6.5. Photodetectors 389
Dynodes345678910μ103104105106Figure 6.5.17: Dependence of gain on
the number of dynodes in the multipli-
cation structure of a PMT operated at
2000 volts.specially at the ends of the photocathode. Furthermore the collection efficiency of
the photoelectrons by the first dynode is usually not the same for the whole of the
photocathode. The termuniformityis used to characterize the variations in the
anode current when photons strike the photocathode at different positions. This
uniformity is generally quantified by measuring the anode sensitivity to directional
light incident on different points on the PMT entrance window. Fig.6.5.18 shows
typical PMT sensitivity curves for head-on and side-on type PMTs. The response
of head-on type PMT is spatially more uniform than the side-on type PMT. The
sensitivity of side-on type PMT to light incident at and around the center of the
photocathode is maximum and decreases to about 50% at the upper and lower
corners. In contrast the head-on type PMT shows an abrupt decrease to almost
zero near the corners of the upper face of the phototube. The anode sensitivity for
the middle region, however, is fairly uniform. It is therefore recommended that the
system is designed such that most of the incident photons strike the middle part of
the PMT entrance window.
A.12TimeResponse.........................
In certain applications the time response of a PMT becomes extremely important.
for example in Cherenkov detectors the timing information is used to construct the
event vertex, that is the point of origination of the Cherenkov light. In such a situa-
tion a time uncertainty of less than a few nanoseconds may be desirable. Similarly in
case of pulse shape measurements the reproduction of the incident waveform requires
good PMT time response.
The response time of any PMT depends almost exclusively on the electron mul-
tiplication process. The reason is that the first stage of the photoelectron collection