The characterisation of the multianode photomultiplier tubes for the RICH-1 upgrade project at COMPASS
Introduction
The COmmon Muon and Proton Apparatus for Structure and Spectroscopy (COMPASS) experiment at CERN [1] is a fixed target experiment, providing two large acceptance spectrometer stages [2] in order to study selected topics in hadron physics. Particle identification is an important requisite of the COMPASS apparatus. Charged hadron identification up to about 50 GeV/c and in a wide angular range (about 200 mrad) is performed by RICH-1 [3]. To better cope with the high luminosity requirements, an important upgrade of the RICH-1 photon detection system has been undertaken and completed between autumn 2004 and spring 2006. This upgrade project is described somewhere else in these proceedings [4], [5], [6].
We report here about the characterisation and quality control of the multianode photomultiplier tubes (MAPMTs) forming the central part of this new photon detection set-up.
Section snippets
Multianode photomultiplier tube
In order to detect with high efficiency single photons at high beam intensities and trigger rates, the photon sensor should have rate independent response up to more than 1 MHz per pixel, provide a quantum efficiency of more than 20%, gain factors of about 107 and exhibit low dark current.
For the RICH-1 upgrade, the MAPMT type R7600-03-M16 by Hamamatsu has been chosen. It provides a common bi-alkali photocathode followed by 16 independent channels, arranged in a pixel matrix with size
Automated test set-up
To guarantee that the complete set of 576 MAPMTs, used in the upgraded RICH-1 detector, fulfills all specified parameters and that all the 9216 channels are operated at the optimum working point, a fully automated test set-up was designed and assembled (Fig. 1).
An LED pulsing system illuminates homogeneously the photocathode of the MAPMT and concurrently generates a trigger signal as gate for the charge-to-digital converter (QDC). The frequency of the LED pulses can be adjusted from 1 Hz up to 2
Results
All tested photomultiplier tubes apart 20 units accomplished all test criteria and their parameters are significantly better than the maximum allowed values. Twenty MAPMTs did not fulfill the dark current limit of less than 2 nA for each individual channel, exhibiting at most 2 out of 16 channels with higher dark current. In general, the dark current registered at the end of the 2 h measurement protocol, is an order of magnitude less than specified (Fig. 2). The uniformity behaviour turned out to
Conclusion
More than 600 MAPMTs with 16 individual channels each have been measured in terms of uniformity, gain, dark current and relative quantum efficiency in a fully automated test-stand, developed for this purpose. From the analysis of these data, the ideal working point for each individual photomultiplier tube has been extracted.
Since 2006, the RICH1 upgraded detector is taking data successfully in the COMPASS environment with excellent performance, exhibiting a resolution of the measured Cherenkov
Acknowledgements
We would like to express deep gratitude to Christian Vogel, whose knowledge concerning DAQ and Linux systems was beyond price. Thanks to Elena Gurzhiy, main responsible of the RubyOnRails project.
We acknowledge the support from CERN and the support by the BMBF (Germany) and the European Community-Research Infrastructure Activity under the FP6 programme (Hadron Physics, RII3-CT-2004-506078).
References (8)
Nucl. Instr. and Meth. A
(2007)Nucl. Instr. and Meth. A
(2005)Nucl. Instr. and Meth. A
(2008)- G. Baum, et al., The COMPASS Collaboration, Proposal, CERN/SPSLC 96-14, SPSC/P 297, March 1996, Addendum,...
Cited by (0)
- 1
On leave from JINR, Dubna, Russia.
- 2
Also at IST, Universidade Técnica de Lisboa, Lisbon, Portugal.
- 3
On leave from IHEP, Protvino, Russia.