Characterization of the stgc Detector Using the Pulser System

Transcription

1 Characterization of the stgc Detector Using the Pulser System Ian Ramirez-Berend Supervisor: Dr. Alain Bellerive Carleton University, Ottawa, Canada

2 Outline Background New Small Wheel Small-Strip Thin Gap Chambers Pulser System Overview System Requirements Implementation Testing Results The SPS for research and development 2

3 New Small Wheel Update: THE STGC DETECTOR 3

4 New Small Wheel Update Small Wheel Location Innermost component of the Muon Spectrometer system 4

5 Small-Strip Thin Gap Chamber QS3 NSW Update includes new detector type: small- Strip Thin Gap Chamber (stgc) QS2 QS1 Requires excellent angular resolution (<1mrad) for Trigger in muon spectrometer system Two wedge sizes (Large and Small), with eight modules of each, covers whole azimuthal region of the NSW Wedge broken up into three sections Carleton contributing one stgc type for each wedge size Focus on the small sector stgc: QS3 (at top of diagram) 5

6 stgc Design Basic design: Multi-Wire Proportional Chambers Two cathode planes (pads and strips) High voltage (~3kV) wires in between Gas gap of ~2.8mm Strip separation: ~3.2mm Wire separation: ~1.8mm 4 detector gaps form a quadruplet, base module of the stgc Pads: region of interest determination Fast 3 of 4 coincidence for trigger Strips: Angular resolution Gives η coordinate Wires: Gives ф coordinate 6

7 The stgc Pulser System SYSTEM OVERVIEW 7

8 The stgc Pulser System Goal: Test electrical connectivity and functionality of all readout elements (pads/strips/wires) Procedure: Pulse HV line of the chamber with square wave at 20 Volts peak-to-peak (20Vpp) measure the response signal from external electronics Collect signals from external adapter boards (AB) Process through Arduino on Pulser board (PB) and digitize with oscilloscope Sort signals to determine results Collect meta-data (amplitudes, variances, means, etc.) to study response of the stgc 8

9 Signal Processing Three Steps of Signal Processing: 1. Signal Smoothing Plots raw data from oscilloscope (noisy), stray peaks caused by noise in analog signal Using averaging function, creates smoothed waveform with clear Vpp value 9

10 Signal Processing 2. Signal Sorting Compares signals to determine if the channel is connected properly or malfunctioning Measures the peak-to-peak voltage, mean, and variance to determine if a signal is present VS. Signal will have large clear regions with high variance, and larger amplitude Can also account for false positives such as high var. noise or low amp. signal 10

11 Signal Processing 3. Signal Mapping Map signal back to two locations: Position on GFZ connector (connection between PB and AB) Position within the chamber (where AB connects to the readout element)» GFZ map organized with following legend:» Green: Channel Passed» Light Green: High Signal» Yellow: Low Signal» Red: Channel Failed» Grey: No Channel Connected Can then locate signal origin in chamber using trace path diagrams of the AB Position in chamber dependent on layer in quad and AB orientation 11

12 Implementation The stgc Pulser System GUI (graphical user interface) Serves two purposes: Configuration of the physical setup Data acquisition and processing Config. window used to communicate with the Arduino (PB) and the Scope Data window used to collect all the relevant information and then run the test, sort the data, map onto the GFZ, and locate a channel within the chamber 12

13 Testing stgc Construction Process: Cathode Board Preparation (TRIUMF) Quad Assembly (Carleton) Pulser Testing (Carleton) Cosmic Ray Testing (McGill) CERN Test at various stages of construction to ensure that none of the readout elements are malfunctioning and that all of the external electronics are correctly installed Testing identifies problematic elements which can be fixed by technicians before construction continues Ensures no gaps in the ATLAS detection system in the form of dead channels 13

