ProtoDUNE SP TPC ADC Calibration Linearity and NL Measurements - - PowerPoint PPT Presentation

ProtoDUNE SP ADC Calibration Introduction Linearity and NL Measurements Moving Forward

ProtoDUNE SP TPC ADC Calibration

Note: Slides updated since 8/15/18 presentation due to mistakes caught by David Adams Richie Diurba (Minnesota)

diurb001@umn.edu

ProtoDUNE SP ADC Calibration Introduction Linearity and NL Measurements Moving Forward

Goals

• Calibrate the ADCs for the entire detector.
• Measure gains, offsets and NL for each channel.

FELIX data was not available for APA3 so it will not be included and neither will a few bad channels on the first APA.

ProtoDUNE SP ADC Calibration Introduction Linearity and NL Measurements Moving Forward

pyADCCalib

A pyROOT package to calibrate all ADCs from raw data to linearity plots

• Takes raw data and runs LAr using RunRawDecoder.fcl.
• Finds peaks and troughs by scanning over decoded data.
• Compiles these peaks across different DAC settings to make a calibration

linear fit.

• Measures NL and creates summary plots

Methodology for scanning function can be found in last talk.

ProtoDUNE SP ADC Calibration Introduction Linearity and NL Measurements Moving Forward

Dataset

Thanks to Flavio and Karol, datasets were made Friday and Saturday.

• gain=14 mV/fC, shaping time=2 us
• Made runs with DAC settings 2, 3, 4, 5, 6, and 7.
• Each DAC step is 21.4 ke.
• For labeling purposes, the highest DAC setting run is used to label datasets.

Thanks to Tom for telling me to update my RunRawDecoder.fcl

ProtoDUNE SP ADC Calibration Introduction Linearity and NL Measurements Moving Forward

Sample Plot

Run 3282 linear fit for channel 7664

ProtoDUNE SP ADC Calibration Introduction Linearity and NL Measurements Moving Forward

Sample Plot

Run 3282 linear fit for channel 13000

Did two separate fits to minimize DNL. DB team says this is ok!

ProtoDUNE SP ADC Calibration Introduction Linearity and NL Measurements Moving Forward

Gain Measurements

Run 3282 summary plot for gains with collection channels (left) and induction channels (right)

ProtoDUNE SP ADC Calibration Introduction Linearity and NL Measurements Moving Forward

Gain Measurements

Run 3258 summary plot for gains with collection channels (left) and induction channels (right)

ProtoDUNE SP ADC Calibration Introduction Linearity and NL Measurements Moving Forward

Gain Comparisons

• Run 3282: 144.6 e/count for collection, 140.4 e/count for induction.
• Run 3258: 143.1 e/count for collection, 139.0 e/count for induction.

Has a 3% uncertainty, so any discrepancy is expected.

ProtoDUNE SP ADC Calibration Introduction Linearity and NL Measurements Moving Forward

DNL

Differential non-linearity defined as the residuals

Run 3282 DNL measurements with collection channels (left) and induction channels (right)

ProtoDUNE SP ADC Calibration Introduction Linearity and NL Measurements Moving Forward

DNL

Run 3258 DNL measurements with collection channels (left) and induction channels (right)

ProtoDUNE SP ADC Calibration Introduction Linearity and NL Measurements Moving Forward

INL

Histogram of endpoint INL, due to plotting errors I am only doing collection for Run 3282.

INL measurements for collection channels for runs ending with 3282.

Note: Ignore the abs on the stat box.

ProtoDUNE SP ADC Calibration Introduction Linearity and NL Measurements Moving Forward

Conclusion

• Gain measurements is approx. 144 e/count for collection channels and 140

e/count for induction channels.

• DNL residuals measured to be approx. 4 counts, meets BNL expectations.

ProtoDUNE SP ADC Calibration Introduction Linearity and NL Measurements Moving Forward

Moving Forward

Jon’s presentation should inform DB requirements.

• Clean code so it can be used during runs and fix INL plots.
• Bad channel metric. Preliminary considering anything with a DNL above 10

counts is bad.