Some clarifications are clearly in order here.
As I look at the first post in this thread, Жека introduced the H290C horn/waveguide I provide in my loudspeaker kits. I see that he showed the measurements, which are very good. He did not make them as a comparison to other horn designs, but rather an introduction of my product.
I appreciate that, Жека, by the way. I am very proud of the H290C horn/waveguide, and I am happy that you like it too. Many other people are also satisfied purchasers of this new waveguide, and I'd venture to say that the four Pi loudspeaker you are putting it in is the single most popular matched-directivity two-way loudspeaker on the market today, certainly with the DIY crowd. There are literally thousands of happy four Pi owners around the world.
The H290C horn/waveguide is not a "clone" of anything. It is a horn that I designed about ten years ago, one iteration of a handful of prototype designs for the H390C wood horn/waveguide that's been in constant production for nearly ten years now. I ended up making my wood horn/waveguide a larger device because I knew it would always be cradle mounted instead of baffle mounted. It is usually purchased for its aesthetic beauty as well as for its acoustic performance.
The design was influenced by several factors. One was a long-time realization that constant directivity was desirable, particularly in the horizontal, but that sharp edges in a horn usually made them sound "spitty" and harsh. So I had been using radial horns with smooth contours and good horizontal patterns instead.
I was also influenced by Geddes opinions about the catenary shape, which he likes to rotate on axis to form the now-famous oblate spheroidal horn that so many people talk about. There were other factors too, but I don't want to turn this into a waveguide history lesson. Anyone interested can search my forum and others for this kind of information. Frankly, I think most of you know it, at least in part.
I had spoken to Earl Geddes at one of the trade shows in some depth before I designed my H290C and H390C horn/waveguides. We continued our discussions, and at one point, I considered entering into an agreement with him to produce Prolate Spheroidal horns. Earl preferred the round OS horns, but I always prefered to limit the vertical beamwidth as well as the vertical distance between sound sources. So the PS horns were more attractive to me.
My H290C and H390C horn/waveguides use a pure catenary in both planes. The edges are also catenary flares, It is a smooth transition from round to rectangular, as is prescribed in the literature. They are mathematically pure.
The H390C has a larger section at the mouth to provide more mouth area. Since it is to be used off a baffle, I wanted more area and a widening section to counter waistbanding. The H290C has less of a roundover because it is designed to be used on a baffle, and doesn't need as much. I found this out when I built the first prototypes.
The reason I had molds made was that Eminence lost their source for the H290 horn. They did not make this horn, they bought it from a vendor. So when the vendor suddenly stopped making them, Eminence was caught with no inventory almost overnight. The Eminence H290 was a radial horn and did not use a catenary flare, but it performed well. In fact, it performed so well that I preferred it to most of the other waveguides on the market, which is why I did not switch. However, when my source dried up, I decided to "make lemonade."
I basically just dusted off the prototype I had made during H390C development, and sent the 3D model off to some plastic mold makers to get bids. I modified it first to add the bolt patterns, but kept it as much the same as possible in order to retain the same acoustic characteristics and performance. This allowed me to use the crossover designs I already had.
It's not exactly the same as the (Eminence) H290 crossover, but very close. Close enough to use it as bolt-on upgrade, which was one of my goals. All my crossover schematics have notes to show what components to use for various options. So this allowed me to leverage the design work I had already done.
This is not some hack job or clone of anything, and I resent that implication. My designs have been in production so long, it seems to me that the "other guys" would be the ones seen as making the clones. I did not copy anyone's compression driver. I did not copy anyone else's horn. I did not copy anyone else's loudspeaker design or methodology. I'm just making speakers that excite me, and hoping that others enjoy them too.
I always liked how the Econowave folks and Pi Speakers enthusiasts saw each other as "kindred spirits", and we always shared information and friendly conversation. But the SEOS crowd seems to be loaded for bear. So since you've thrown down the gauntlet, I have some comments for you.
First, you can't claim the SEOS device to be "Geddes approved". I'm not sure this is relevant, but you do seem to think it's a trump card or something. So if that's your goal, then let's get this part straight.
Geddes has said numerous times that he feels the OS shape is the best way to reduce internal reflections, the high-order modes. In his mind, the flare profile must be catenary, revolved to a minimum surface oblate spheroid. He doesn't want it truncated, elongated or squished.
Even though Geddes patented the PS waveguide
, he seems to distance himself from it. In his patent, he declares that the ratio of "squish" can be only 2/3rds or so, basically a 90x60 beamwidth. Otherwise, he is concerned that the shape might fracture the expansion enough to increase high-order modes.
The flare profile also has to be a catenary. But a catenary is like conical, in that it forms a right triangle. For a 90 degree pattern, the mouth radius MUST be the same as the depth. You can't make it too shallow, or it isn't a catenary. This also increases high-order modes. For a given mouth size, throat size and wall angle, there is a given depth. It isn't a variable, it's fixed.
Now then, I do know that Geddes sees some deformations as "forgiveable". I think he sees it as a continuum of acceptable shapes. The more the horn is squished, the more compromised it becomes, in his eyes. But it is pretty evident, at least to me, that Geddes yardstick for HOM-friendly deformations is still a bit of a guess. Nobody really knows what is acceptable or not, or how much each modification from the basic revolved catenary does.
I think we all would agree that some mouth radiusing is desireable, even though that deviates from the basic catenary shape. But I think there is some question as to whether or not deep
radiusing to reduce waistbanding is going to increase high-order modes or not. I've not seen that brought into question, but I do know that Geddes doesn't do it. That's a throwback from Mantaray and other diffraction horns, designed with pattern control as the primary (perhaps only) design goal.
So is it a problem or not to widen out the flare profile to counter waistbanding? What if it increases high-order modes? After all, it is not part of the catenary shape. It is not included in the OS flare profile, nor in the PS, for that matter. If it increases high-order modes a little bit but makes the beamwidth more consistent, is it a good thing?
Those are rhetorical questions. I'm not going to give any more of my opinions in this forum, but I am going to say that the SEOS cannot be a catenary flare because its geometry isn't right for that. At best, it is a catenary for the first couple of inches and then it flares wider, to counter waistbanding. That is what I assumed it was, anyway.
I do not want to go any further into a critique of the SEOS horns. As I said earlier, I wanted to see us as kindred spirits, much like the Econowave guys were. Maybe we can get back to that, I don't know. But I do know you have taken a position that the SEOS horn is superior, and I do not see it that way, not at all. It is a good horn, but it is not right for me or I would have used it. For now, I will leave it at that.