As part of the ongoing analisys of data collected during the RHIC 2000 run, a measurement of the linearity of the photomultipliers used fort the scintillator tiles, was performed in the lab.
The measurement was done with a bright blue LED inside an integrating cube. The LED is biased with a driver that produces sharp pulses in time. The source of light is mounted on a track and can be moved away from the PMT with a remotely driven stepping motor. The signal from the PMT is read with a LeCroy 2249 ADC (10 bit). Because of the small range of the ADC used for the test, attenuation of the signal was necessary to cover the range of the LeCroy 1885 ADC used in the experiment. The measurement consists of a comparison of the source position dependance for different degrees of attenuation.
The plot shown above summarizes the measurement. The mean value of the ADC distributions is plotted in the vertical axis. The numbers of steps of the stepping motor is displayed in the horizontal axis (1000 steps is ~1cm). For all attenuations settings, the points at 0 steps had a pulse height value of 660 mV as can be seen in the next figure.
For the unattenuated measurements (blue dots), the point at step zero corresponds to the end of the high resolution (50fC/channel) region of the 1885 ADC. And the same point of the most attenuated set (26 db black squares), corresponds to 30000 channels of the same ADC.
The figure shown above shows that the pulse shape is basically the same for the different attenuation settings. At the level of the scope picture, a deviation from linearity can not be discerned.
The figure shown above is one of the pages of log book 2 showing the gate timing for a particular tile. The pulses in these pictures are slower than the ones produced with the LED in the lab. This can be due to the long cables that bring the signal to the FEH or a slower light collection in the tiles.
During prototype test we obtained numbers of photo-electrons per MIP ranging from 8 to 30. If we use a PMT gain of 2x10^6 and pulse durations of 30 nsec, we calculate the instanteneous current produced by 100 MIPs to be equal to 20 mA, well below the 150 mA that is listed in the specification sheet of the R580-17 Hamamatsu tube used to read the tiles.
The result of this measurement indicates that the Tiles PMTs remain linear for the entire range of the LeCroy 1885 ADC.