Error during a sequence

[CarCarlo147]

Good morning Robin, help please, I'm trying to set up a sequence that:
take a target,
do the plate solve, center it and sync the mount
continues until the passage to the meridian,
perform a meridianflip
return to the target and plate solve again, center it and sync the mount
and then finish at a certain time.

I'm attaching the SharpCap log where you can also find the sequence I set, but as you can see I get an error when I try to do the plate solve.

Can you help me understand what I'm doing wrong?

Log_2024-07-14T21_10_09-17196.log

(297.91 KiB) Downloaded 39 times

ps.: Why during the sequence together with the fits save a file does it generate the txt file with the settings each time? see attached image. Only one would be enough as normally happens when starting a capture.



Thanks, Carlo

[admin]

Hi Carlo,

well, from the log, it looks like somehow the image that was being captured to plate solve didn't make it to the plate solving code - from the plate solving code's point of view the error was 'Failed to capture a single frame to plate solve'. However, the frame looks like it worked fine. Sounds like a bug of some sort, possibly related to the camera being in still mode rather than live mode at that point (since live mode is the default, it gets much more usage and bugs get flushed out faster). You could try adding a 'switch the camera to live mode' step just before the plate solve to see if that helps as a workaround - I will try to figure out what went on in more detail.

As to the capture settings files - they get created with every capture file saved by design really. Some file formats (PNG, TIFF) get less internal metadata, so the capture settings files are the record of either manually changed camera settings or varying camera state (temperature, auto-exposure values, etc). By default you can't change the settings manually in the sequencer, but you can for non-sequenced capture (and you can in the sequencer too if you want to enable it), so having all those settings files does have a use case.

cheers,

Robin

[CarCarlo147]

ok robin I will try this evening.
I can't find many commands in the sequence editor in the documentation, can you give me information on this command?

"Set exposure/gain for plate solving and focus while running these steps"

As for txt files, can you tell me how and where to disable continuous saving, or have I misunderstood and it can't be deactivated?

Thanks, Carlo

[admin]

Hi Carlo,

you can turn the text files off completely (see https://docs.sharpcap.co.uk/4.1/#Saving ... nformation), but there is no option to limit them to just a single image for the reasons I mentioned earlier.

The 'Set exposure/gain for plate solving and focus' step tries to increase the frame rate but maintaint he image brightness by reducing exposure down to 4s and boosting gain to compensate - it will then run the enclosed steps and put things back how they were (gain/exposure) at the end of the sub-steps. It's actually a building-block that the GOTO target functionality uses. This is documented in the 'Available Steps' section of the documentation (https://docs.sharpcap.co.uk/4.1/#Available%20Steps - see the 'Camera' related steps section)

Still trying to work out why the frame that was seemingly captured without error never made it to the plate solver...

cheers,

Robin

[CarCarlo147]

Hi Robin, I ran some tests last night, this is the result.

By adding as per your suggestion "Put camera in Live Mode" he managed to resolve the plate, then center the object perfectly and continue with the sequence.

Subsequently, he correctly performed the meridian flip, performed new plate solving and attempted to center the object, but failed.

(It seems like the mistake he made some time ago with the decimal separator problem)

I stopped the sequence and did a "Goto Catalog Target/Co-ordinates" which brought the object to the center.

See attached before and after images.

After Meridian Flip.png (534.59 KiB) Viewed 2135 times

After Meridian Flip - GOto Catalog - Result OK.png (542.12 KiB) Viewed 2135 times

I also attach the session log.

Log 15-07.txt

(756.85 KiB) Downloaded 35 times

Greetings, Carlo

[admin]

Hi Carlo,

looking at your longer sequence including the flip, there is nothing that would force the mount to return to the same position after the flip (as far as I can tell). Because you have guiding enabled, the co-ordinates of the mount when the flip occurs will be different to those when the sequence started (every guiding pulse moves the mount a bit). The 'Mount Flip' command will preserve the mount co-ordinates, but only those co-ordinates that it reads off at the start of that step, not the initial co-ordinates at the beginning of the sequence. I *think* that this is the cause of the offset after the meridian flip.

