Sure, you can take a spline and try to approximate a superellipse (which has zero curvature at both ends *, which is really what you need) 🙂
* I know it's not zero, but nearly so, to any practical meaning. Just not to induce another confusion.
The thing is that if you have a box with flat walls to be used, the best rounding curve will still be something like a superellipse, that connects smoothly to the walls, without curvature discontinuities. So in this sense, it doesn't really matter with what you start, in the end it will be something like what you already have, only the exact dimensions may differ. But that's what you can do (i.e. optimize) at this point anyway.
Free standing waveguides are very different in this regard, and much, much easier to optimize, of course, as they can stay axisymmetric. Actually, that's the whole reason I started using those almost exclusively. It's just so much easier and with great results.
Remember, the Tritonia waveguide was created when infinite baffle simulation was still everything we had at the time. The enclosure modeling in Ath came only after that, and then it was immediately obvious it won't be easy to keep the same quality in a (rounded) box. And then the free standing horns were implemented...
Last edited:
All the discussion about freestanding waveguides is motivating me try out an MEH configuration with a suitable (90ish degree beam width for most of the range) free standing waveguide.. 🙂
In fact this thought has been there from the beginning.. It is just the bad simulation results shown in ATH thread with MEH configuration (in terms of mid ports in the waveguide resulting in not so benign blips in polars) that held me back from trying it out..
Another reason was that everyone trying out MEHs sort of took rectangular ish sort of waveguides and got good results with them.. This is why I went back in time and started out with the Tritonia..
The only decent results that i have seen with MEH free standing guides is this one and to some extent, the Monihorn that came after it: https://www.diyaudio.com/community/...-design-the-practical-way.354772/post-7455192
Maybe I should seriously think about atleast trying out a prototype of a freestanding horn based MEH...
In fact this thought has been there from the beginning.. It is just the bad simulation results shown in ATH thread with MEH configuration (in terms of mid ports in the waveguide resulting in not so benign blips in polars) that held me back from trying it out..
Another reason was that everyone trying out MEHs sort of took rectangular ish sort of waveguides and got good results with them.. This is why I went back in time and started out with the Tritonia..
The only decent results that i have seen with MEH free standing guides is this one and to some extent, the Monihorn that came after it: https://www.diyaudio.com/community/...-design-the-practical-way.354772/post-7455192
Maybe I should seriously think about atleast trying out a prototype of a freestanding horn based MEH...
The ease is of course dependent on one's skills and aims. You do give up the choice to be in control of the radiation space if you move to the freestanding style.
Maybe I would disagree if I knew what that sentence means 🙂
@AllenB: Could you please elaborate a bit more about the 'control of radiation space' aspect?
What benefits do we get with the boxed waveguide compared to the free standing one in this particular application..
Is the concern that once we move the optimized free standing guide to a practical mounting location such as on top of a box with some overlap between part of the waveguide and the box resulting in similar sort of trade offs that I have been seeing so far.. i.e., putting the waveguide in an environment it was not optimized for..
What benefits do we get with the boxed waveguide compared to the free standing one in this particular application..
Is the concern that once we move the optimized free standing guide to a practical mounting location such as on top of a box with some overlap between part of the waveguide and the box resulting in similar sort of trade offs that I have been seeing so far.. i.e., putting the waveguide in an environment it was not optimized for..
That's not a concern. Putting a free-standing horn above / near a woofer enclosure, whatever it is, harm very little acoustically (still a lot less than making holes into the horn wall). Without this being already known, all these thoughts wouldn't make much sense, of course. That was the first thing verified.
That said, I still see a lot of potential in what you are doing now, no doubts. And I'm very curious how it will all turn out eventually.
Last edited:
The ease that a freestanding waveguide gives in design, comes from the freedom for it to keep opening in an almost infinite way. This allows diffraction to extend freely which is a way of fulfilling the goal of smooth offloading, and explains the smooth polars.
A freestanding waveguide has to be larger than other types in respect to managing lower frequencies. Patrick Bateman showed a simulation yesterday which partially shows the wider radiation, limited by the brightness settings used.. but Bjorn Kolbrek has shown that following the roundover to 360 degrees is appropriate and worthwhile for this style of horn. One example is LeCleach horns, which share the infinite opening property even if they differ in other ways to a typical waveguide.
An optimised baffled waveguide has complex choices too, that's par for the course. Building a rollback style waveguide and putting it close to objects or walls can cause it to sound restricted, and give distinct reflections.
A freestanding waveguide has to be larger than other types in respect to managing lower frequencies. Patrick Bateman showed a simulation yesterday which partially shows the wider radiation, limited by the brightness settings used.. but Bjorn Kolbrek has shown that following the roundover to 360 degrees is appropriate and worthwhile for this style of horn. One example is LeCleach horns, which share the infinite opening property even if they differ in other ways to a typical waveguide.
An optimised baffled waveguide has complex choices too, that's par for the course. Building a rollback style waveguide and putting it close to objects or walls can cause it to sound restricted, and give distinct reflections.
Finally something to disagree! 🙂 And I'm happy to say I disagree with all of it, as it counters all my data.
I'd expect you to say something like that, Marcel 😉
I have built and discovered this for myself 🙂
I have built and discovered this for myself 🙂
Yeah, the exact opposite of bascially everything you just wrote is true. This is easy to disagree with, I just don't want to go one by one and prove it wrong. It's everything in the ATH thread already.
The distinction between a freestanding and an enclosed/baffled device is a very loose one, in fact it's a continuum, and typically, starting with a simple freestanding one and making it more extended "around itself" or into a box (it's all about continuous change of the rear surface) just doesn't shift things in the right direction. And I studied maybe hundreds of comparative models like this in the last years. It may help in some special cases, typically suboptimal from the start, but it's certianly not a general rule, it's actually quite the opposite.
