As most of you already know, MACRO's PMTs are slowly but steadily
being contaminated by Helium. This effect is seen as afterpulsing at
the characteristic (for He) time of 1.2 usec following an event with
sufficient energy. The He contamination is restricted only to EMI tubes,
while HAMAMATSU PMTs appear to not be affected (different glass).
Among the EMI tubes, bottom and center have 1/2 of the helium the top
counters have. Although ventilation should be hold responsible for this
effect, due to the nature of the diffusion we haven't really had so far
a direct demonstration of that.
The most immediate effect of He afterpulsing is an increase of the
TOHM trigger rate, seen in particular for counters where both sides
(-0 and -1) are heavily contaminated (>10% as measured by the ratio
of the charge of the afterpulse over the charge of the primary pulse)
by He, and thus trigger even on coincidence of radioactivities.
As a result, for the worst cases, we have counters that present
a TOHM trigger rate of more than 20mHz and thus
(1) increase dead time
(2) reduce run duration
(3) introduce background for slow monopole searches
In facing the next 9 months or so of MACRO's running, one may ask
of a possible plan to deal with this problem. There are a number of
alternatives for this:
1) As long as acquisition can handle dead time (i.e. <2.5%) and rate,
do nothing about it. Afterpulsing can be viewed as a feature of the
problem counters, a continuous monitor of the slow monopole trigger.
If it becomes necessary, counters with extreme rate will be disconnected.
10-20 counters are expected to reach an unacceptable rate by May 2000.
2) If Bottom and Center counters have less helium than the Top ones we
can perform massive swap of EMI tubes between the T and B/C layers.
(e.g. move all -0 side top EMI's to the bottom-0 and v.v.) Performing
such operation detector wide, implies 2-3 weeks of downtime of the entire
detector but it is probably the only solution that can guarantee the same
operating conditions for MACRO for another 2-3 years *without* buying any
tubes.
It can be an attractive solution if other detector wide retrofits are
scheduled at the same time (maybe when Stu will be on shift?)
3) Operate the Slow Monopole Trigger (i.e. the trigger that mostly
triggers on the afterpulses) in a two face coincidence. This will
reduce the SMT rate by a factor of 7-8. This option is practically in
place for what concerns data distribution to US institutions: the rare
dsts are imposing in software a two face requirement. Continuity in MACRO
running for its nominal duration does not favor this option, i.e.,
we'd rather have the whole MACRO data set with the SMT triggering on
single counters.
This may ultimately be the option for running the slow monopole trigger
system beyond the nominal end of MACRO data taking (May 2000).
4) Given that the Helium contamination is more strongly present in the attico
counters, we can use the ability of the TOHM/LI to set thresholds
(LI counts) on a counter-per-counter basis and suppress the rate of
the hot counters by increasing the LI threshold. Increasing the LI threshold
will affect the system's sensitivity on both the high and low velocity
region. This can be simulated in the detector Monte Carlo, however,
it will add a difficulty layer that will most likely imply abondoning
of the simulation of these variations. Moreover, any LI threshold
adjustment in the problem counter require involvment of experts.
5) Replacing the EMI tubes with the highest He contamination is the most
natural solution to this problem; the question is where to find spare
EMI tubes. As of today, we have roughly ~10 good spares and these will
most likely be needed for inevitable replacements during the next 9 months
or so of MACRO's running.
The detector's end caps (South/North) are instrumented with EMI tubes that
are in very good state and more importantly are low in He concentration
(~1/2 respect to the B/C faces and 1/4 respect to the T face).
We can use these tubes to repair the top counters and at the same time
use our spare HAMAMATSU tubes to instrument the South/North counters.
As I'm told, streamer tubes used for the South face are the rejects
from the construction of the rest of the detector and the use of it
in various analyses is questionable (if people could commnent on that?).
Pursuing this option will require 1-2 weeks during which few counters
will be disconnected.
In conclusion, the He contamination of our counters already has and will have
an even more significant impact on data rates during this last period of
MACRO's running. If we don't want to start disconnecting channels from the
Slow Monopole Trigger system we need to take action very soon by replacing the
highest in He concentration tubes. Having practically no spares available and
very limited possibility to buy new tubes (and have them delivered at GS within
a month or so), option #5 I've outlined above is the only realistic scenario
that will keep MACRO as uniform as possible for the rest of its running, will
attact the rate problem drastically and will have minimal effect on analyses;
we should thus proceed in its implementation the earliest possible (Sep 99).
Please let me know of any comments or thoughts of yours on this issue.
--Erik