"Flat" efficiency curve

[LPGaffney]

The "flat" efficiency curve parameterisation for OP,RAW defined in the manual (7.21.2), uses the Gremlin parameterisation and it isn't flat! Before calling the efficiency curve, the stopping in the degrader material is calculated and passed to the efficiency curve calculation. Gremlin uses this value, while the other functions essentially ignore it by redefining it. Previously I have always used the 'Leuven' parameterisation with the following values and this appears to be truly flat, except for energies that are stopped in the degrader/absorber material:

xx 0.0 0.0 0.0 0.0 0.0 0.0 -999.9

where the "xx" at the start can be replaced by ln(1/N), where N is the number of detectors being summed together in OP,RAW. This gives an efficiency of 1.0 for the total experiment.

Of course, if somebody has made an efficiency calibration with a source, all degrader material is intrinsically included, so the efficiency curve should ignore these corrections. In fact, of the different parameterisations, only the Gremlin curve uses the absorber corrections. Maybe it is worthwhile moving the GOTO calls before this section to avoid the low-energy cut off this causes. It may be that somebody wasn't careful enough to define all of the thicknesses correctly in OP,GDET and simply relies on a "flat" efficiency curve and efficiency corrected yields. This may have consequences, however, for upper limit warnings set with UPL that rely on being defined in the raw spectra...

Regards,
Liam