Originally Posted by dane10
Yes, the diaphragm break-up as you suggest is my suspicion now. My first thought was that the ringing and the bump might be due to a reflection or diffraction at the throat as the wavelength was about right. I was hopeful that removing the bug screen and smoothing the transition between the throat and the entrance to the wave guide would help, but even Geddes's models which use the DE250 have a similar bump at that frequency. I also see similar ringing due to that 17 kHz bump in the impulse response of his ESP15's. He removes the bug screen and smoothes the transition between the CD and the waveguide with clay or putty. So it does imply to me it's happening within the driver, maybe at the phrase plug or the diaphram breakup as you suggest. The SEOS12's with the DE250's sound so good right now that I'm starting to think it's not worth tampering.
If you measure at different horizontal/vertical angles, you will see that the bump varies across space, which is also a symptom of break-up. In this region, the diaphragm is no longer moving like a piston.
If you have EQ capability or are able to create a passive circuit, you can suppress the break-up to some degree, but you will only be able to improve the response in a spatially-averaged sense being that the bump will vary with angle.
If done properly, I believe it can improve sound quality, depending on its severity, but keep in mind that by suppressing the break-up, you will also suppress the overall output at the top. Without compensation via a broad boost, this modification may diminish overall presence up there. That might not be a good thing.
I have practically unlimited EQ capability and am content to do a lot of measurements at different angles, so I can and do fix these things. In a passive design, I'm not sure it's worth the trouble. It's your call.