Test in Naples on
Scintillator Strips:
Time Resolution
In order to test the scintillator strips used in
the 3 planes of the ToF system of Pamela, a telescope
was
built with 4 layers of streamer tubes ( Y coordinate ) each one equipped
with pick-up strips ( X coordinate ).
The streamer tubes used are these of the MACRO exp.
at Gran Sasso, so their section (3x3 cm2) is not optimized for
our goal. In the X direction, pick-up strips of 0.5 cm pitch or less, obtained
by OR-ing several wires of 64 conductors flat cable [1],
provide a better space resolution, then the scintillators to be tested
are placed along this axis, on the third plane of the telescope.
The trigger is provided by a couple of 30x
30cm2 paddles, on the top and at the bottom of the telescope.
Cosmic ray tracks are reconstructed in the Y-Z and
X-Z views and the position in the test scintillator is extrapolated
with a space resolution of about 1 cm (Fig.1
)
The X position of the track in the scintillator
(in the plots called 'xscin') is compared to the half
difference in time recorded by the 2 PMT's = (T1 - T2)/2 as shown
in ( Fig.2 ) . Taking
into account light guide lengths, the light
propagation speed in the scintillator is estimated
to be about 19 cm/ns,
in agreement with the value of n=1.58 provided by Bicron.
The time resolution
obtained (without time walk corrections) is about 130
ps for S11,S12 as is possible to see in Fig.3
and is in the range
140-150 ps for S31 and
S32. This is probably due to the presence
of a couple of cookies (6 mm of silicon) behind the PMT's of S3 scintillators
instead of only one (3 mm) in the case of S1 scintillators.
Fig.4
shows the different response in charge of the same PMT with 3 mm silicon
cookie or 6 mm. The mean value of the charge distribution with 6 mm silicon
is typically 15-20 % less than in the other case. The silicon cookies
are the solution adopted after the vibration tests, in order to avoid
a too hard contact between light guides and PMT's.
The S3 and S1 scintillator layers are submitted
to vibration tests in the Galileo labs in Florence. Fig. 7 e Fig. 8 show
for S31 that time resolution before and after
vibrations is not modified. Results of vibration
tests also suggested to give a different shape to light guide edges. Time
resolution is not affected by the introduction of this different
coupling (Fig.9).
Each scintillator is characterized in charge and
in time with a couple of PMT Fig.5.
Charge distributions obtained on PMT 984 and 955 are well fitted with a
Landau curve ( Fig.6 )
Notes:
-
[1] To have small pitch pick-up strips we use
a 3M 2010 - 28 AWG stranded copper conductors in halogen free insulation
flat cable. Several flat cables are glued on a side of a polycarbonate
layer, with a copper film on the other side. The technique is the
same used in the Monolith prototype. See for reference
NIM A 456, NIM A 457, NIM A 461.