To run TRF, timing calibrations must be available (at least ECA). Because the number of channels used for the measurement must be a true representation of the number that fired (otherwise false inefficiencies will appear), TRF must throw out events when even one tube has a bad calibration. Therefore it is recommended that a good calibration data set is taken and verified before the trigger efficiency runs.
TRF also requires data sets in which one or more of the raw trigger signals being measured (the ECL digital outputs of the MTC/A's) are translated to TTL and are connected to the FECD through inverting transformers. TRF can measure the efficiencies of up to five signals (all three NHIT100's and the two NHIT20's) and it will output all five to the titles file regardless of which ones are actually being measured. The Monte Carlo trigger simulation uses only the NHIT100MED trigger efficiency measurement, so the titles file archived should contain measurements of at least that trigger.
TRF can process several data files at once (to cover a wide dynamic range of laser intensities) and these must all be specified in the input command file (an example, test_trf.cmd, is provided).