Sure seems like PS2 is not getting the correct hints and it is not until ANSVR solves and syncs once then PS2 is able to pickup from there. I’m wondering if where the mount/scope thinks it is pointing is actually to far away for the hints that PS2 needs. The reason I’m thinking this is after ANSVR solves, not needing any hints, and now the mount is sync. Once this first sync is done the hints are close enough for PS2 to finish, I wonder after your slew how close you really are to the target? Maybe your Polar Alignment is to far off so when you slew you are not close enough for PS2 to solve from the hints given to it.
This is not atypical and is really what the blind solve failover is designed for… Innacuracy in the slew, polar alignment, mechanical imperfections, encoder issues all contribute to a mount reporting a location it is not really at after a slew. This, of course creates a situation where location hints are out of tolerance for PS2 and it fails the solve.
As for the scale creep, that does not appear to be a big thing… scale can shift very slightly due to rounding errors in the plate solver’s math. If we saw a significant change in scale, that would be a different story…
So… I am still not entirely sure what is going on…
Does PS2 get a rotational cue before solving? If so, is that shown in the log? If it doesn’t get a cue, then I think you must have it right, Mark and Ken. It’s possible that alignment has gotten a bit off (though I"m permanently set up), so I’ll recheck it next session.
Thanks Ken. So if it doesn’t get a cue, it’s hard to figure out what’s going on. Tonight when I started up, PS2 solved without trouble, then the rig slewed to target. Here is what happened:
Slew 1: PS2 solved again. Center is off about 100 arcsecs and rotation off about 8 degrees, all perfectly normal so far
Slew 2: SGP adjusts the rotator, then PS2 fails, although it’s only a few arcsecs (and 8 degrees) off the first solve. Blind solve succeeds.
Slew 3; PS2 succeeds. Still a tiny bit off target, but rotation is spot on now.
Slew 4: PS2 fails, blind solve takes over and succeeds.
If blind solve succeeded, why is PS2 failing when the scope has barely moved?
Dang - you guys were right. My pointing model was about a light year off. I have no idea how that happened. PS2 looks much happier now.
I still need to keep an eye on the rotator, though. After my last session it started to do a cordwrap andI had to power down quick. It’s always something, isn’t it? Thanks for the help.
Rotator cord wrap can be a challenge to solve. The Optec rotators have a max position to which they will rotate before reversing position. So the goal is to put your rotator in that max position and then adjust the cord lengths to allow the rotator to move to that max position without pulling the cords.
When homing my Gemini rotator, it will rotate counter clockwise to find the homing sensor at an instrumental position of 75 degrees; then reverse direction stopping at 0 degrees. So I move the rotator to a 74 degree position and then adjust the length of the cords to fit. I know the rotator will never move past that point and will not end up wrapping the cords.
When homing my Pyxis LE rotator, it rotates counter clockwise untill it finds the homing sensor at 180 degrees; then it moves CW stopping at 0 degress. So, again, I move the Pixis LE to 179 degrees and adjust the cord lengths.
A little experimenting in the day time can help clarify exactly what is going on. It is important to test the full motion of the rotator. For example, with the Pyxis, move the rotator to 179 degrees; adjust the cord lengths; then command it to move to 181 degrees. This forces a 358 degree rotation allowing you to see how the cord behaves throughout the move. Also, do this test with the scope pointing at the zenith – cord behavior changes dramatically between horizontal and vertical rotation.
One note - after homing be sure to set the Sky Position Angle (Sky PA) to also be 0 degrees so that the readout is showing the internal or instrumental position angle and not some random Sky PA.
Thanks for all the details, Charlie. With the Pyxis 2", Is there any chance that the home position can get reset somehow, maybe through some interaction between platesolving,SGP, and the Pyxis ASCOM driver? I followed just the procedure you laid out when I did my wiring. But somehow the rotator exceeded those limits. The only thing I can think of is that that home position somehow changed.
I believe the homing position of the Gemini and Pyxis LE rotators is detected by the use of a small, magnetic sensor that is mounted in a fixed position; so the home position and the resulting 0 position cannot be altered even by a bug in the software or other glitch. I assume the 2" Pyxis is the same.
The park position of the Gemini rotator can be set anywhere you want it to be but the Pyxis LE has the park position hard coded to be at the 0 instrumental position angle. I suspect the 2" Pyxis is hard coded to be 0 as well. However, you might look through the Optec Pyxis control app to see if you can find a way to change the park position.
At power up, my Pyxis LE will always do a homing operation and that can’t be disabled. At power up, the Pyxis LE will begin rotating CCW looking for the home sensor. If the rotator happens to be just CCW of 180 degrees (190 degrees, for example), the rotator will go all the way to 180 in a CCW direction – potentially 359 degrees of rotation. This can be a problem with cord wrap. So, my policy is to always “park” my Pyxis LE at the end of the evening. Then at the next power up, the rotator will move from 0 to 180 CCW and then back to 0.
Note to Ken and Jared: at end of sequence, it would be nice to be able to park the rotator as well as park the scope before disconnecting the equipment.
!!! I had no idea it could go over 180 degrees! I’m sure that’s what happened. And yes, agree 100% that including a rotator park at the end of the session is the way to go. I’ll put in a feature request for that. Thanks very much, Charlie!
Charlie, is there a surefire way to prevent cordwrap using SGP and the Pyxis? As long as it has one point on the circle that it will never cross - even across multiple nights and multiple targets - I’m fine. Is there something I need to do to make sure that it won’t cross that point?
I’m assuming my Pyxis LE behavior is the same as the 2" Pyxis. The magnetic sensor doesn’t move, of course, and everything is relative to that. The home function performed at power on finds that position and then returns the rotator to instrumental position 0. I would suggest you also make the homing position be Sky PA of 0 as well and do not attempt to use Sky PA in the Pyxis Commander. Do all your rotator moving in SGP.
In the Pyxis LE “Advanced Setup” dialog, you can set rotator data there, as well, but again, I recommend leaving them at factor default. The “reverse” option tells the rotator which way to move when commanded by a controlling app (SGP). Unfortunately, I don’t remember my setting off the top of my head. For example, if the rotator is reporting 90 degrees to SGP and in SGP you enter the command to move to 100 degrees, the rotator will move 10 degrees BUT when SGP does a plate solve it may show 80 degrees. The rotator went 10 degrees but in the wrong direction. If you see that happening, you will need to change the “Reverse” option on the Advanced Menu.
So, the rotator will home at power up and leave the rotator at instrumental position angle 0. If you now connect to the rotator in SGP, SGP will show a position angle of 0. At some point before starting your sequence, you need to do a “Solve and Sync” to get an accurate DEC and RA and to get the rotator’s current Sky PA. As soon as the solve and sync completes, you will see SGP change the rotator’s position to match the plate solve’s Sky PA. SGP now knows how to map the camera’s Sky PA to the rotators instrumental PA. For the rest of the night, only move the rotator from SGP or simply let SGP manage the rotator position by itself.
Note, you can “park” the rotator from SGP by using the “Zero” button. This moves the rotator to instrumental position 0 – same as doing a park.
To test the “reverse” option, do the following:
After connecting to the rotator in SGP, you will see position 0.
Do a solve and sync to set the current rotator position to the Sky PA of the camera.
Enter a value in the SGP rotator position box that is 10 degrees bigger than the current Sky PA
Click “Set” to move the rotator.
Do a frame and focus frame
Right click and select plate solve
Does the plate solve show the expected Sky PA? If it is off by 20 degrees the rotator moved in the wrong direction. This means the current reverse option is incorrect and needs to be flipped.