Recently, a script to plot the HR diagram from an image was released, but my results look strange.
I followed the steps from the video:
https://www.youtube.com/watch?v=rfmWigz66v0
Linear image with SPCC
Download the 16 parts of the Gaia DR3 catalog
Download the repository
Any idea what might be going on?
📷 image.png
my output
📷 image.png
linear image
📷 image.png
STF image
📷 image.png
Console output (only 270 stars detected)
Ari, here’s what I’d suggest.
start a fresh PI session;
drag and drop your R, G & B autocrop.xisf masters into the workspace from where they live on your local data drive;
close the crop_mask frames for each master;
use ChannelCombination to generate a fresh linear RGB;
use DynamicCrop to crop out all but the immediate vicinity of your target globular;
run the imagesolver script to rebuild the astrometry blown away by the DynamicCrop;
run BXT coorect only;
run SPCC. Be careful with your white reference in SPCC as it can influence the resulting HR diagram.
for my test I did yesterday on 47 Tuc data, I used G0III as my reference. whether that was a good choice is debatable but since globs are predominantly late giants, I figured a yellowish type 3 star white reference would be good.
Finally, run the HR diagram script.
in the settings for the script, be sure to turn on “Adjust for distance “ and “Proper motion adjustment” and also you can play with the Circular filter setting to exclude surrounding non-cluster stars (it’s like a circular mask).
have fun!!!
Gordon Pegue · Oct 15, 2025, 03:44 AM
Ari, here’s what I’d suggest.
start a fresh PI session;
drag and drop your R, G & B autocrop.xisf masters into the workspace from where they live on your local data drive;
close the crop_mask frames for each master;
use ChannelCombination to generate a fresh linear RGB;
use DynamicCrop to crop out all but the immediate vicinity of your target globular;
run the imagesolver script to rebuild the astrometry blown away by the DynamicCrop;
run BXT coorect only;
run SPCC. Be careful with your white reference in SPCC as it can influence the resulting HR diagram.
for my test I did yesterday on 47 Tuc data, I used G0III as my reference. whether that was a good choice is debatable but since globs are predominantly late giants, I figured a yellowish type 3 star white reference would be good.
Finally, run the HR diagram script.
in the settings for the script, be sure to turn on “Adjust for distance “ and “Proper motion adjustment” and also you can play with the Circular filter setting to exclude surrounding non-cluster stars (it’s like a circular mask).
have fun!!!
Thank you for the tips!
I’ve been a bit busy and was only able to get back to working on this today.
Unfortunately, I haven’t made any significant progress.
I recorded my screen showing the workflow so it’s easier for you to identify where the mistake might be, if you could help me with that.
Here’s the YouTube link:
https://youtu.be/7jTqSK_6dl8
Thank you very much!
Try cropping to include only the GC without background/foreground stars, or by using the circular filter tool int he script