CdC won't slew near meridian.

[dhmiller]

My workflow is to start a session with CdC, where I load EQMOD, then slew to my target. Then I jump into SGP and set up a sequence, then jump back to CdC to reslew, whereupon I would start PhDt2, then start the sequence. But it seems that CdC won't move at all if my target is too near the meridian - it does slews away from the target initially, but then starts to heard back to the target and stops well short, in the same position it was before I initiated the slew. The attached shows the position of the scope after the slew and the target. Any way to adjust this, if there is a limitation on how close to the meridian CdC will slew?  Or maybe there is another explanation? Thanks for any assistance.

[symmetal]

In Eqmod (or EQAscom) have you set your sync mode to 'Dialog Mode' and not 'Append On Sync'. Append on Sync can cause a similar behaviour to what you have, though it can be anywhere and not just near the Meridian. 

I'm curious as to why you don't let SGP do the slewing and centreing, as the Mosaic and Framing Wizard makes it very easy to do. As it uses EQAscom syncs like CdC does, you can see if the slewing behaviour is any different and so narrow it down as to what program's causing the error.

Alan

[dhmiller]

11 hours ago, symmetal said:

Thanks for the reply. I will check the EQMOD setting - thanks for the tip.  I have tried using SGP to slew and center, but it also does not move the scope when I enable those options - I believe it is supposed to move the scope when the sequence starts, if I'm not mistaken. Also, it always fails when doing the final stage of the plate solving,.. I don't believe it indicates a specific error other than "Fail."  It also will not start the sequence when I use the Frame and Mosaic tool because it seems to expect an auto-rotator.. I have tried mightily to find the Manual Rotate page, such as shown in the manual where you simply turn the camera yourself to satisfy the expected orientation, but I cannot find that anywhere in 4.2.0 (1158), so that is also an issue. All user error, no doubt, but that all leads me back to using CdC for the final alignment with the target.  

 

In Eqmod (or EQAscom) have you set your sync mode to 'Dialog Mode' and not 'Append On Sync'. Append on Sync can cause a similar behaviour to what you have, though it can be anywhere and not just near the Meridian. 

I'm curious as to why you don't let SGP do the slewing and centreing, as the Mosaic and Framing Wizard makes it very easy to do. As it uses EQAscom syncs like CdC does, you can see if the slewing behaviour is any different and so narrow it down as to what program's causing the error.

Alan

 

[symmetal]

A little bug is that the Framing and Mosaic tool by default expects a rotator to be present even if one hasn't been specified in the equipment list. When you click Create Sequence, on the next panel deselect the Rotate or Validate Camera Angle box as below. Your sequence will then start without the auto-rotator error you've been getting. An unfortunate side effect is this will then not display the camera rotation angle when it's centreing so before running the sequence do a manual plate solve which will display the camera angle.

I always manually slew to the first target and do a plate solve just to check the rotation angle and adjust it if necessary.

I assume you're using PlateSolve2 as the SGP plate solver. This can be a bit troublesome if you're quite a way off target of haven't edited its settings to make it more tolerant of star sizes. I suggest you use ASTAP as the plate solver. This is an option in the Control Panel's plate solve tab. You need to install ASTAP from here first.

Just download the two files indicated to the same directory and run the astap_setup.exe file. I assume you're using 64 bit Windows. If you use the suggested default directory then SGP will find it right away. This should eliminate your plate solve fail problems. Ensure you've set the correct camera scale (arcsecs / pixel) in the Control panel camera tab (below), or the plate solvers will likely fail. This may be why Platesolve2 is failing but I suggest you use ASTAP anyway.

Hope this helps. 🙂

Alan

Edited November 11, 2023 by symmetal

[dhmiller]

Thanks so much, Alan. Very helpful indeed. The only problem is that when I plate solve, I do not see the target, given that I am looking at only one step/frame (even with auto stretch enabled). So how does one determine the current rotation if the target is not actually visible?  Moving forward with your other suggestions (all your assumptions are correct ;-)). Thanks again. 

