1. 1 PMT Test @ Univ. of Tokyo Yasuko HISAMATSU ICEPP, MEG Collaboration meeting Feb. 10th, 2004

2. 2 PMT Output Deterioration  During the pi0 beam test @PSI, output from some PMTs reduced by 20-30% when beam was ON. (The background was estimated to be ~ 1.5*10^6p.e. per sec)  Similar change was observed under the background simulated using LED pulses -> The Deterioration of old type PMT output due to the high rate background.  Output from Al Strip Type PMT (new type) needs to be tested under the high rate background, 10^6 ~ 10^8 p.e. per sec. Yasuko HISAMATSUMEG Collaboration Meeting @ PSI February 2004

3. 3 PMT Test @ Univ. of Tokyo Set up PMT (Al Strip Type) TB0094 TB0102 TB0239 TB0268 TB0284 TB0302 TB0415 TB0439 TB0568 PMT Alpha source( 241 Am ) LED Liq. Xe 5.5MeV alpha peak ADC channel Yasuko HISAMATSUMEG Collaboration Meeting @PSI February 2004

4. 4 PMT Test facility @Univ. of Tokyo Xe tank Purification system Liq.Xe chamber alpha source LED PMT Chamber Inside

5. 5 Condition and Procedure alpha source : ~200Hz, LEDpulse height:500p.e. ~ 12000p.e. per event pulse shape: ~10nsec rate: 100Hz ~ 10KHz Trigger: alpha self trigger (veto by LED driver pulse) Procedure Pedestal Run & Gain calibration using LED Alpha Run @ LED OFF Alpha Run @ LED ON (LED : high rate background) -Change LED Pulse height, rate and PMT gain Yasuko HISAMATSUMEG Collaboration Meeting @PSI February 2004

6. 6 Condition and Procedure  DAQ: -Alpha run with LED OFF -Alpha run with LED ON  The cause of the change in PMT output  Breeder current ?  Deterioration of photocathode? Low gain run Peak/Sigma (gain independent) The change in PMT output? Yasuko HISAMATSUMEG Collaboration Meeting @PSI February 2004

7. 7 Thermal neutron background estimation Condition of the estimation  16 n/cm^2/sec (MEG TN022)  From all direction 6*10^5 photons ( : ~9MeV gamma ) are generated in LP MC 5.2 * 10^6 p.e. /PMT/sec @LP Scaled to the final detector (surface area) 3.2 * 10^7 p.e./PMT/sec @final detector = 5.0 *10^-6 A @10^6gain PMTs were tested under the background of 10^6 ~ 10^8 p.e./sec

8. 8 For instance…TB0415 5*e^5 gain peak (LED ON)/peak (LED OFF) 1*e^6 gain peak (LED ON)/peak (LED OFF) 1*e^6 gain (peak/sigma @LED ON)/ (peak/sigma @LED OFF) 10^7p.e. per sec

9. 9 Result 5*e5gain X: number of photoelectrons per second from LED Y: alpha peak (@LED ON) / alpha peak (@LED OFF) Yasuko HISAMATSUMEG Collaboration Meeting @PSI February 2004 TB0094TB0302 TB0268 TB0568TB0102 TB0415 TB0284 TB0239 TB0439

10. 10 Result 1*e6gain X: number of photoelectrons per second from LED Y: alpha peak (@LED ON) / alpha peak (@LED OFF) Yasuko HISAMATSUMEG Collaboration Meeting @PSI February 2004 TB0094TB0302 TB0268 TB0568TB0102 TB0415 TB0284TB0439 TB0239

11. 11 X: number of photoelectrons per second from LED Y: (peak/sigma @LED ON) / (peak/sigma @LED OFF) Result 5*e5gain Yasuko HISAMATSUMEG Collaboration Meeting @PSI February 2004 TB0094TB0302 TB0268 TB0568TB0102 TB0415 TB0284TB0439 TB0239

12. 12 Result 1*e6gain X: number of photoelectrons per second from LED Y: (peak/sigma @LED ON) / (peak/sigma @LED OFF) TB0094TB0302 TB0268 TB0568 TB0102 TB0415 TB0284 TB0439 TB0239 Yasuko HISAMATSUMEG Collaboration Meeting @PSI February 2004

13. 13 PMT Output vs Elapsed Time X: event number Y: ADC count TB0239 5*10^5gain Background: 8.32*10^7pe/sec 1*10^6gain Background 1.33*10^8pe/sec ~10sec~20sec  DAQ started 2 ~ 10 sec after the LED had been turned on.  Stable output up to 5*10^7 photoelectrons/sec Yasuko HISAMATSUMEG Collaboration Meeting @PSI February 2004

14. 14 Summary  9 Al Strip Type PMTs (new type) have been tested under the high rate background environment at the university of Tokyo.  So far, for the 9 PMTs, the stability of PMT output is better than 10% under the background up to 10^7 p.e. per sec.  For the 8 PMTs out of 9 PMTs, no output deterioration under background of ~1.5*10^6p.e. per sec.  For a PMT, TB0239, the output deterioration due to the breeder circuit was found under the background of ~1.9 * 10 ^6p.e. per sec.  The background number of photoelectrons should be suppressed to 10*6 p.e./sec.  Necessary to suppress thermal neutrons to 1/10.  Neutron background measurements @piE5 using LP is recommended. Yasuko HISAMATSUMEG Collaboration Meeting @PSI February 2004