I looked at your images this evening. The sky background
is negative in the region around the apertures you sent. The
negative background is what is causing the centroid operation to
fail. It sounds like you have everything working now by aligning
using stars in a region with positive sky background, and then turning
centroid off to run photometry on the larger set of stars,
some of which have negative background counts.
If you are still having trouble, an alternate approach would be
to process each of the three images independently with a custom set
of apertures generated from WCS headers by John's python script, one
for each image. Since the apertures from the python script are
supposed to be aligned to the stars to within a pixel, you can leave
centroid turned off when running photometry on each image. You will
just need to open the appropriate aperture file for each image (or
drag and drop it on the image, whichever is easiest to you), before
you run multi-aperture. As long as you leave the measurementstable
open, new measurements will be appended to the current table.
On 7/19/2013 5:15 PM, David Emory [via
I open a set of apertures and then try to do a stack
align or multi-aperature photometry.
I get a "Centroid ERROR: no signal in aperture for moment widths!'
There is clearly a star in the aperture.
I have tried this numerous times.
If you reply to this email, your
message will be added to the discussion below: