[KIDS] ccd Crosstalk

[aniello grado]

Hi Ewout,


On Tue, May 8, 2012 at 12:11 PM, Ewout M. Helmich <helmich at astro.rug.nl>wrote:

> Hi Lino,
>
> There is a difference between measuring the crosstalk and correcting for
> it.


Indeed I was speaking about correcting. But your answer triggered another
question: what about case of very crowded fields? ok it is a special case
but still possible to happen.



> Two things will likely happen:
>
> 1) Your example would produce outliers in the fit to measure crosstalk for
> all "overlapping" pixels (this in both plots #95 -> #96 and #96 -> #95).
> Presumably the overall fits will not be affected much because clipping is
> done before the fit. There will be other (and more) pixels with e.g. value
> 40000ADU that fall on the background in the other CCD rather than another
> star.
>
> 2) Saturated pixels in #96 will be affected by crosstalk from #95, and
> therefore you get a comparable issue as happens on the low end when
> negative crosstalk is larger than the bias level: you get a dark spot on
> top of the saturated star in #96.
>
> I was not considering the case of saturated stars. You have two sources in
the same place in the respective ccd. The counts of one source will affect
the other and the other way around. May be I am wrong but in this case it
seems to me that you cannot recover the correct values.

ciao

Lino



> Regards,
> Ewout
>
>
> On 05/08/2012 11:51 AM, agrado wrote:
>
>> Dear Ewout,
>>
>> There is something about crosstalk that it is not clear to me. Let suppose
>> that on ccd95 you have a source at a certain position and on ccd96 you
>> have
>> another source at the same position. How will be corrected the crosstalk
>> in
>> this case?
>>
>> Thanks
>>
>> Lino
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listman.astro-wise.org/pipermail/kids/attachments/20120508/f6f89123/attachment.html