I think you are on the right track with respect to cancellation due to the port as that would explain the dip I am seeing. On the rest, that is not how the system works. Klippel NFS has no visibility into the design of the speaker. It simply measures the response around the speaker and uses that to construct an equation that predicts the response. What a speaker is, is as a black box to NFS.
Now, there may be an inherent issue with the measurement but it is not due to this factor.
Thanks look it up a pity its not so easy to solve or could solve itself
Probably sounds stupid but feel its like NFS analyze has clearly seen that port output is inverted and marked as so and probobly has full IIR track of phase data, but then under compute process say for range below where gating kick in then software remove IIR track of phase because calculation will benefit in speed caculate only on amplitude curve and then it goes wrong for the part below gating range because some part of it is still marked as inverted and in phaseless FIR land summing inverted to non inverted seems makes cancelation interference, same as don't invert one of two summing 2nd order slopes if they of FIR where IIR land will work perfect optimal with one inverted.
Last edited: