Re: weak pixels, etc.

[Tom Droege <tdroege2@earthlink.net>]

Andrew, Arne and all,

Well, what can I say.  I only paid $2000 for this particular chip.  A 
really good one would have cost $40,000.  It has a dark streak at about x = 
1960 as Andrew notes.  It also has a bright streak at x = 1726 or so.  The 
bright streak is obvious when you look at an image.  The dark streak is 
probably a worse defect and it is much harder to see.

Once you cut out a 10 tall by full width strip and blow it up, it is easy 
to find the dark (and bright) streaks.

I think that you just cut these areas out of the chip.  I would cut a half 
dozen pixels out on each side of the defect and just put up with the 
additional boundaries.  Sigh!  I must say that these chips are much better 
in quality than I had been  led to expect.

Seems that I recall that there are defects that eat electrons (traps?) I 
suppose that this is one of those.  Really bright stuff gets by the defect, 
but is smeared out as Andrew notes.  I suppose it can only eat so many 
electrons and then it is satisfied until it rests up a bit.  Burp!

Folks, I think that these are expected.  What say Arne, you have experience 
with these devices in the old days when they were not so good?  I believe 
that I sent you Data Set 20.  You will find the particular file on 
DS20D5.  But you can also see it on the flat in DS20D2.

Tom Droege


At 12:57 AM 2/10/02 +0000, you wrote:
>On Fri, 8 Feb 2002 09:48:20 -0700, <aah@nofs.navy.mil> wrote:
> > ...
> >  I still don't understand the problem with weak pixels.
> >Andrew says:
> >>Not in this case! The defect is hard to see on both the Darks
> >>and the Flats but is much more obvious (i.e. I found it when I
> >>knew exactly where to look) on an actual image. It looks as though
> >>electrons smear out over the full affected length of the column
> >>(or is that row? It's vertical on my plot: AXIS2 direction)
> >This doesn't make sense to me, unless what he is experiencing is
> >blooming.  I'd like to see four images posted:  a raw, a flat,
> >a dark, and the processed image for one of these cases.  Again,
> >processing is supposed to make all pixels equal.  If this is not
> >the case, then something is wrong.
> >Arne
>
>Yes - it looks like some sort of blooming (says he, not
>knowing the precise definition of the term.)
>
>An obvious example is to be found near X1=1960, X2=712
>raw coordinates, FITS style, counting from 1, in file
>hira2199915.fits. The smeared columns are X1=1960,61,62
>starting abruptly at X2=459 and extending to the data
>cutoff at X2=2032.
>
>Perhaps somebody with the appropriate software could
>crop out this bit from this image and from a corresponding
>raw flat (e.g. hira2244571.fits) and dark (e.g. hira2244620.fits)
>for us? I don't have suitable software at the moment.
>
>Andrew Bennett, Avondale Vineyard