request for help with plate solving in 1.4

Hello team,

I am trying to stack a set of MilkyWay pics. i must be doing something fundamentally wrong when plate solving. i keep getting “not enough stars” error when i try with siril solver. I followed the other messages about the servers being down and so setup a local star database and ran it against that. that did not work as well. i then uploaded one of the pics to astrometry.net to see if it identifies anything and it did and the results are at Astrometry.net submission 12446407. i then downloaded the generated wcs.fits image and test ran to solve it locally, and even that failed with local astrometry index files setup. any pointers as to what i am missing?

Hello and welcome,
if the error is ‘not enough stars’, the problem must lie with the star detection (see this FAQ entry, the next one can be useful too: Siril - FAQ).

Use the Dynamic PSF tool to adjust the detection parameters and see if you can have enough stars detected everywhere in the image.

1 Like

Thank you for the immediate response.
I did try the Dynamic PSF tool. I see logs as below in the console

( I see that some of the detected stars are shown with a purple circle and most in a red/orange colour. any reason why is that? )

19:39:35: Findstar: processing for channel 1...
19:39:36: Execution time: 471.29 ms
19:39:36: Found 3066 Gaussian profile stars in image, channel #1 (FWHM 4.022458)
19:39:46: Findstar: processing for channel 1...
19:39:47: Execution time: 752.69 ms
19:39:47: Found 2739 Moffat profile stars in image, channel #1 (FWHM 3.684900)

after that I tried plate solving again.

19:41:43: Plate solving image with astrometry.net for a field of view of 154.29 degrees
19:41:43: Using 2739 detected stars from image.
19:41:43: Astrometry.net solving blindly
...

after trying through all the index files, it fails with the message

19:42:47: solve-field: Field 1 did not solve (index index-4212.fits, field objects 11-20).
19:42:49: solve-field: Total CPU time limit reached!
19:42:49: solve-field: Field 1 did not solve (index index-4211.fits, field objects 11-20).
19:42:50: solve-field: Field: /Volumes/nvme/Milkyway-siril/result_375s.xyls
19:42:50: solve-field: Did not solve (or no WCS file was written).
19:42:50: No astrometric solution found
19:42:50: Plate Solving failed. The image could not be aligned with the reference stars.

are my source images itself bad? or am I missing anything.

ok. after multiple retries with different options selected, it finally gave the result. I am redoing the process with a different set of pics. I will be back with the results.

If you could share the image that was difficult to solve, that would be great!

that’s always complicated

the purple ones are saturated, the red ones are not

1 Like

Sometimes drawing a selection around the center of the image first, will allow it to platesolve. Like @vinvin mentioned, the large FOV is the issue at times.

1 Like

that is helpful info. I now am able to understand the numbers a little better.

I was following your beginners video from your channel. I went from absolute 0 to at understanding the other discussions from the forum, setting up my local star database, and using it to render my stacks. thank you for the very helpful info.

1 Like

it took me a little time to understand how to use the Dynamic PSF tool. I applied that and then tried plate solving and it worked.
thank you!