That said, it's pretty difficult to make a typical freestanding waveguide with a very wide radiation pattern, I'll give you that. That's where a rounded enclosure comes really handy.
The distinction between a freestanding and an enclosed/baffled device is a very loose one, in fact it's a continuum, and typically, starting with a simple freestanding one and making it more extended "around itself" or into a box (it's all about continuous change of the rear surface) just doesn't shift things in the right direction. And I studied maybe hundreds of comparative models like this in the last years. It may help in some special cases, typically suboptimal from the start, but it's certianly not a general rule, it's actually quite the opposite.
That said, it's pretty difficult to make a typical freestanding waveguide with a very wide radiation pattern, I'll give you that. That's where a rounded enclosure comes really handy.
Last edited:
I was just checking out the looks of the speaker with the new MEH waveguide...
SInce the waveguide width is less than that of the cabinet, it looks slightly odd I think.. Maybe it is a little hard to like the overall looks.. 😀
(Obviously the woofer will be black unlike the pic 😛 )
SInce the waveguide width is less than that of the cabinet, it looks slightly odd I think.. Maybe it is a little hard to like the overall looks.. 😀
(Obviously the woofer will be black unlike the pic 😛 )
Last edited:
Vineeth - One way to change the aesthetics would be to include a short trapezoid transition from the woofer cabinet to the MEH
Omit roundovers on bottom of the upper box and add them to the top of the woofer cabinet to smooth the transition. Bonus points awarded if both cabinets use the same radius on the top edges.
Thanks a lot @tktran303, @hifijim, and @HammerSandwich for the suggestions and especially the pics..
It really helps with building and refining design inspiration..
Today, over at ASR, I saw most of the concepts we discussed above in action at the Ascilab speaker page..
Their full range cardioid speaker concept (copy-pasting pics from ASR):
At his pace with speaker development, he will cover MEH teritorry as well pretty soon I think (if he has not conceptualized a system with it already.. 😀)
Anyway, back to my project, I am waiting for some wild design inspiration.. 😀
Because the B&C 4NDF16 drivers & the small B&C compression driver are very nice looking from the back side and I feel they like to shown off and not hidden completely inside a box Like I have done above..
The monihorn speaker handles that aspect pretty well I think. In those speakers, the B&C driver might be utilizing back radiaion as well for cardioidish pattern in some range I guess..
I like the looks of the round horn+rectangular cabinet as well as in my current EXAR 400 system
Just waiting for some idea to strike like in the case of my smaller cardioid speakers with SICA drivers.. 🙂
It really helps with building and refining design inspiration..
Today, over at ASR, I saw most of the concepts we discussed above in action at the Ascilab speaker page..
Their full range cardioid speaker concept (copy-pasting pics from ASR):
At his pace with speaker development, he will cover MEH teritorry as well pretty soon I think (if he has not conceptualized a system with it already.. 😀)
Anyway, back to my project, I am waiting for some wild design inspiration.. 😀
Because the B&C 4NDF16 drivers & the small B&C compression driver are very nice looking from the back side and I feel they like to shown off and not hidden completely inside a box Like I have done above..
The monihorn speaker handles that aspect pretty well I think. In those speakers, the B&C driver might be utilizing back radiaion as well for cardioidish pattern in some range I guess..
I like the looks of the round horn+rectangular cabinet as well as in my current EXAR 400 system
Just waiting for some idea to strike like in the case of my smaller cardioid speakers with SICA drivers.. 🙂
Optimizing horn/wg termination with roundovers is not necessarry in multiways, but of course it means some compromises... KEF R series, Buchardt etc. are doing just fine in boxes with sharp edges.
From https://www.erinsaudiocorner.com/loudspeakers/
From https://www.erinsaudiocorner.com/loudspeakers/
@Juhazi: I guess waveguide/horn termination might probably be secondary compared to the design of the core of the waveguide itself, if its performance is good enough for the intended frequency range it covers.. Companies are also probably bothered by overall cost considerations... But who knows may be tomorrow they might add a small termination to the waveguide also and say that those speakers now have even cleaner looking polars. Now it will just come with an added cost.. 😀
I am just playing around with all this just because we have free tools (because of the generosity of the people who have made all this available for free) and all the help available here to do so.. 🙂 To think about/implement that last bit of optimization if it can be done with not much additional burden.. Plus the math behind all is some times fascinating 🙂
I am just playing around with all this just because we have free tools (because of the generosity of the people who have made all this available for free) and all the help available here to do so.. 🙂 To think about/implement that last bit of optimization if it can be done with not much additional burden.. Plus the math behind all is some times fascinating 🙂
Integrating the circular horn into the rectangular body of the bass cabinet would look really tricked out and difficult to make. Having the edges flush, but laying on top of the bass box could have a really nice look to it.
Personally, I think the original rectangular horn on rectangular box could be fine, but making it stand out would be a challenge. Finishing it like a JBL speaker with some subtle, darker coloring and highlights may favor the design overall. Compressing these spaces would involve a lot of trickery as mentioned above-- rounding the corners, swooping the back and creating open space around the top to disguise height. The other style would be to acknowledge the scale and not highlight it by using a variety of dark colors. At this scale though, it's just like the acoustics-- the room matters most.
Personally, I think the original rectangular horn on rectangular box could be fine, but making it stand out would be a challenge. Finishing it like a JBL speaker with some subtle, darker coloring and highlights may favor the design overall. Compressing these spaces would involve a lot of trickery as mentioned above-- rounding the corners, swooping the back and creating open space around the top to disguise height. The other style would be to acknowledge the scale and not highlight it by using a variety of dark colors. At this scale though, it's just like the acoustics-- the room matters most.