FTCal energy calibration tutorial

This page provide instructions on how to perform energy calibration for the FT-Cal. The goal of this calibration procedure is to extract the charge-to-energy constants that are used to convert the measured signal charge for a given crystal to the energy deposition. This calibration is based on events with a pi0 decaying to two photons, both detected in the calorimeter. Events with two clusters detected in the FT-Cal are selected and filtered applying the following cuts:

• the energy of both clusters, as reconstructed based on existing calibrations, is larger than 500 MeV;
• the size of both clusters, i.e. the number of crystals involved, is larger than 3;
• the opening angle between the two clusters is larger than 2 deg.

The last cut is useful to reduce backgrounds resulting from split clusters, i.e. events in which a secondary particle originating from the electromagnetic shower creates a second cluster at a close distance to the primary cluster. For each crystal, events in which the crystal is the seed of one of the two clusters are accumulated and the ratio between 1) the measured cluster energy for the given crystal and the energy calculated from the nominal pi0 mass and 2) the other cluster energy is computed. The distribution of such ratios is fit with a Gaussian function to derive a correction factor for the charge-to-energy calibration constant of the selected crystal. The procedure is applied iteratively until the pi0 mass spectrum for all crystal is within 0.5 MeV of the nominal value.

The calibration procedure can be summarized as follows:

1. launch the calibration suite without loading initial parameters for the energy calibration; in this case, the suite will extract the energy correction factors with respect to the calibration constants currently available in CCDB for that run;
2. launch the calibration suite loading the new constants from step 1;
3. repeat step 2 loading the constants from the previous iteration until the correction factors are less than 1% and the pi0 mass is within 0.5 MeV from the nominal value.

Step 1: energy calibration with respect to the default (CCDB) constants

Launch the calibration suite. The initialization window will be displayed. Don't load any file with previous calibrations and click on Finish. Click on the H4 button at the bottom-right of the suite and load the input data file. Click on the "play arrow" on the left of the bottom bar to start the file processing. Go to the EnergyCalibration tab of the GUI: this should look like in the figure below. The 8 plots from top-left to bottom-right show:

1. The pi0 mass spectrum with the currently available calibrations integrated over all events/crystals;
2. The angle between the two clusters as a function of their invariant mass, showing the separation between the pi0 events that form a vertical line and the split clusters that accumulate in the bottom-left corner of the histogram;
3. The calibration constants correction factors as a function of the crystal ID; these are defined as the ratio between the new charge-to-energy calibration constants derived from the suite in the current iteration and the previous calibration constants;
4. The distribution of the calibration constants correction factors for all crystals (same as 3); this distribution should be centered at 1 and have a narrow width (<1%) when the calibration process has converged;
5. The pi0 mass spectrum for events in which the seed of one of the clusters is the selected crystal;
6. The measured energy in the selected crystal as a function of the calculated one as defined above;
7. The correction factor distribution for the events at 5; the gaussian fit of this distribution is what is used to. derive the correction factors and therefore the new calibration constants;
8. The distribution of the new charge-to-energy calibration constants.

Check that all the fits have properly converged selecting "View All" from the Histogram drop-down menu: the relevant fits are the ones in the right column. Ignore the fits of histograms where the maximum bin content is less than 4 counts (for such crystals the calibration constants are automatically set to a default value). Redo failed fits of the histograms with more than 4 counts by using the "Adjust fit..." option from the Histogram drop-down menu. For this purpose:

1. select the relevant crystal by clicking on the corresponding box in the detector view or the corresponding row in the table;
2. select the "Adjust fit..." option from the Histogram drop-down menu; this will open the following panel:

3. adjust the fit parameters to best match the histogram and click on "Fit"; the fit function will be updated and shown in green;
4. repeat 3 until the fit is satisfactory; if the fit keeps failing, set the amplitude, mean and sigma parameters to your best guess from looking at the histogram, select the "Fix" checkbox and click on "Fit": this will the Gaussian function parameters to the selected ones and calculate the correction factor and new calibration constant accordingly;
5. close the fit adjustment panel and repeat the procedure for the other crystals with failed fits;
6. update the table by selecting the "Update table" option from the Constants drop-down menu.

Save the constants by selecting the "Save" option from the Constants drop-down menu and quit the calibration suite.

Step 2: energy calibration with respect to the "first" iteration

Launch the calibration suite. Load the energy calibration results from Step 1 in the corresponding section of the initialization window and click on the Finish button. Load the input file and start data processing.

Since now the calibration constants that were loaded are different from the "default" ones read from CCDB, the top-left plot of the suite EnergyCalibration tab will show two histograms, the original one based on the default constants in green and the new one based on the calibration constants that were loaded in blue. Depending on whether the pi0 peak in the green histogram is above or below the nominal value, the peak in the blue histogram will be below or above it. The width of the pi0 peak in the blue histogram should be smaller than the one of the green one, indicating the spread in the individual crystals is reduced. Similarly, two histograms will be displayed in the bottom-left plot for the selected crystal: the blue histogram is the one based on the new calibration constants.

Check that all the fits have properly converged. Redo failed fits of the histograms with more than 4 counts by using the "Adjust fit..." option from the Histogram drop-down menu and following the procedure outlined above. Remember to update the table by selecting the "Update table" option from the Constants drop-down menu. Save the constants by selecting the "Save" option from the Constants drop-down menu and quit the calibration suite.

Step 3: energy calibration with respect to the "previous" iteration

Repeat the procedure outlined at step 2 until the correction factors distraction is centered at 1 within less than 1% and the pi0 mass is within 0.5 MeV from the nominal value. As the iterative process continues, the pi0 peak in the blue histogram of the top-right plot should become closer and closer to the nominal mass value and its width should decrease.

Go to the JLab electronic logbook and select the Hall B CLAS12Calib logbook: https://logbooks.jlab.org/book/CLAS12CALIB

Make a new logbook entry with subject "FTCAL energy calibration for run xxxx", including:

• The run number that was calibrated and any detail you have been given about the data file (version of the cooking, etc.);
• The snapshots of the final results;
• The text files with the energy. calibration constants from the last iteration;
• Any relevant remark on problematic fits or issues encountered.