The LUX Experiment Trigger and Data Acquisition Systems. Eryk Druszkiewicz April 15 th 2013

Transcription

1 The LUX Experiment Trigger and Data Acquisition Systems Eryk Druszkiewicz April 15 th 2013

2 Principle of operation Two-phase operation: Initial interaction produces scintillation light and free electrons (the number of free electrons differs for nuclear and electron interactions). Scintillation light is detected immediately in the PMTs (S1 signal). The free electrons drift through the Xenon and escape from the liquid. They are amplified in the gas region and produce a second pulse in the PMTs (S2 signal). The time difference between the S1 and S2 signals provide information about the vertical position of the interaction. 2

3 LUX DAQ & Trigger Overview Specially shaped signals for: The Struck digitizers Digital trigger CAEN discriminators Jeremy Chapman (Brown) Pre-Amp Post-Amp Trigger channel summing UC Davis Harvard 3

4 LUX DAQ Based on the Struck SIS3301: 100 MHz sampling with 14-bit resolution Customized firmware allowing for Pulse Only Digitization (POD) 1.5 khz acquisition rate with zero dead time and >99% zero suppression 95% single photoelectrons > 5σ of baseline noise > 200 kev ee dynamic range with dark matter search gains 4

5 LUX DAQ Sample Event 1.5 kev ee event 5

6 LUX Trigger Overview DDC-8DSP Trigger Builder (TB) 64 MHz sampling at 14-bit resolution Processing Mbytes/second Trigger decision reached within a few microseconds (mostly due to integration length of the filters) Four lane LVDS links over HDMI between DDCs and TB Major operation modes: S1, S2, S1&S2 We have full control over the firmware! (developed by E.D.) 6

7 Digital filter for S1 and S2 detection Primary goals: Response proportional to the area of the pulse Remove the baseline S1 filter specification: Integration width = ~ ns Dynamic range = 800 phe (linear), 2000 phe (non-linear) S2 filter specification: Integration width = ~ ns Dynamic range = phe (linear), phe (non-linear) 7

8 Trigger example Generated event S1 found S2 found 2 μs Trigger 3.1 μs Necessary to exclude the possibility of the pulse being an S2-type. Coincidence window (2 μs) + time to transfer trigger data to the TB and make the final trigger decision. 8

9 Fiducialization of events Key points of the maximum based trigger: The top PMTs provide X,Y localization We find the group of PMTs that saw the maximum signal We generate a trigger if the group belongs to the fiducial volume Top PMT trigger grouping The efficiency is energy independent. 9

10 System monitoring The DDC-8DSPs perform constant channel sweeps in the background without affecting the trigger performance. Loose cable Monitoring data is incorporated into the Slow-Control of the experiment and allows for setting automated alarms and notifications. 10

11 Ease of reconfiguration Configuration via GUIs The operation and control of the DAQ acquisitions and the trigger is closely tied with the use of a centralized MySql database. This leaves less room for operator error and also gives a precise record of all activities and usage of the system. Configuration via XML files 11

12 Final remarks Both DAQ and Trigger have been extensively tested over the last year. We suppress % of electronics noise baseline. Diagnostic tools that have been developed to monitor the data flow and the performance of the system have proven to be priceless. Being in full control of the trigger firmware has been critical for the success of the project. Main supported trigger modes: S1, S2, S1+S2, VETO large S1 signals, VETO large S2 signals, maximum signal detection and also generating gating signals for the acquisition. The flexibility of the Trigger has been important in improving data collection efficiency under a wide range of conditions. Thank you. Questions? 12