14 The stgc Pulser System RESULTS 14

15 Current Results Four complete quads (numbered accordingly) with adapter boards attached Quads 1 and 2 have been fully tested (pulser and cosmic ray) and sent to CERN Quad 4 has been pulser tested and sent to McGill for cosmic ray testing Quad 3 was damaged and testing has stopped until further notice Quad 5 was closed recently, waiting for AB mounting to begin pulser testing Of the Quads that have been pulser tested: 100% of the readout elements (pads, strips, and wires) on Quads 1 and 2 are functional One strip on Quad 4 is non-functional, can have up to two per quad without great effect on overall efficiency of the NSW Relevant Measurements: At 20V and 500Hz, the current input is ~20nA, too small to damage any of the external electronics 15

16 RESEARCH AND DEVELOPMENT Attenuation Studies and Shorting Simulations 16

17 Attenuation Studies Measure the attenuation of readout element signals due to Wire Adapter Board (WAB) electronics Three layers of WAB: Bare Wires, Wire 1 Boards (includes capacitors and diodes), Wire 2 Boards (routs signals to Pad boards for processing). Results: Without W1B: Average Amplitude = 51mV ± 1mV With W1B: Average Amplitude = 27mV ± 0.6mV With W2B: Average Amplitude = 25mV ± 0.5mV Measure attenuation of wire signals due to frequency of input pulse: At low frequency: high signal distortion At high frequency: low signal amplitude Optimal range between 500 and 1000 Hz» See plot to right 17

18 Short to Ground Simulation Very clear result, complete loss of signal Easy to identify and fix Plot Legend: Signal Amplitude Sorting Function Response Variance 18

19 Shorting Simulations Short between elements: Assumed to double signal Observational result: signal averaged between shorted channels Only observed between channels with large difference in signal amplitude 19

20 Conclusion The stgc Pulser system is a necessary part of the characterization of the detector It has been crucial for identifying problems that occur during construction, and rectifying them before the process continues The pulser system, in particular the GUI, is almost at the level of automation where it can be fully implemented by technicians It has also been a useful tool for external research into the signal response of the stgc Looking Forward: a similar characterization model will be necessary for the other stgc type being built at Carleton, the QL2. 20

21 Figures: [1] ATLAS Small Wheel Location. Image From: [2] stgc Small Sector Wedge. Image From: New Small Wheel, Technical Design Report. The ATLAS Collaboration, CERN. [3] stgc Design Layout. Image From: New Small Wheel, Technical Design Report. The ATLAS Collaboration, CERN. [4] stgc Pulser Board v1.0. [5] Before/After Signal Smoothing Comparison. [6] Signal/Noise Sorting Comparison. [7] Sorted Channel Mapping to GFZ. [8] Trace Drawing of Adapter Boards for the QS3. Image From: stgc Adapter: QS3. A. Toro, The ATLAS Experiment. 21

22 Figures: [8] stgc Pulser System GUI. [10] Completed QS3 Quad with Adapter Boards attached. [11] Amplitude vs. Frequency plot for attenuation study. Produced By: S. Weber, Carleton University. [12] Chamber Amplitude Distribution with simulated short to ground. [13] Chamber amplitude Distribution with Shorted/Un-Shorted comparison. [14] New Small Wheel Overview. Image from: Performance of the new small-strip Thin Gap Chamber for the ATLAS Muon System at the LHC. A. Bellerive, Carleton University. [15] Muon Trigger Data. Image from: New Small Wheel, Technical Design Report. The ATLAS Collaboration, CERN. 22

23 EXTRA CONTENT 23

24 New Small Wheel Update One of the major upgrades intended for ATLAS during the LS2 to achieve HL-LHC. Provides Level-1 Trigger for muon spectrometer system Introduces two new detector types: small-strip thin gap chambers (stgc) and micromegas (MM) stgc: angular resolution in η and ф MM: muon tracking 24

25 NSW Motivations Plan to double peak luminosity between LS2 and HL-LHC, results in higher radiation and event pileup Increased radiation in small wheel region, old detectors deteriorating Improve trigger determination to minimize false muon triggers 90% of muon triggers in 2012 were fake (mainly late stage protons from secondary collisions) 25