I have been using SGP for about three years and have never had a problem with meridian flips. The last three nights I’ve been having problems.
I watched the imaging closely last night. I was taking 10 minute subs. When sub 13 was saved there were 8 minutes remaining until the meridian flip was scheduled. No flip occurred and Sub 14 started. Unfortunately, PHD2 thought it was now on the west side and ran away from the guide star. SPG parked the telescope. Here is a snippet of the log starting with the successful save of sub 13. I have also attacxhed a picture of my “flip” options.
What are your settings on the Telescope tab of the control panel? That is where the flip is set up. What you are showing just causes a center after the flip (if any).
Well, the flip didn’t happen because you told it not to.
Wait For Meridian is unchecked. This means if the mount isn’t to the time to flip, SGP will start the next frame…regardless of how soon the flip comes. If this is checked, then if “time before flip” < “exposure time of image”, SGP will wait for the flip, perform it, and then start imaging again. Since you state “8 minutes remaining until the meridian flip was scheduled”…SGP started the next frame.
Minutes Past Meridian to Flip is 8. Line 56 of your log shows this point had not been reached.
PHD appears to have “run away from the guide star” during a dither settle…not sure why that happened. I don’t THINK SGP ever tells PHD (or should have to) anything about “Scope is on this side”…that should, if memory serves, be the mount driver’s job, if anyone’s. Your log snippet suggests the same thing…doesn’t appear to be any signal form SGP that PHD should be doing anything different, or that mount side of pier has changed.
Frankly, I have a hunch that SGP not flipping is a bit of a red herring…true, it didn’t flip…but it had been told not to yet. So, it appears to have been cruising along quite happy, well aware of where everything was and what it was doing.
PHD log for that time period might prove helpful, to see if perhaps the mount driver told it something erroneous?
As for whether the Reverse Dec box should be checked…hard to say, since you don’t say what mount you’re using. That setting’s pretty mount-dependent ime.
Exact same issue happened to me. Only change from my last session was I updated PhD to the latest version.
I had to recalibrate after the meridian flip to keep tracing on the guide star.
In almost all cases I would recommend leaving the “Minutes past meridian to flip” at 0 unless you really have a reason/need to track past the meridian. The reasoning is that the longer you track past the meridian the less “wiggle room” you’re going to have before your mount stops tracking or your gear and the tripod get “friendly”.
The earlier you can flip the more likely it’s going to be that you’re going to be successful. Also if you have a very narrow margin for error with your pier/tripod then the “Wait For Meridian” option is needed. This will stop SGP from starting a new image if it will complete prior to the mount passing the meridian. Yes you’ll lose some imaging time but you’ll flip right at the meridian rather than sometime around it.
The point that PHD2 ran away when the mount had tracked past the meridian but had not flipped makes me wonder how PHD is determining that the flip has happened.
What should happen is that PHD checks the pointing state from the mount by reading the PierSide property and uses that to determine what dec correction direction to use. But if it’s using the hour angle to determine the side of the meridian then it could start changing the dec correction direction before the flip.
And of course the mount must report the pointing state correctly to both PHD and SGP.
I suspect the PHD2 run away was not related to the meridian flip. On two occasions, I have seen a guider run away and both times it appeared to be associated with a dither.
@Jared - in earlier posts, we previously came to the conclusion that flipping on the meridian could be unstable, on account of the various devices/software not quite knowing which side they were on. 5 minutes or so is safe and ensures that all the device status’ are on the same page?
Yes, I would agree with that. I think in most cases that setting the “Minutes Past…” to 0 will actually get you that. Setting it to a smallish number (like 5 minutes) should guarantee that the mount and PC were in agreement that you were past the meridian. It’s always best that the PC and Mount time be exactly the same where possible though. If the mount allows the time to be populated from the PC this helps a lot.
