Even with the correct receiver models neither Intricon's ethos.dll (or SoundDesigns ARK for that matter) will correctly predict the total sound pressure in the ear canal. The reason for this is that in both cases only the output from the HA is modeled. The total sound pressure also includes the sound pressure that gets in on its own. Ideally we would develop a model for that acoustic pathway as well, but as a start lets assume it is simply unity gain at all frequencies. For the purposes of the on screen display we should really sum these 2 effects together. The trick is that the summation can't be done with dBs, we have to convert back to linear pressure first, and needs to mean the RMS sum

HApressure     = 10^(HA gain in dB)/20
DirectPressure = 1                                      ' assume unity at all frequencies for now
TotalPressure  = sqrt(DirectPressure^2 + HApressure^2)  ' find total power
TotaldBs       = 20 * log10(TotalPressure)              ' convert back to dBs

In the case of Ethos, there are 65 frequencies at which this needs to be calculated. Using this formula should get our on screen display to more accurately reflect on ear performance. Ideally the DirectPressure would not be fixed at 1 for all 65 frequencies but rather would be frequency dependent …Perhaps we should have use a 65 point array as a place holder to allow us to get smarter as we go along. This is actually all part of a bigger discussion about modeling and generating targets, but it is the low hanging fruit from my perspective. To validate this, we could compare this correction against measuring the FLX on the Open coupler in the Frye.