Time-of-Flight System Commissioning

From Hall D Ops Wiki
Jump to: navigation, search


Time-of-Flight is a 254cm x 254cm wall of scintillator counters organized in 2 planes (horizontal and vertical). It has a total of 92 scintillator modules and 176 PMT/readout channels. The detector commissioning goals are:

  1. Verification of hardware (cabling, possible light leaks, etc.);
  2. Verification and calibration of electronics;
  3. PMT gain balancing;
  4. Geometrical alignment;
  5. Detector efficiency measurements;
  6. Study of the magnetic field impact on PMT gains;
  7. Study of the detector performance under high rates;
  8. Timing calibration of TOF as a standalone detector;
  9. Study of time-walk calibration procedures;
  10. Verification of pattern-recognition methods for TOF hits;
  11. Measurement of the positional resolution with reconstructed tracks;
  12. Timing calibration of TOF in the framework of the overall GlueX timing;
  13. Determination of the achieved overall timing resolution.

Detector commissioning prior to beam

The following tasks will be performed prior to beam, mostly with cosmic rays.

  • Signals checks. Raise HV to minimal values (~1200 V) and verify the shapes of all signals with a scope.
  • Cabling check 1. Turn HV for each channel on and off in sequence and monitor the corresponding channel in a) the discriminator scalers; b)the CAEN TDC in free-running mode; c) fADC250 scalers (or raw signal shape). This will verify proper cabling correspondence between HV channel, discriminator channel, TDC channel and ADC channel.
  • Cabling check 2. Physically disconnect and reconnect signal cable at each PMT in turn and monitor the corresponding discriminator scaler to verify proper cabling correspondence between the geographic location of the PMT and all electronics cabling verified in the previous step.
  • Light leak test.Measure and compare scaler rates with lights on and lights off (darkness) in the Hall to test for possible light leaks.
  • Signal reflection tests.Verify fADC signal shapes for each channel for the absence of signal reflections at the passive splitter.
  • High-voltage test. Raise HV in small steps from the minimal to the predetermined optimal value and monitor scaler rates for any sudden increase in rates due to light leak, cathode discharge, ringing, etc.
  • fADC calibration.With cosmic trigger, measure ADC pedestals (baseline) and adjust each individual DAC to have all pedestals at 200 counts. Make multiple measurements at different times to study pedestal stability. Optimize fADC latency and timing window to fit the raw shape of all triggered signals.
  • Discriminator calibration.By varying discriminator thresholds and monitoring the minimal magnitude of the accepted fADC signal, establish the scale of discriminator thresholds in mV. Optimize and universally equalize the thresholds for all channels.
  • CAEN TDC calibration.Put TDC in the free-running (continious storage) mode and analyze data for equalization of RC-delays. Adjust if necessary. Optimize TDC timing window and latency to accept all triggered signals but reduce background.
  • Cosmic data calibrations.
    • Collect about 1M cosmic events. Verify the presence of straight cosmic tracks on the event display.
    • Verify pattern-recognition algorithm which should find such tracks.
    • The first calibration is done for each plane independently. The fitted per channel parameters are timing offsets and time-walk parameters. The fitted per module parameters are effective speed of light and module's gap (displacement) from the neighbouring channel. Hit location along a module is determined from the time difference of the signals from the opposite ends. Each recognized track is fitted as a straight line with the least square method, and the overall minimization quantity is the sum of all distances from hit location to such straight lines over all hits in all events.
    • The second calibration requires matching portions of the same tracks in both planes. The fitted parameters are the single timing offset and xy-displacement of two planes relative to each other.
    • The third calibration step will involve half-length modules with PMT on one end only (which were excluded in the previous step). The fit of their parameters will minimize the difference between hit locations established from their timing along and from tracks found in the full-length modules.
    • At this point, TOF should be time-calibrated as a standalone detector. First determination of the per-module positional resolution can be made by comparing fitted tracks lines and positions found from the time difference.
  • Gain balancing.With good positional resolution achieved, the cosmics will be used to collect hits in the vicinity (6cm) of each PMT (to eliminate light attenuation effects). Average ADC spectrum of minimum ionizing particles in such hits will be measured. HV for each channel will be iteratively adjusted until average ADC spectra are similar in each channel.

TOF commissioning with beam and no field

  • Rates.Measure per-module rates and occupancy and compare with the ones expected from the azimuthal symmetry of charged tracks. Make such measurements at a few values of beam current (from low to high) to verify that all rates are proportional to the current and there is no change in normalized distribution of hits on TOF surface (i.e., no rate-dependence in the performance of PMT/LED/TDC/ADC). Check fADC shapes and average spectra for the sign of signal pile-up or deterioration with the increased beam current.
  • Geometrical alignment.This step will require straight tracks reconstructed in FDC (and, possibly, CDC, depending on the upstream target position). First, the location of the "average" hit location along each module's width should be established from the shape of the overall hit distribution on the face of TOF (such "average" hit location may not necessarily be at the module's center). Next, pattern recognition should be verified in order to find the approximate correspondence between TOF hit locations and reconstructed track positions. Finally, each module's geometrical displacements should be fitted to minimize the difference between maximum of track location distribution and expected "average" hit location.
  • Refined gain balancing.HV adjustment should be repeated with much more precise selection of hits right next to each PMT. However, the TOF corners may not be sufficiently illuminated. A different procedure should be studied in each all hits are used to equalize fADC spectra after accounting for light attenuation of each hit.
  • Refined timing calibration.First, step 2 of cosmic calibration (time matching of two planes) should be repeated because of better matching of hits in two planes with straight beam). Second, two approaches will be tried for calibrating TOF to the overall GlueX timing from the start counter: minimizing a single time offset of TOF as a self-contained detector and minimizing each channel timing offset individually.

TOF commissioning with beam and solenoid field

  • Remeasure rates and compare with field off at the same beam current.
  • Compare fADC spectra for hits at the same location as in field off measurements. Notice any change in the PMT gains.
  • Redo gain balancing through HV adjustment. Notice and investigate any significant change in HV values from field off to field on conditions.

--- to be conitnued ----