Re: Minutes posted, calib goals requested (fwd from George)

[Elton Smith <elton@jlab.org>]

---------- Forwarded message ----------
Date: Mon, 21 Apr 2008 12:12:21 -0600
From: George Lolos <gjlolos@uregina.ca>
To: Elton Smith <elton@jlab.org>
Cc: halld-cal@jlab.org
Subject: Re: Minutes posted, calib goals requested

Hi Elton:

For the BCAL, the needs of a calibration/monitoring system, as the one
pursued by NKUA, serves two purposes:

First, it is very valuable for the initial setting (and subsequent
start-ups after shutdowns) of the relative timing of all inner and
outer layer sensors and their electronics/cabling/signal processing
modules.  In other words, we need to know that all signals are timed
for the same light pulse arrival time so the only timing differences
will result from hit location in the BCAL.  I know that there are many
ways to do this, however, nothing beats the ease and speed of a common
flashing system with properly designed cabling to accomplish this goal.
   Clearly, it also makes diagnostics of bad or malfunctioning sensors
much easier and faster than using cosmics.

Second, if we know that the light output of the flashing system is
uniform enough as distributed to the sensors, a first order
equalization of gains can be made and then we use other more precise
methods to make sure gains are equalized to high precision.   However,
such a system makes on-line diagnostics of gain variations due to
degradation/damage/ temperature effects easy and fast.  In fact,
software "alarms" can be set for each sensor to alert us when the
flasher ACD and/or TDC means fall out of a predetermined window.

How useful the use of the system to equalize gains will be is, in the
end, determined by the reproducibility and stability of the flashing
system and its light pulse distribution system.   It's perhaps too
demanding a requirement to have a system that is uniform and stable to,
say, 10%, but in any case, being able to equalize gains of sensors to
within 20% initially, saves time when we use software with physics data
to further improve calibration constants.

As for the number of photoelectrons, the light input into the sensors
should have an adequate number to give rise to a decent timing
resolution and "representative" pulse heights.  For the BCAL, in
general, we have two very different expectations for the inner and
outer layers, respectively.   Having 1000 photons incident on an inner
layer SiPM (for example layers 2-4 from the inner face) is no big deal,
while the outer sensors may see only 100 photons incident.   I would
say that if we aim for something like 25-50 photoelectrons generated by
the SiPM's (probably about the same for the outer layers) we should be
O.K.  This would correspond approximately to 200-400 photons incident
on the front face of the light guide-WC collectors.   These are rough
numbers.

For the SiPM's, the flat response spectrum is between 490-550 nm, give
or take.    Since we are talking either about LED's or plastic
scintillator as the medium of the calibration system, I would think
that picking a source around 500-530 nm is just about perfect.

Cheers,

George

On 21-Apr-08, at 10:14 AM, Elton Smith wrote:

>
> HI all,
>
> Zisis posted the minutes of today's meeting on the wiki. Thanks! I did
> add
> some comments at the end on questions of calibration.
>
> We need to clearly identify the goals of the calibration system, so
> that
> we build what we need. At the moment, the calibration system is geared
> toward gain monitoring of pmts. For the next meeting, we would like the
> Fcal and Bcal and other pmt-based systems to specify what needs to be
> monitored at to what accuracy. Pavlos suggested that it would also be
> useful to give an estimate of the signal size (e.g. number of
> photoelectrons) as well as wavelength response of the detector.
>
> Cheers, Elton.
>
>
>
> Elton Smith
> Jefferson Lab MS 12H5
> 12000 Jefferson Ave
> Suite # 16
> Newport News, VA 23606
> elton@jlab.org
> (757) 269-7625
> (757) 269-6331 fax