LArIAT Calibration with Stopping Tracks LArTPC Calibration & - - PowerPoint PPT Presentation

LArIAT Calibration with Stopping Tracks

• LArTPC Calibration & Reconstruction Workshop

December 10-11, 2018 Jen Raaf

LArIAT TPC

¤ Small! Tabletop-sized, and so does not have some of the same challenges seen in larger TPCs

¤ Shorter wires à lower noise levels ¤ Shorter drift distance à less diffusion and less sensitivity to impurities

¤ Non-recirculating cryogenic purification system

¤ Single pass through filtration system to remove H2O and O2 before filling

• cryostat. LAr boils off, and cryostat

gets topped up every few hours ¤ Because of this, LArIAT sees a much larger range of argon (im)purity than most other LArTPC experiments

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

2

Caveats:

NB: LArIAT uses BNL v4* LArASICs hosted on a cold front-end motherboard designed at MSU (not shown in this image)

47 cm 40 cm 9 c m For Run-I and Run-II:

• 4-mm wire pitch
• 1 shield plane (vertical)
• 1 induction plane (+30º from vert.)
• 1 collection plane (-30º from vert.)

Before stopping track calibrations…

¤ ASIC response

¤ Check that gains & shaping times of all channels are uniform via calibrated input pulses

¤ Wire-to-wire variations

¤ Use crossing beam MIPs (at ~constant drift time) to check/calibrate variations in charge collection response seen from wire to wire.

¤ Electron lifetime correction

¤ LArIAT does not recirculate & repurify its argon. If we put dirty argon into the cryostat, we just have to correct for it …

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

3

I will not cover the following types of calibrations that we do before we try to do the ADCs-to-charge calibration:

LArIAT Run-I Preliminary Filter regeneration HV Drift Studies Filter regeneration LArIAT Run-II Preliminary

Stopping tracks

¤ Beam particles

¤ Use beamline instrumentation to ID particle types entering the TPC, and to measure their momenta ¤ Measure hit amplitude (in ADC counts), for clean samples of particles identified by beamline instrumentation ¤ Conversion from ADCs to charge via an electronic calibration factor (determined at ASIC design stage & measured on test bench). Not discussed here. ¤ Convert charge to energy, assuming Modified Box recombination model with parameters from ArgoNeuT best fit ¤ Tune calorimetry constant to make dE/dx vs. momentum agree with Bethe- Bloch expectation ¤ This method is susceptible to biases in beamline momentum measurement, but we use 100 MeV/c bins (much larger than uncertainty in momentum measurement)

¤ Cosmic muons

¤ Collect sample via light-based trigger (muon + delayed decay electron signal) ¤ Reconstruct TPC track energy loss vs. residual range ¤ Tune to achieve agreement with expectation

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

4

LArIAT has two sources of stopping tracks

Beamline particle type selection

¤ Measure momentum by determining how much a particle track bends in the magnets

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

5

LArIAT has two sources of stopping tracks

Dipole Magnets Multi-Wire Proportional Chambers (MWPCs)

TPC

LArIAT TPC & cryostat Beamline spectrometer (MWPCs + bending magnets)

Upstream MWPCs Downstream MWPCs

Beamline particle type selection

¤ a

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

6

LArIAT has two sources of stopping tracks

Dipole Magnets Multi-Wire Proportional Chambers (MWPCs)

TPC

LArIAT TPC & cryostat

m = p c c⋅TOF ℓ ⎛ ⎝ ⎜ ⎞ ⎠ ⎟

2

−1

LArIAT Data Preliminary

TOF

Energy loss vs. momentum

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

7

Strategy 1

¤ Select “light” beamline particles (pions/muons) via beamline spectrometer + TOF ¤ For each track in the TPC that is matched to a beamline track

¤ Measurements from the first 5 cm

• f this track will go into a

momentum bin that is determined by the beamline spectrometer ¤ Fill a histogram with reconstructed dE/dx for each of the spacepoints in the first 5 cm of the track ¤ Repeat for every track in the data sample ¤ Tune value of CalAreaConstant to make momentum bins match as best possible with Bethe-Bloch dE/dx prediction

Beamline track quality cuts

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

8

TPC

WC4

¤ Keep only the cleanest beamline tracks

¤ One and only one hit in each of the 4 wire chambers

¤ Reconstructed track in TPC must match nicely with beamline track entry point

¤ Project wire chamber track trajectory to front face of TPC

¤ Track in TPC must be at least 10 cm long

¤ To eliminate electrons (reconstructed as many short tracks) ¤ To avoid including pion interaction points that could affect the dE/dx, we use only the first 12 spacepoints (~5 cm) for the calibration ¤ Momentum at the most upstream face of TPC is adjusted by a flat correction factor to account for energy lost while traversing material between the 4th (downstream) wire chamber and the entry point of the TPC

DSTOF

Halo veto Beam window WC4

TPC

Beam window “Excluder” Evacuated volume

3.2 cm “dead” LAr

Positive Polarity “Light” Particles

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

9

Procedure ¤ Start with light positive particles (𝜈/𝜌), tune calorimetry constants to achieve best agreement of data dE/dx vs. momentum and predicted dE/dx vs. momentum (Bethe-Bloch) ¤ Apply same calo constants to heavier positive particles (K/p) ¤ Apply same calo constants to negative particles (𝜈/𝜌/K)

LArIAT Preliminary

Tuning & verification

¤ Fit Landau to data dE/dx distribution for each momentum bin of pion sample ¤ Adjust collection plane calorimetry constant until fit MPV matches Bethe-Bloch expected value at that momentum ¤ Choose calo constant that gives best agreement across full range of momentum bins ¤ Verify calibration by applying same calorimetry constants to kaon and proton samples à good agreement!

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

10

LArIAT Preliminary

LArIAT Preliminary LArIAT Preliminary LArIAT Preliminary

Post-tuning

¤ dE/dx vs. momentum showing all particle species ¤ Run-I pos, Run-I neg, Run-II pos, Run-II neg

¤ All Run-I with same calo constants ¤ All Run-II with same calo constants

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

11

Run-I K+ Data Run-I K- Data Run-I 𝜌- Data

LArIAT Preliminary LArIAT Preliminary LArIAT Preliminary LArIAT Preliminary

Post-tuning

¤ dE/dx vs. momentum showing all particle species ¤ Run-I pos, Run-I neg, Run-II pos, Run-II neg

¤ All Run-I with same calo constants ¤ All Run-II with same calo constants

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

12

Run-I K+ Data Run-I K- Data Run-I 𝜌- Data

Run-I tuning sample

LArIAT Preliminary LArIAT Preliminary LArIAT Preliminary LArIAT Preliminary

Run-I test samples Run-II tuning sample Run-II test samples

Calibration with stopping cosmics

¤ Michel electron sample

¤ For ~30 seconds after each beam spill, we collected cosmic ray triggers ¤ Trigger on light from muon and delayed coincidence of decay electron ¤ Reconstruct tracks in TPC ¤ GausHitFinder à TrajCluster à PMAlgTrack à Calorimetry ¤ Select events containing a single stopping 3D track ¤ Identify boundary between muon and electron by same technique as MicroBooNE (JINST 12 P09014 (2017)) ¤ Tune calorimetry constant via dE/dx vs. residual range of track

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

13

Strategy 2

Event selection

¤ Constructed muon endpoint must be close to identified start of Michel cluster

¤ Start of Michel cluster based on hit charge and local linearity conditions; endpoint of muon track from PMAlgTrack ¤ Require separation, projected along 3D track direction to be < 2 mm

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

14

For more details on reconstruction and uses of Michel electron sample, see W. Foreman’s talk earlier today.

LArIAT Preliminary Preliminary

Event selection

¤ Require hit pitch to be less than 1.2 cm ¤ Track inclination angle (angle relative to E field) must be > 20 deg

¤ Vertical tracks = 90 degrees ¤ Parallel to E field = 0 degrees ¤ Distribution of angles matches well with ArgoNeuT sample used for parameterizing Modified Box model of recombination (see B. Baller talk)

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

15

Preliminary Preliminary

Muon endpoint resolution

¤ MC study determines how well we find the muon endpoint

¤ On average, reconstructed muon endpoint is ~1.6 mm short of the true muon endpoint ¤ Correct for this reconstruction bias in calibration

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

16

Preliminary

Muon endpoint resolution

¤ Fit data slices of residual range (convoluted landau & gaussian) ¤ Assume Modified Box model recombination, R, with values suggested by ArgoNeuT: ¤ Tune calorimetry constant, Ccal, to best match expectation

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

17

dQ/dx [ADC/cm] = Ccal × dE/dx Wion × R(dE/dx, E)

Preliminary

Calibration results

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

18

¤ Fit to find Ccal ¤ Recombination model parameters fixed to ArgoNeuT values ¤ Wion = 23.6 MeV (ionization energy) ¤ Consistent with results from beamline-based calibration

dQ/dx [ADC/cm] = Ccal × dE/dx Wion × R(dE/dx, E)

Preliminary Preliminary Preliminary Preliminary

Summary

¤ LArIAT can determine calibration constants using two independent samples of events

¤ Beamline particles ¤ Stopping cosmic muons

¤ Both methods give consistent results ¤ Disadvantage: both methods must assume a recombination model

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

19

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

20

Tiank you

Extras

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

21

LArIAT cryo system (non-recirculating)

¤ a

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

22

Electronics Calibration

¤ Pedestal runs

¤ Baseline noise measurement ¤ Useful also for monitoring overall health of channels

¤ Pulser runs

¤ Injected pulses into ASIC channels used to check for cross-talk, bad channels, etc. ¤ Different ASIC gain/shaping settings for signal response and gain calibration

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

23

Electronics Calibration: Pedestals

¤ a

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

24

Before filling with LAr (RMS ~1.65 counts)

Electronics Calibration: Pedestals

¤ a

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

25

After filling with LAr (RMS ~1.17 counts)

One noisy channel in this run period (Run-IIIB), damaged during assembly of wire plane.

