Originally Posted by markus767
Yes, that's one of the drawbacks of coax designs. There are diffraction effects from the joint between tweeter and woofer. Furthermore the woofer's cone movement is modulating the output from the tweeter.
Yes, a potential drawback of concentric drivers is diffraction of the wavefront due to the edge termination of the cone to the baffle surface. However, these effects manifest themselves mostly directly on axis. Typically by 10 deg off axis they are gone. However, the way Stereophile plots the directivity response, they choose the on axis response as the reference, then divide every other curve by that. Therefore, if the on axis response has an irregularity, and every other response was
perfectly flat, then by this approach every other curve in the plot would look uneven!
It therefore makes it more difficult to truly assess what the directivity function is. I prefer to divide through bt the 10 or 15 deg off axis response, then Only the on axis directivity response contains the irregularities, and I can then more easily see the true directivity performance.
This doesn't only affect concentric drivers. All tweeters have some degree of diffraction irregularities on axis, especially if a tweeter is not flush mounted into the baffle. But the cabinet edge causes problems (rounding of the baffle edge rarely fixes this unless the rounding radius is very large) and he frames of the grill will do this also.
Regarding modulation of the tweeter by the midrange cone, you might be surprised to know that the adjacent woofer on a conventional two way will also do this!! It is part of the baffle surface, and it is moving. The effect is not difficult to measure, and its effect is to create Doppler distortion. However, Doppler distortion has been shown to be not very audible at all. I remember back at KEF when we were developing the UNIQ, this question came up so we did some research. This is when we discovered that even the adjacent woofer can modulate the tweeter. We also recognized that if Doppler distortion is indeed audible then any speaker that reproduces simultaneously, from one diaphragm, both low and high frequencies should be subject to the effect. That means ALL large panel full range electrostatic loudspeakers!! Their diaphragms are moving up to 6mm peak to peak on bass signals, and we measured similar levels of Doppler to those that we got with UNIQ, yet know one has ever questioned the audible significance of the effect in their electrostatic speakers. At this point we decided that Doppler was largely a non-issue.
However, I did realize that there is another effect that can be significant, and that is modulation distortion due to changes in frequency response of the tweeter caused by the cone position changing the diffraction response of the tweeter wavefront. Of course, as I stated earlier, this happens primarily only directly on axis, but the war to mitigate against it is to make sure to minimize diffraction edges, but also to minimize cone displacement. This is done by limiting how far down in frequency the driver operates. For every additional octave of low frequency response, cone displacement quadruples, so I therefore pretty much always use a concentric driver only in a three way system. Crossing over at 270Hz as I do in these speakers results in cone movement that is barely observable except under very high drive levels, levels such that this form of distortion is the least of my worries!
This of course then requires a three way system , but I have no fear of three way systems. In fact, in an interview for the Absolute Sound about designing compact speakers, I stated that the best two way is a three way :-)
Designing speakers at any price point is all about choices, and how one choice will enhance or detract from some of the other choices, hence the skill comes in making those decisions that will provide the best balance between all of them. For me, concentric drivers give me more positives than they do negatives, and I think the proof has been in the reaction to the designs I have previously offered.
Of course, one negative is that it is not easy to extend the technology down to low price systems. It's not just the cost of the concentric driver, it's that you need an additional bass driver and crossover. I have not yet solved that issue :-)