Two questions about "Mount SolveAndSyns" in sequence editor

[photon08]

In the sequence editor, one of mount commands is 'Mount SolveAndSyns', which does the following:

1. take a snapshot
2. plate solves it
3. sync the mount with the plate solution
4. re-goto the target.

My first question is how to set the target coordinate the mount goes to in #4 above. The sequence editor does not seem to have a way to set the target coordinate (RA,Dec). Deep Sky Planner offers a way to set the target RA,Dec, but it is not active unless SharpCap is connected to a mount, which is inconvenient because I want to do the sequence editing sitting at my desk away from my mount).

Instead of "Mount SolveAndSync', I can use 'Mount SolveAndSyncOnly' and 'Mount Goto {Ra, Dec}', but it's a bit inconvenient to copy {Ra, Dec} to multiple places.

My second question is if there is a way to do the following:

1. Take a snapshot
2. Plate solve
3. Sync mount
4. Go to {Ra, Dec] / target
5. wait for the mount to settle
6. Take a snapshot
7. Plate solve
8. Sync mount
9. measure the offset between the solution in #6 and the target in #4.
10. If the offset is less than the pre-set tolerance, end.
Otherwise, repeat steps 4 - 10 until it reaches max attempt count (say 3 attempts)

NINA has this feature and I like it, but it does not support live-stacking. I'd like to have this feature in SharpCap for live-stacking with my dobsonian because its tracking and goto quality is poor and I need to recenter every 2 minutes or so (inside LOOP in sequencer).

If the sequencer allows me to get the offset in step 9 above and have a conditional branch (if statement) in step 10, I'd be able to write a function/subroutine (assuming that there is a way to add a function/subroutine).

Thank you for any help in advance.

[admin]

Hi,

so, to answer the easy question first, the co-ordinates used as 'target' co-ordinates for the recenter are the co-ordinates that the mount is pointing to prior to the 'plate solve/resync' step being run. So, the intended use pattern is

* Move the mount to the RA/Dec co-ordinates of the target ('Move the mount to co-ordinates' step)
* Plate solve, sync and re-goto the target

The first step may well not put the right co-ordinates in the center of the frame, but the second should fix that problem.

Once you have plate solve+sync'd in a particular area of the sky, you can do GOTO movements nearby with good accuracy without needing to plate solve again.

There are a number of other steps that you can use in your sequence to set the target - most notably the 'repeat for each RA/Dec co-ordinate in <Co-ordinate list file>' which will read co-ordinates from a text file (one co-ordinate pair per line), move to each co-ordinate in turn and run the sub-steps for that co-ordinate before moving on to the next. Note that you can also put a target name at the beginning of each line before the co-ordinates - see the documentation for more info : https://docs.sharpcap.co.uk/4.0/#Repeat

As to the loop of 'plate solve, recenter, check distance, redo if too large', there is no current way to do that in SharpCap or using the sequencer. I have resisted the temptation to introduce conditional branches or 'if' statements into the sequencer (and variables too, for that matter) as I feel it would increase the complexity too much and also make it harder (and maybe impossible) to estimate how long a sequence will take to run in lots of situations.

I would suggest that there is (probably) limited value to the plate solve loop itself - the plate solving itself is basically accurate, so any pointing errors after the first plate solve and recenter are going to be due to inaccuracies in mount GOTO movements. If a second plate solve/recenter does not fix the issue (and presumably the second one would only have a very small adjustment to make), then the mount movement must be inaccurate for even very small movements, so it's unlikely that a third, fourth or even more would make a significant improvement on the second except by simple chance of the randomness of position happening to turn out to be small on one occasion.

Personally I find that a single plate solve+recenter is usually enough to put the target at the center of view (to within a handful of pixels).

cheers,

Robin

[aklatt]

Hello,
Thought I’d ask here rather than start a new thread. I am able to plate solve in Nina, but it always fails in SharpCap and I can’t figure out what I’m doing wrong. All my settings seem correct, mount is connected, scope is 50mm at fl of 200 with asi 120mc-s, so field of view is sufficient.
I am using astap and have updated the files to the most recent set (d50 I believe it was).
Thank you for your thoughts,
Arthur.

[admin]

Hi Arthur,

lots of possibilities from incorrect settings to the image being too faint. Some pointers

* Disable the use of focal length in the plate solving settings if you are having issues - if the setting is enabled and wrong it can fail. If it is disabled it will work but is slower

* Make sure you can see stars *without having the SharpCap display stretch enabled)* . 4s exposure and mid to high gain is a good starting point

* Look in the log to see how many stars Astap is detecting - ideally somewhere in the 50 to 300 range. Any lower, increase exposure/gain. High hundreds, reduce gain and possibly exposure.

* If all else fails, press the 'Snapshot' button to save a frame that fails to solve and post it here in PNG or JPG format and I will investigate and advise further.

cheers,

Robin

[photon08]

I would suggest that there is (probably) limited value to the plate solve loop itself - the plate solving itself is basically accurate, so any pointing errors after the first plate solve and recenter are going to be due to inaccuracies in mount GOTO movements. If a second plate solve/recenter does not fix the issue (and presumably the second one would only have a very small adjustment to make), then the mount movement must be inaccurate for even very small movements, so it's unlikely that a third, fourth or even more would make a significant improvement on the second except by simple chance of the randomness of position happening to turn out to be small on one occasion.

Yes, both goto and tracking quality of my mount is certainly subpar. It's Orion XT10G (Dobsonian - Alt-az). Needless to say, it also depends on how good my 2-star/brightest star alignment was, but I tried hard to get it as good as possible, but it's still not great.

That's the very reason I asked for this feature in the first place. And, my experience with NINA that does support what I asked here is that usually the 2nd attempt puts the mount close enough (NINA allows setting the distance error threshold) , which is why I asked for this feature in Sharpcap.

Personally I find that a single plate solve+recenter is usually enough to put the target at the center of view (to within a handful of pixels).

I guess your finding is based on your experience with an equatorial mount. As I wrote above, poor Dobsonian users like me can benefit from the feature request I made here.

Actually, I found another user requesting exactly the same feature. See viewtopic.php?t=6025 .

[admin]

Hi,

thanks for the explanation. Would you be able to do an experiment for me to give me some numbers on how many tries are needed and how close the mount gets in the end?

If you set up SharpCap to use the internal SharpSolve plate solving tool (which should be nice and fast) and then move to a target and repeatedly press the Solve+Resync button in the mount controls (or from the menu). Each time, you should see a notification bar telling you what the offset was - ie

Mount synced to <Some co-ordinates>, recentering on target at <target co-ordinates> (offset of X.XX degrees)

I'm looking for info on how those offset numbers drop as you try multiple resyncs - that just gives me some guidance on what threshold might be appropriate or what number of loops might be appropriate when I get a chance to add this.

cheers,

Robin

[photon08]

Thank you for the reply. December/January in my place have been mostly cloudy/rainy (especially Friday-Saturday). I'll try to do what you suggested once I have a clear night.

In the meantime, a new 'auto-center' feature might make what I asked for in this thread unnecessary. I'll also try that and see how it goes. Some Dobs goto users reported that auto center works well for them as you know (e.g. in viewtopic.php?t=6531&sid=0254480298b5c2 ... cb477cec96)