TOF Data Validation

TOF Data Validation examines the quality of the TOF data. Separate summary set of histograms are generated for the scintillators installed on wall TB and on wall TC. This document contains a description of the histograms that are generated as part of this process, and what aspects of these histograms should be used when deciding to validate or not to validate a run.

Each scintillator generates four signals:

• Two pulse height signals: one measured using the PMT connected to the top of the scintillator and one measured using the PMT connected to the bottom of the scintillator.
• Two time signals derived from the pulse height signals using leading-edge discriminators.

The absence of any of these four signals makes the information provided by the other three useless. TOF Data Validation checks that all four signals are present for each scintillator.

TOF Data Validation generates a number of efficiency distributions

• E Efficiency: These distributions show the portability (in %) that the pulse height signals measured with the two PMTs mounted on top and bottom are above some user defined threshold as a function of the scintillator number (the threshold used in Data Validation is at the lower end of the MIP peak). A value of 80 indicates that if the signal of the top PMT is above threshold, there is an 80% probability that the signal of the bottom PMT is also above threshold. Since the gains for the anodes in a given PMT can vary by as much as factor of two, the variations in the efficiencies of the raw data are expected to vary between 50% and 100%.
• T Efficiency: These distributions show the portability (in %) that two prompt time signals generated by the leading-edge discriminators connected to the two PMTs, mounted on top and bottom, as a function of the scintillator number. A value of 80 indicates that if the signal of the top PMT is above threshold, there is an 80% probability is also above threshold. Since the gains for the anodes in a given PMT can vary by as much as factor of two, the variations in the efficiencies of the raw data are expected to vary between 50% and 100%.
• T and E Efficiency: These distributions show the portability (in %) that the pulse height signals measured with the two two PMTs, mounted on top and bottom are above some user defined threshold, and are in prompt coincident with the time signals generated by the corresponding discriminator channels, as a function of the scintillator number (the threshold used in Data Validation is at the lower end of the MIP peak). A value of 80 indicates that if the signal of the top PMT is above threshold, there is an 80% probability that the signal of the bottom PMT is also above threshold. and in prompt time coincidence with two timing signals from these PMTs. Since the threshold in the discriminators are set well below the MIP peak, it is expected that this distribution is very similar to the "E Efficiency" distribution.

The TOF uses PMTs with segmented anodes, and one PMT is connected to four scintillators. A failure a single PMT will thus effect the signals from four scintillators. Such failure will show up in the efficiency distributions as a dip in efficiency (to 0%) of 4 neighboring channels. A failure of a discriminator or TDC channel will produce a dip in the T Efficiency distributions, while a failure of an ADC channels will produce a dip in the E Efficiency distributions.

The TOF Data Validation generates the following additional histograms:

• # TB sensors/TB event vs SeqNumber: This histogram shows the average number of TB scintillators that generate signals for each event where there is at least one scintillator hit in Wall TB. The average number of scintillators hit depends very much on the triggers used during data collection. The main feature to look for is abrupt changes in this number in a given run.
• Fraction of events with TB hits vs SeqNumber:This histogram shows the fraction of the total number of events in a particular sequence in which at least one TB scintillator records a hit. As before, this number depends very much on the trigger conditions used during data collection. The main feature to look for are abrupt changes in this number in a given run.
• OK E and T U and D Section TB: This histogram shows the number of events in which two good pulse height and two good timing signals are recorded inn a given scintillator as function of the scintillator number.

Here is a sample of the output generated by TOF Data Validation (click on each individual histogram to get a description of its contents).

Wall TB

Wall TC

In order to validate the data please use the following guidelines:

• Look at the graphs of the average number of TB/TC scintillators and fraction of TB/TC events as function of sequence number (top left, bottom left). There should be no abrupt variations in these quantities as function of run number. If no abrupt changes are observed, continue with the data validation. If abrupt changes are observed, do not validate the run, and contact the experts.
• Look at the T and E efficiency histograms (top right) for wall TB and wall TC. The efficiency should be between 50% and 100% for all scintillators. If this is the case for wall TB and wall TC, then the TOF data can be validated. Any dips in the efficiency, such as those observed in the spectrum for wall TC. If dips are observed, determine whether similar dips are observed in the E Efficiency and T Efficiency spectra (middle top, middle bottom). Do not validate the run, and contact the experts.

