[Halld-cal] We are proceeding with the SiPM contract award

[Elton Smith]

Dear collaborators,

At our calorimeter working group meeting yesterday we made the final
decision to use the SiPMs for the readout of the Hall D barrel
calorimeter. A brief summary of the evaluation is given in Eugene's note
below. We have instructed the procurement department to prepare the
necessary paperwork to make the award.

Cheers, Elton.

---------- Forwarded message ----------
Date: Thu, 14 Oct 2010 20:52:33 -0400 (EDT)
From: Eugene Chudakov <gen at jlab.org>
To: Glenn Young <gyoung at jlab.org>
Cc: Elton Smith <elton at jlab.org>
Subject: Hall D SiPM procurements

Dear Glenn,

after deliberation we confirm our commitment to SiPMs for the BCAL
readout.

Extensive radiation tests performed allow us to evaluate the lifetime
of the sensors to be at least 9 years, under certain conditions:

1) Detectors are kept at about 0C while running, but heated to
     room temperature or higher for some period every year to anneal them;

2) The experiment is running at high luminosity: 100 MHz of 8.4-9GeV
     photons on 30cm long LH2 target, effectively for 1/3 of the calendar
     period;

3) The neutron flux is 30 Hz/cm^2 (GEANT prediction). [Note that FLUKA
    simulations predict lower fluxes by approximately 30%].

In 9 calendar years of running the noise from the downstream SiPMs
is expected to increase by a factor of less than 5, and be still tolerable
for GlueX. The damage of the upstream sensors is expected to be about 4 times
lower.

It was agreed that the radiation hardness of SiPMs was acceptable for GlueX.
Two other issues were brought up:
a) after-pulsing and its temperature dependence;
b) the summing scheme.

The issue a) has been investigated. An upper limit for "slow"
afterpulsing (decay time of about 200ns) was found to be about
10%. "Fast" afterpulsing which can be confused with cross-talk is also
about 10%.  No temperature dependence of this effects have been
observed. Such level of afterpulsing is acceptable for the experiment.

We plan to address b) in parallel to moving forward with the SiPM
procurement, as it is an optimization of their use in our detector.

Regards,
Eugene