Cosmic-ray Test of MiniPlug Calorimeter

Setup
We tested a 60^o-section (wedge) of the east MiniPlug calorimeter using cosmic-ray muons. A wedge is divided into 3 "SumTowers", with each SumTower subdivided into 4 or 5 towers, as shown in Fig. 1. The wedge we tested comprises SumTowers 2, 8 and 14.

The tower outputs from a SumTower are read out by one 16-channel MAPMT (Hamamatsu, H6568-10/R5900 M16). Each tower has 3 readout channels, which are grouped together before they are read out by the front-end electronics. In addition to the outputs from the 16 anodes, the MAPMT we use has an extra output, which is the sum of the last dynodes of the 16 channels The signal from the extra output is proportional to the sum of the outputs from the 16 anodes (See Fig. 2). We call this output "sum-output".

The cosmic-ray trigger was made of a 2-fold coincidence of two scintillation counters mounted above and below the MiniPlug module. The scintillation counters were read out by RCA 8575 PMTs. You can see the picture of the cosmic-ray test setup on the right.

The front-end readout system we employed is a CAMAC based LeCroy 2249W 11-bit 12-channel charge integrating ADC, the same one used for an LED calibration of the MAPMTs. The ADC has a full scale of 512 pC and a sensitivity of 0.25 pC/count. The system was controlled by a Windows NT operated Gateway PC E4200 using National Instrument "Labview" data acquisition software.
Single Photoelectron Measurement
A measurement of the signal from single photoelectrons was carried out using a ⁶⁰Co gamma source. The source was placed just outside the MiniPlug vessel. The light emitted by the liquid scintillator (Bicron 517L) in the MiniPlug when irradiated by the source is rather week and usually results into a single photoelectron emitted by the MAPMT photocathode. We triggered on about 100,000 events using a randomly-generated gate of width 1 µs. Most triggers do not contain any signal within the gate, but a small fraction of events have a single photoelectron signal.

Three MAPMTs mounted on SumTowers 2, 8 and 14 (2KC297, 9H24D1 and 9K08C9) were operated at 960 V in this cosmic-ray test. According to specifications provided by the manufacturer, the gain of the Hamamatsu H6568-10/R5900 M16 MAPMT at HV=960 V is about G=5.0x10⁶. At this gain, the expected single photoelectron response of the LeCroy 2249W ADC unit is approximately 3.2 counts. A LRS 234 linear amplifier was used to amplify the signal by a factor of 10.

Results are shown in Figs. 3 and 4. Fig. 3(a) shows the ⁶⁰Co source signal for Tower 7. Since the tower consists of 3 PMT channels which have slightly different gains, the single photoelectron peak is broader than the resolution of each channel. For this reason, we also measured the signals from the 3 PMT channels individually, shown in Figs. 3(b), (c) and (d). These single channel signals show clearer single photoelectron peaks. Figs. 3(b), (c) and (d) show that PMT channel 5 has a higher gain than channels 6 and 7, in agreement with the results of the LED test. The distributions in Fig. 3 are fitted with a Polya distribution, i.e.
which is the appropriate one for describing single photoelectron pulse height distributions. In the fits in Fig. 3, p1 is the normalization constant, p2 the parameter m, p3 the parameter G_o, and p4 the pedestal mean value. From Fig. 3(a) we obtain 41.28/10=4.1 ADC counts for the single photoelectron response of Tower 7. Similarly, from Fig. 4, we obtain 4.1, 3.7 and 2.9 ADC counts for Towers 6, 8 and 9, respectively.
Cosmic-ray Test Results
In this cosmic-ray test, the outputs from Towers 6, 7, 8 and 9, and sum-outputs from SumTowers 2, 8and 14, were read out by the DAQ system and recorded by the computer.

The sum-output distributions for SumTower 14, 8 and 2 are shown in Fig. 5. They do not show distinctive cosmic-ray muon peaks, since cosmic-ray muons do not necessarily traverse a single SumTower. A better separation of the cosmic-ray muon peak from the pedestal is obtained by plotting the SumTower response imposing an isolation cut consisting the requirement that the other two SumTowers have a signal smaller than their own cosmic-ray muon signal. Fig. 6 shows sum-output distributions after the isolation cut, which have clearer cosmic-ray muon peaks.
However, even with this isolation cut, the cosmic-ray muon path length differences within a given SumTower still contribute to the width of the muon peak. So, although a cosmic-ray muon spectrum is expected to follow a Landau distribution, the measured distribution is a sum of many different Landau distributions and is better described by a Gaussian function. In Fig. 6, the distributions are fitted with a Gaussian function.

Figs. 7, 8 and 9 show the response from Towers 7, 6 and 8 with and without the isolation cut. The distributions with the isolation cut are fitted with a Gaussian function to obtain ADC counts for the cosmic-ray muon peak. As shown in these figures, the cosmic-ray muon peak values for Towers 7, 6 and 8 are 479.4, 435.7 and 344.9 ADC counts, respectively. Dividing these values by corresponding single photoelectron ADC counts, we obtain 116, 105 and 93 photoelectrons/MIP for Tower 7, 6 and 8, respectively.

The single photoelectron response for SumTower 8 is estimated to be about 1.1 ADC counts by averaging the single photoelectron responses of Towers 6, 7, 8 and 9 and dividing it by 3.4, which is the ratio of the sum of the anode outputs to the sum-output from the last dynodes (See Fig. 2 and LED test result). So, the yield of the cosmic-ray muon peak for SumTower 8 is estimated to be 166 (=182.6/1.1) photoelectrons/MIP. The LED test results indicate that the sum-outputs of the 3 MAPMTs mounted on SumTowers 2, 8 and 14 have approximately the same gains. So, assuming that the single photoelectron responses for SumTowers 2 and 14 are 1.1 ADC counts, we estimate the cosmic-ray muon yield for SumTowers 2 and 14 to be 175 and 172 photoelectrons, respectively.

In order to study the spread of the cosmic-ray muon signal, we plot in Fig. 10. the ratio of the light yield of Tower 7 (8) to the sum of the light yields of SumTowers 2, 8 and 14 for events whose Tower 7 (8) response is in the neighborhood of its cosmic-ray muon peak. This plot shows that cosmic-ray muons leave about 30 % of their energy in the "seed" tower.

This study is documented in CDFnote-5557
Links

The CDF Forward Detector Project Page

Kenichi Hatakeyama

Last modified: Wed Jan 21 15:13:23 EST 2004