC angles than at others. APCC also allows you to specify what happens when the limit is reached. In my case, I have it stop tracking so the scope doesn’t run into the pier. APPC in effect is fooling SGP into thinking the meridian is past the actual meridian, per the limits the user set.


#4

Hi Eric,
I’ve also wondered why SGP uses a plate solve rather than the given coordinates. I can think of a case, however, when using PHD2 comet tracking, where the FOV might well have drifted considerably off the original coordinates. Plate solving just before the flip would keep the comet centered whereas using the original coordinates would cause the comet to be offset in the FOV for multiple exposures. An example is 6 5-min exposures of a comet to be stacked to make a 30-min equivalent exposure.

Regards,
Kent


#5

Kent, good example.
Adding the option as I requested would allow for both conditions. Comet trackers would use SGP as it currently is, and people like me would check the box to skip the initial plate solve and use target coordinates.
Eric


#6

The primary reason is because you may recenter and not update your coordinates. This also means that your current position may be past the meridian but your target has not passed the meridian yet which gets a little tricky for certain mounts (AP being one of them)

We’ll consider changing it to recenter on the target. I think that makes sense assuming you have the coordinates populated.

Thanks,
Jared


#7

Jared, it would probably be best to allow the user to specify to re-center on the current coordinates versus re-center on the target coordinates, rather than just changing it. This way it won’t break anything. It’s also important that the re-center on target coordinates NOT do an initial plate solve - in the case of AP mounts (and probably others), the mount may have already stopped tracking.
Eric


#8

Right, in that case the plate solve wouldn’t be needed as SGP would just slew to the target coordinates. We use the plate solve to determine where to slew to as getting the coordinates from the mount is very inaccurate after guiding.

Thanks,
Jared


www.mainsequencesoftware.com