[dhmiller]

1 hour ago, symmetal said:

"A little bug is that the Framing and Mosaic tool by default expects a rotator to be present even if one hasn't been specified in the equipment list. When you click Create Sequence, on the next panel deselect the Rotate or Validate Camera Angle box as below. Your sequence will then start without the auto-rotator error you've been getting. An unfortunate side effect is this will then not display the camera rotation angle when it's centreing so before running the sequence do a manual plate solve which will display the camera angle."

 

-->And I assume I need to copy the database to the ASTAP folder and run the exe?  Doesn't appear to be in the dir at present after just installing ASATP...

 

[symmetal]

You can just run the database exe file from wherever it was downloaded to. As long as you've installed ASTAP, when you run the database exe it will ask where you want them installed, and will fill in the ASTAP directory, C:|Program Files\ASTAP by default, so just hit return. You don't need to manually run anything in the ASTAP folder once it's all installed as SGP should run it automatically when you click Solve and Sync.

2 hours ago, dhmiller said:

Thanks so much, Alan. Very helpful indeed. The only problem is that when I plate solve, I do not see the target, given that I am looking at only one step/frame (even with auto stretch enabled). So how does one determine the current rotation if the target is not actually visible?  Moving forward with your other suggestions (all your assumptions are correct ;-)). Thanks again. 

Glad it helped. 🙂 You don't need to see the target visually. Just as long as SGP can get a useable image. Using the Control Panel Plate Solve tab click the Solve & Sync button and it will take an exposure of length as set and binning as set. I recommend binning 2x2 so that a faster exposure can be used. Depending on your scope and the filter used the exposure will likely be around 2 to 5 secs for LRGB or around 30s for narrowband. When solved the current location will be displayed along with the camera rotation angle. If you want the camera image long side to be aligned with RA then you want the camera angle to be 0 or 180 degrees (it will of course rotate by 180 after a meridian flip anyway so being one or the other is fine. If it reports say 352 degrees then you need to rotate the camera by 8 degrees, or just guess it, and plate solve again to see how much closer you are. The direction you need to rotate the camera to increase the rotation reading depends on what type of scope and camera you have, so it's best to to do a small rotation at first and see if the resultant plate solved camera angle has increased or decreased. Then you'll now for the future. 🙂

When you select Centre Target, SGP will then keep running Centreing up to the number of times specified, or until the image centre is less than 10 pixels from the correct location, using the settings shown, though you can set then however you wish. Ignore the rotator error box setting, as SGP will too, once you've told it as I mentioned in my ptrvious post.

Alan

[dhmiller]

Alan, you are a gentleman and a scholar - thanks so much for your help. I will try this tonight, now that the sky is clear again. I was also reading on another thread that I may have too many points in my EQMOD sky map - have not cleared that out in a while, so perhaps that is another issue I need to tackle... one thing at a time, though. If you don't mind, I will check in tomorrow or shortly after I complete tonight's session.  And thanks again for your help.

[symmetal]

No problem. 😊 Yes I forgot to mention it's worth clearing out EQMod's pointing data before each session, although selecting Dialog Mode as I mentioned previously, helps to mitigate this. With platesolving, the pointing data saved in EQMod is more of a hinderance than a help. It's a pity there isn't an option to disable it. I went over to using GSS Server a year or so ago instead of EQMod and it works fine without these pointing model slewing problems. It also gives a nice 3D picture of where your scope is looking too. 😁

Alan

[dhmiller]

Hello - thought i might give you a bit of an update.  I changed my workflow to eliminate the slewing in CdC entirely. So now I unpark the scope and head over to SGP. Tonight I gave it Andromeda for a target and I drew the square in Frame and Mosaic to create a single sequence starting about two hours prior to Andromeda crossing the meridian. The sequence ran fine and finished, though Andromeda had just crossed when it completed. I wanted to reorient my scope, so I added a new event of 90 x 120 second exposures, opened up the Target window and selected slew and center, thinking the scope would flip around to the other side, but it did not move despite my requesting the slew and center command. So I parked the scope to home and then unparked, created a new sequence via typing Andromeda into the Frame and Mosaic window and again chose Slew and Center, but the scope still did not move. At this point, I was completely off the target, so I opened CdC and chose Slew to Andromeda and the scope did indeed move, but the first few frames I saw after restarting the sequence did not look at all like upside down versions of what I had already captured - for the first two hours, I could see Andromeda clearly in the mid center of the frame, but now I couldn't see it at all. And again, even after requesting that CdC slew to Andromeda, the scope position and the target were not synchronized, so I really don't know what I will be getting until I integrate everything overnight.