Histograms Defined:

• # TB sensors/TB event vs SeqNumber: This histogram shows the average number of TB scintillators that generate signals for each event where there is at least one scintillator hit in Wall TB. The average number of scintillators hit depends very much on the triggers used during data collection. The main feature to look for is abrupt changes in this number in a given run.

• E Efficiency: These distributions show the portability (in %) that the pulse height signals measured with the two PMTs mounted on top and bottom are above some user defined threshold as a function of the scintillator number (the threshold used in Data Validation is at the lower end of the MIP peak). A value of 80 indicates that if the signal of the top PMT is above threshold, there is an 80% probability that the signal of the bottom PMT is also above threshold. Since the gains for the anodes in a given PMT can vary by as much as factor of two, the variations in the efficiencies of the raw data are expected to vary between 50% and 100%.

• T and E Efficiency: These distributions show the portability (in %) that the pulse height signals measured with the two two PMTs, mounted on top and bottom are above some user defined threshold, and are in prompt coincident with the time signals generated by the corresponding discriminator channels, as a function of the scintillator number (the threshold used in Data Validation is at the lower end of the MIP peak). A value of 80 indicates that if the signal of the top PMT is above threshold, there is an 80% probability that the signal of the bottom PMT is also above threshold. and in prompt time coincidence with two timing signals from these PMTs. Since the threshold in the discriminators are set well below the MIP peak, it is expected that this distribution is very similar to the "E Efficiency" distribution.

• Fraction of events with TB hits vs SeqNumber:This histogram shows the fraction of the total number of events in a particular sequence in which at least one TB scintillator records a hit. As before, this number depends very much on the trigger conditions used during data collection. The main feature to look for are abrupt changes in this number in a given run.

• T Efficiency: These distributions show the portability (in %) that two prompt time signals generated by the leading-edge discriminators connected to the two PMTs, mounted on top and bottom, as a function of the scintillator number. A value of 80 indicates that if the signal of the top PMT is above threshold, there is an 80% probability is also above threshold. Since the gains for the anodes in a given PMT can vary by as much as factor of two, the variations in the efficiencies of the raw data are expected to vary between 50% and 100%.

• OK E and T U and D Section TB: This histogram shows the number of events in which two good pulse height and two good timing signals are recorded inn a given scintillator as function of the scintillator number.

• # TC sensors/TC event vs SeqNumber: This histogram shows the average number of TB scintillators that generate signals for each event where there is at least one scintillator hit in Wall TC. The average number of scintillators hit depends very much on the triggers used during data collection. The main feature to look for is abrupt changes in this number in a given run.

• E Efficiency: These distributions show the portability (in %) that the pulse height signals measured with the two PMTs mounted on top and bottom are above some user defined threshold as a function of the scintillator number (the threshold used in Data Validation is at the lower end of the MIP peak). A value of 80 indicates that if the signal of the top PMT is above threshold, there is an 80% probability that the signal of the bottom PMT is also above threshold. Since the gains for the anodes in a given PMT can vary by as much as factor of two, the variations in the efficiencies of the raw data are expected to vary between 50% and 100%.

• T and E Efficiency: These distributions show the portability (in %) that the pulse height signals measured with the two two PMTs, mounted on top and bottom are above some user defined threshold, and are in prompt coincident with the time signals generated by the corresponding discriminator channels, as a function of the scintillator number (the threshold used in Data Validation is at the lower end of the MIP peak). A value of 80 indicates that if the signal of the top PMT is above threshold, there is an 80% probability that the signal of the bottom PMT is also above threshold. and in prompt time coincidence with two timing signals from these PMTs. Since the threshold in the discriminators are set well below the MIP peak, it is expected that this distribution is very similar to the "E Efficiency" distribution.

• Fraction of events with TC hits vs SeqNumber:This histogram shows the fraction of the total number of events in a particular sequence in which at least one TC scintillator records a hit. As before, this number depends very much on the trigger conditions used during data collection. The main feature to look for are abrupt changes in this number in a given run.

• T Efficiency: These distributions show the portability (in %) that two prompt time signals generated by the leading-edge discriminators connected to the two PMTs, mounted on top and bottom, as a function of the scintillator number. A value of 80 indicates that if the signal of the top PMT is above threshold, there is an 80% probability is also above threshold. Since the gains for the anodes in a given PMT can vary by as much as factor of two, the variations in the efficiencies of the raw data are expected to vary between 50% and 100%.

• OK E and T U and D Section TC: This histogram shows the number of events in which two good pulse height and two good timing signals are recorded inn a given scintillator of wall TC as function of the scintillator number.