[Nottingham] LibreOffice cloud?

Duncan notlug at pendinas.org.uk
Tue Jan 21 16:40:52 UTC 2014

> To my eye and cynicism, that looks to be chosen just because it
> photojournalistically "looks pretty" and the "public are all ignorant"
> and the technical merit be damned.

More than likely.  The web page information doesn't help though.

"Hipparcos data give a parallax value for Sadalmelik, also known as
Alpha Aquarii, of 6.23 ± 0.19 mas, corresponding to a distance of
161 ± 5 pc. The Gaia error bar will be 30 times smaller, resulting
in a distance measurement that will be 30 times more precise"

Under normal operations Gaia will not transmit data for stars this bright;
only for stars fainter than (G) magnitude 5.6 (to 20th).

> Hopefully being not so ignorant...
> Why the bifurcation for the vertical blooms?
I don't think they are blooms but are part of the PSF.
I suspect the bifurcation is because the 5 mirrors involved
aren't correctly focused yet so you are seeing a double image.

Without more information about this image it is difficult to
know.  If it were a "raw" SM (star mapper) image then a hi-res
full CCD image would be 2250x990 (2x2 on-CCD hardware binning)
and a low res one 1125x495 (2x2 on-CCD hardware binning followed
by 2x2 on-board software binning).  Since this image matches
neither of those it is obviously an extract and/or been manipulated.

> Is that an artefact of the star image falling an a seam between CCD
> chips or something more interesting?

This should be an extract from a single CCD.  There is one stitch
boundary in (what we call) the along-scan direction and 8 stitch
boundaries in the across-scan direction from the CCD lithography.
One or more of the faint horizontal lines could be stitch boundaries
but their apparent separation in the image isn't right for them all
to be stitch boundaries.

> Also, what determines the angle of the flaring? Is that from optical
> interferometry effects from internal multipath, or an artefact of the
> CCDs? (Is there in reality just a clean bright spot hitting the sensor?)

The flaring is the point spread function (PSF) -- the diffraction of light
through the telescope aperture and off the mirrors, CCD QE etc.
The mirrors aren't "normal" so the PSF has complex wings.

> Could a bright star on the very edge of a CCD splatter or channel light
> through/along the surface of the CCDs to create ghosting?... (OK, lots
> more questions but those can wait for a few beers ;-) )

Each CCD is 4500 pixels in the along-scan (AL) direction.  Our PSF models
suggest Sirius (magnitude -1.46) will have a measurable influence, similar
to that of a 20th magnitude star more than 10,000 pixels away in the AL
direction (we measure stars down to 20th magnitude - more than 100,000,000
times fainter than Sirius).

So yes, when Sirius is transiting the focal plane we will see its
influence in neighboring CCD but that is just normal optics rather
than chance alignments.

Jupiter is even worse because it will be like a bright, saturating pizza box
going across the focal plane.

> The limiting of the blooming looks blooming good for giving well defined
> edges to the blooming! :-)
> Good luck!

Cheers, but yes, let's grab a beer sometime too.

Have fun,

More information about the Nottingham mailing list