When overhauling these files some time ago, I took a closer look at
the paper (Murrel et al.) describing the tests, and noticed that the
measured droplet diameters were median values, not Sauter means. I
assume they were volumetric median Dv50.
As the PDPA DEVC in FDS cannot measure the median diameter, I have
chosen to measure D43, which is usually quite close to median value
with the spray distribution shapes used.