This is certainly not a high priority issue, but knowing this in advance may save a bit of time and effort. VPhot stacking doesn't work on images which have been "solved" using the WCS tool. I have run into this issue previously, but I just moved on and didn't try to understand exactly what was happening.
In a set of images one of the three B images didn't solve. I applied the WCS tool to that image, and the box turned green. When I examined the stacked image, all the stars were doubles. When I stacked the two images which did not need WCS correction the stacked image was fine. When I stacked the WCS corrected image with one of the other images (which did not require WCS correction) the stars were doubled.
Conclusion: Don't bother trying to stack an image which has been WCS corrected.
Perhaps there is a work-around. If so, I'd like to know it.
Phil
Hi Phil:
The Update WCS tool often works but it uses a very simplistic approach. It just copies the the WCS headers from a good image and applies it to the 'bad' image.
If there was a scope shift between the good and bad images (btw, I bet your first B image was the bad one from the aavsonet scope?), even after the update wcs, the shift is not aligned properly and double stars appear. I have seen this before on occasion. I did not find a resolution.
IF you really need to stack all three images for SNR reasons, ask for your aavsonet plan to include 4 B images.
I think you are correct for these specific images but I do not agree that your conclusion is necessarily true for every image that has been WCS updated. We haven't looked carefully for a resolution since it doesn't occur too often. BTW, did you notice whether the single 'bad' image already had double stars? We have seen that happen sometimes during imaging of the first exposure where the pointing disagreement was just large enough (1 arcminute) to cause ACP to repoint the second image. Haven't determined why this happens but tried to fix it in other settings. Looks like it still doesn't always work. ;-(
Ken
Hi Ken,
"did you notice whether the single 'bad' image already had double stars?"
The "bad" image looked fine when viewed by itself. The "doubled stars" were shifted 38 arcsec from the position of the other stars' position in the stacked image (according to the Measure PA and SEP tool).
"I think you are correct for these specific images but I do not agree that your conclusion is necessarily true for every image that has been WCS updated."
Yes, you're right. I now recall one occurrence of this when stacking did work correctly with a WCS corrected image.
I should change my conclusion to: "Just don't be surprised if it doesn't work" :- )
Phil