Timing of PMTs
The relative timing of each hit-PMT is important for determination of the vertex. This relative timing depends on observed charge because of the discriminator slewing effect. The timing difference caused by pulse height has to be measured precisely in each PMT. The timing resolution also depends on pulse height and can be measured at the same time.
Fig 1 shows the system for measuring the relative timing of hit-PMTs. The N laser
light source can emit intense light of wavelength 337nm within a very short time (less
than 3 nano seconds). The wavelength is converted to 384 nm which is near the Cherenkov
light wavelength by a dye laser module. The light intensity is changed using an optical
filter, and the measurement of PMT timing at various pulse heights is made. After going
through the optical filter, the light is split into two, one goes to the diffuser ball in
the water tank through an optical fiber, and the other signal is used for monitoring and
making trigger signals. The schematic view of the diffuser ball is also shown in Fig 1.
The diffuser tip which is located at the center of the ball is made from 
suspended in optical cement. Light emitted from the tip is further diffused by LUDOX,
which is silica gel made with 20nm glass fragments. The combination of diffuser tip and
LUDOX can make modestly diffused light without introducing timing spread.
|
Figure 1: The system of the measurement of the relative timing of hit-PMT. |
A typical 2-dimensional plot of timing and pulse height is shown in Fig 2, which we call a ``TQ-map''. Each PMT has its own TQ-map, because the character of each PMT is unique. The vertical axis of this figure shows the hit timing; larger values indicate earlier hit timing. As shown in the figure, higher charge hits have earlier timing and better resolution. The typical timing resolution at the single p.e. level, which is important for the solar neutrino analysis, is 3nsec.
|
Figure 2: A typical 2-dimensional plot of timing and pulse height distribution. (TQ-map) |