What I need to determine is the best way to continue to capture more data when my target transits. I'm not yet ready to try an automated meridian flip, so I need a reliable"manual" approach that will slew/realign the scope using SGP, which did a great job for the time prior to my stopping it. If you have any thoughts or if the above is not clear, please let me know. And again, thanks for all your help - seems like I am halfway to where I want to be.

 

[symmetal]

I generally always do a manual Meridian flip anyway and do what you did. After the target's passed the Meridian stop the sequence and select slew to target.

Using EQMod it always slowly slewed back to the Meridian and then did the flip correctly. However since I changed to GSS it does ignore the manual slew command when the sequence has passed the Meridian. I have to home the scope, unpark, then manually select slew and it now slews to the target correctly. It's probably a setting in GSS I need to change but I haven't really checked as going via home doesn't add much time.

I assume you're still using EQMod. Have you selected 'Dialog mode' as previously mentioned, and did you clear out the EQMod pointing data before your session began.

It's worth looking through the SGP log which should say what happened. Looking through the log, at the time you gave the slew command after it passed the Meridian, did SGP acknowledge the slew command and state  why it couldn't do it. If your pointing data seemed to have gone off at some point before you gave the slew command it's possible SGP didn't think you had crossed the Meridian yet. After you gave the home command did the scope actually move fully to the home position. It's likely it didn't and ended up some way off.

In the SGP Help menu there's a 'View log' option and an 'Open log folder' option if you want to check previous logs.

When it failed to slew initially it would be worth doing a manual plate solve and sync. This may fix your problem, or SGP will likely say 'Unable to sync' or similar. It says this when the manual sync data is significantly different from where EQMod reports the scope is currently pointing.

Alan

[dhmiller]

Thanks, Alan.I probably missed a step along the way . Will check all tonight. I wonder if the target was just too close to the meridian...

Also, still struggling a bit with camera angle. I watched a video by Chris Woodhouse that said to enable the rotation alignment toggle even to use manual rotation to properly align the target, but I have yet to find a window that allows me to do that, hence the attached, which I have just started to edit (lots of issues).  I'm currently using v. 4.x and wonder if that still matches what I saw/heard in his tutorial, which didn't actually show the manual rotation window but does allude to ti. 

Anyway, onward and upward and thanks again for your help.

 

[symmetal]

I usually wait until the target is 5 or 10 mins passed the Meridian before flipping. SGP should allow a flip if it's only 1 second passed but the issue here is that after the flip the scope may overshoot slightly and end up back on the other side of the Meridian. Then when it does a target centre the plate solve reports it's on the wrong side of the Meridian and it flips again. 😬 Waiting at least five minutes any slew overshoot when flipping shouldn't cause this issue.

It's worth having the Telescope Docking Module visible as shown here. When your slew problem occurred what was the 'time to pier flip' saying. A positive value shows time to the Meridian while a negative value is the time past the Meridian. SGP won't flip if this value is positive. Also check that the reported RA and Dec is close to what your target RA and Dec is. It will be slightly off if you enable dithering but not by much. These reported RA and Dec values are from EQMod and SGP uses these to calculate the time to the Meridian.

The Chris Woodhouse videos are several years old and refer to an earlier version of SGP possibly before rotator options were available. I'm fairly sure earlier versions did report the camera angle after each target centreing attempt, which was useful, but the later versions don't if there is no autorotator present. The only way now to get the camera rotation angle I believe, is to do a manual 'Solve and Sync' from the Control Panel's Plate Solve tab. 🤔

Alan