Re: signal detector question
- To: mathgroup at smc.vnet.net
- Subject: [mg39651] Re: signal detector question
- From: "John Doty" <jpd at w-d.org>
- Date: Thu, 27 Feb 2003 00:32:26 -0500 (EST)
- Organization: Wampler-Doty Family
- References: <b3hs60$ikb$1@smc.vnet.net>
- Sender: owner-wri-mathgroup at wolfram.com
In article <b3hs60$ikb$1 at smc.vnet.net>, "The Leddons" <jleddon1 at attbi.com>
wrote:
> Does anyone know of an example of pulse pile up and how it is detected
> in a data set? This is when two photons or particles hit a detector at
> the same time and get counted as one particle/or photon. Is there a way
> to uncover this problem in a residuals analysis of the data distribution
> for example? Would appreciate an example notebook.
It's a complicated question.
Just seeing it in a pulseheight spectrum is impossible without some idea
of what the spectrum would look like without pileup. It's generally more
useful to infer an incoming rate from the data and *calculate* the
expected pileup percentage. With a parameterized model of the incoming
spectrum and an instrument model that includes pileup you may, in
principle, adjust the parameters to fit your pulseheight spectrum, thus
taking pileup into account. It is difficult to get this right in practice.
Sometimes you may have a more direct indication. X-ray CCD detectors often
register some information on the shape of the charge cloud left by the
X-ray (the jargon is "event grade"). An unusual number of the more exotic
grades indicates the presence of overlapping charge clouds ("pileup").
--
| John Doty "You can't confuse me, that's my job."
| Home: jpd at w-d.org
| Work: jpd at space.mit.edu