Electronics Calibration: Pedestals

¤ a

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

26

Blip when LAr flow stopped (pump turned

• ff and valve closed à

similar to noise seen in regular LAr top-offs of cryostat)

Finding unexpected noise sources

¤ Occasionally saw noise in LArIAT TPC, every few hours, that looked like ASIC

• corruption. Event displays

also showed “strange” behavior, most visible during pedestals (~10 pedestal values recorded before each beam spill).

¤ Correlated with LAr top-ups to cryostat (every ~4 hours) ¤ Suspect wire vibrations during turbulence of refill (valve open/pump on) ¤ Pedestal triggers very helpful in eliminating other hypotheses (e.g. beam- induced ASIC corruptions)

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

27

Signal on one wire during “noisy” period

Pulser runs

¤ Inject known signal to each channel

¤ Tests everything in readout chain except small amount of electronics between ASIC and TPC (connector and input clamping diode) ¤ Simultaneous injection on all channels shown here (left) ¤ More useful is “walking pulser” runs (right), which allow checks for cross- talk and/or miscabling, also ASIC gain/shaping time variations

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

28

Calibration of wire-to-wire variations

¤ Use crossing MIPs to check/calibrate variations in charge collection response seen from wire to wire

¤ If using cosmics, this calibration would be tied together with the electron lifetime measurement ¤ In LArIAT, have the benefit of being able to decouple electron lifetime measurement from channel-specific variations using beam particles at “constant” drift distance

¤ Example: LArIAT Run-I had a misbehaving PMT that caused reduced charge collection efficiency, especially in the central wires of the TPC (where the PMT was located); fixed before Run-II.

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

29

LArIAT Preliminary LArIAT Preliminary

Muon-based Calibrations

¤ Crossing Muons

¤ Electron lifetime measurement ¤ LArIAT does not have in-situ purity monitors, but even if we did, this would be a nice cross-check ¤ Calibration of wire-to-wire variations

¤ Stopping Muons

¤ Michel electrons for energy scale calibration (although broad E distribution)

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

30

Electron Lifetime: Multi-track Method

¤ Measure electron lifetime by slicing crossing tracks into bins of drift time, measuring attenuation of charge in bins farthest from anode planes compared to bins closest to anode planes

¤ Method used by ICARUS & LAPD ¤ Multiple tracks needed to achieve good measurement (laser would be great!) ¤ Choose size of time bins to have negligible variation in dQ/dx across bin width

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

31

Fill MPVs, fit exponential to measure attenuation Get Landau MPV from fit to dQ/dx in each time bin Loop over many tracks, filling dQ/dx histograms for relevant time bins.

Electron Lifetime: Multi-Track Method

¤ In practice, LArIAT needed 24 hours of data (~5k cosmics) to have good fits for multi-track method ¤ Caveat: variations in argon purity are much larger for LArIAT than they will likely be for DUNE (LArIAT has no recirculation system, and adds new filtered argon to cryostat every ~4 hrs)

¤ Interesting feature: electron lifetime increases (electronegative impurities decrease) as TPC operates, up to a point, and then filter saturation overcomes the improvements.

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

32

LArIAT Run-I Preliminary Filter regeneration HV Drift Studies Filter regeneration LArIAT Run-II Preliminary

Crossing Muons: Single-track method

¤ Faster but less precise measurement, suitable for quick/online monitoring ¤ In practice, we ended up not using this method in LArIAT, since we could track changes in argon purity day to day with multi-track method.

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

33

Multi-track vs. Single-track

¤ Agreement within errors for the two methods ¤ Apparent progressive trend seen for single-track method diverging from true simulated lifetime due to small size of TPC (charge attenuation at lifetimes > ~1.5ms is comparable to width of typical dQ/dx distribution)

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

34

Stopping Muons: Michel Electrons

¤ Delayed coincidence trigger on light seen by PMTs

¤ Collected light + collected charge à combine to eliminate need for recombination corrections (as are necessary for a charge-only measurement

• f deposited energy)

¤ Also using this sample to directly measure recombination probability

¤ Michel energy spectrum also gives handle on calibration

• f overall energy scale
• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

35

Light-based calibration of N2 contamination

¤ Nitrogen contamination effectively shortens the slow time component of scintillation light

¤ Fit PMT light timing distribution to extract slow component

¤ The light collection system is more sensitive than our gas analyzer

¤ Agreement between the two measurements in region where both are sensitive ¤ Trend of light analysis agrees fairly well with expectation from WARP model (2010 JINST 5 P06003)

• Dec. 10, 2018

LArIAT Calibrations with Stopping Tracks | J. Raaf

36

Gas analyzer measurement Slow time component analysis