The trick is to add a 'SAVE MOUNT COORDINATES' step at the beginning of the sequence - just after the select/goto target. Then after the flip, add a 'GOTO SAVED COORDINATES' step, which will put the mount back to the exact initial co-ordinates chosen regardless of any guiding drift. The following solve/sync should line everything up correctly.

Glad to hear that the live mode option fixed the solve/sync as a workaround. I couldn't work out how the frame capture managed to fail in your log in a way that was consistent with the log entries that did (and didn't) show up at that point, so I have had to add more logging around that area for the next version to try to catch it happening (I couldn't break it myself either).

cheers,

Robin

[CarCarlo147]

Ok Robin, this evening I will try your advice immediately and will let you know.

I add a request, is it possible via sequence to add the possibility of "GoTO target at RA/Dec" as in the following image?

Goto catalog target-Co-ordinates.png (573.84 KiB) Viewed 2044 times

I tried to enter the command "Move the mount to (RA,Dec) <RA/Dec co-ordinates>" but it opens the window for selecting a target from the catalog.

Greetings, Carlo

[admin]

Hi Carlo,

essentially, what that window does is accept the user input regarding what target to GOTO and the various options, then it builds a sequence containing all the instructions needed and runs the sequence. That means that putting that particular window into the sequencer doesn't really fit nicely, since it would be trying to run a separate sequence inside a sequence and would all get very complex...

If you look in the log after using the GOTO target window, you will actually see the sequence that it builds recorded - something like

Code: Select all

SEQUENCE
    SET PLATESOLVE/FOCUS SETTINGS EXPOSURE 4
        MOUNT TRACKING Sidereal
        MOUNT GOTO RA=09:43:52.0,Dec=+13:11:56
        TARGETNAME Mercury
    END SET
END SEQUENCE   

By using the 'select and goto target' step along with the additional steps you will see the GOTO Catalog tool adding for various options, you should be able to pretty much build the equivalent in the sequencer yourself. The only things I think that you are missing are

1) The 'quick choice' drop down list of Messier/Solar system/Caldwell targets
2) The option to enter co-ordinates manually

cheers,

Robin

[CarCarlo147]

Hi Robin,
last night I carried out the tests using your advice to find the object centered after the meridian flip, everything worked perfectly.
Thank you.

I didn't mean to insert that whole window, but, given the successes achieved so far in creating a complete sequence for deep sky, the final touch is the possibility of being able to insert customized coordinates (outside the catalogues) for the goto.

However, while I was writing, I think I found the solution to this too (big mistake of mine) I didn't notice the text box in the "Sequence Step Properties" section.
I entered the coordinates there and they were accepted. (See attached image)

ps.: Don't notice that I inserted the command "Move the mount to (RA,Dec) <RA/Dec co-ordinates>" before the "Promp to Select and goto a catalog target" it's just a test to see if it accepted coordinate input. Obviously I will only insert one of the two depending on my needs.

pss.: You suggest I modify the sequence manually, by inserting some code, but the relevant area (bottom left) is not editable, see attached image here too.

Immagine 2024-07-19 192728.png (377.08 KiB) Viewed 1963 times

Tonight I will carry out these final tests.
Thanks again
I love SharpCap

Carlo

[admin]

Hi Carlo,

great to hear that you are making progress getting this working

The 'code' in the bottom left is deliberately not available for editing, as I decided there was too much risk of people randomly breaking it by making the wrong edits in that area. If you do want to manually edit, save the sequence as an SCS file, open that in notepad, edit then reload.

For working in the sequence editor itself, it's all set up around drag/drop for moving the blocks then adjusting the properties of each block - a bit like some of the simple kids programming languages like 'Scratch' or 'Lego Mindstorms'. I think this gives less scope for error than text based programming - you can still tell it to do the wrong thing by choosing the wrong blocks or settings, but typing the wrong command or the parameters in the wrong order is not a problem.

cheers,

Robin