You did, I ignored it because it was not advantageous to the use case of the design. I did fully understand it, but I was not trying to fill a room. I was filling my very specific listening position.
This is just plain not true at all. I measured a bunch of different horns and waveguides. On average, the waveguides all improved smoothed the frequency response (which is erratic naked), and reduced distortion. The worst testing one was it completely bare. I spent like a whole month doing this. I must have 20 some different designs of waveguides and 25 some different horns. I also tested a bunch of different back cups and diffusion chamber shapes. I was thorough. I would like to know why you think this though. Why do you think it being mounted flat on a baffle is best? Have you tested waveguides yourself? I am pretty sure you don't 3d print stuff so I'm not sure how you would even make the different shapes to be able to test them.
Did you ever think, and I am not saying this is true as I have zero basis of saying this, but purely theoretically, that you might be bad at making passive crossovers and extraordinarily good at making DSP ones? This would explain your entire experience.
I had a friend I used to race with. He always lost. In a circuit car on slicks, he was slow. He quit racing. No one likes to lose over and over.
About 5 years later he got a seat in a rally car. AWD and sliding all around. He was fantastic, a natural. To this day he's slow in on a circuit but if you put him in the snow he's an absolute artist.
With static friction it just never snapped into place but once the tires switched to kinetic friction, his brain just clicked.
Maybe you just found your muse.
There are other choices than what is seen here and there are some smaller brands. But there is clearly less of a mass market for large scale PA systems than there is for home hifi. Plus, the reality is that the home hifi market is much more susceptible to marketing than the Pro Audio market.
Yes. Which is strange to me because I think Digicos are inferior to many other brands, but they are flexible.
It's all DSP for sure. Has been for over 20 years. Though, I think they're all a bit more than slightly different when you get down to it.
But Clair Global owns more D&B touring rigs than anybody else now. You can get almost any flavor (see what I did there) of PA system from Clair.
What's interesting about this is that you will still see more L-Acoustics than any other brand at all levels. I would say it's the most overall preferred PA system. But only D&B has the full range cardio capabilities.
But they all operate by the same laws of physics.
Nope.
I'm very skilled with passive. I've designed about every kind of speaker there is. Analog motional feedback subwoofers to OB to TLs to back loaded horns to line arrays. Active and passive. Home audio and pro. I've done cars too. At times was obsessed with full range drivers. Designed OEM speakers as an acoustical engineer at Jensen in the 90s. Acura Vigor, Ford Probe, Jeep Cherokee, Honda Civic.
The Shaded Arrays, the Bitches Brews, and the Live Edge Dipoles which took 1st place against 58 entries at Parts Express speaker design competition in 2023... all three of those are hybrid designs that combine both active and passive.
I've posted various passive designs here for years. https://www.diyaudio.com/community/...+crossover&c[users]=perrymarshall&o=relevance
Passives can sound great. But some objectives are impossible to pull off without active.
Last edited:
Is one of these the 100 hz caused by floor bounce from woofers? Because I would love to know how to fix that
I wouldn't say floor bounce is one of the things you need DSP to solve. In fact that would tend to be the wrong way to address that problem.
I would start by reading this hoary classic paper by Roy Allison
"The Influence of Room Boundaries on
Loudspeaker Power Output"
https://audioroundtable.com/misc/Influence_of_Room_Boundaries.pdf
The way I recommend you deal with floor bounce is the way Allison Acoustics and Boston Acoustics did it back in the 80s which is a 3 way with a woofer crossed over below the frequency where the floor is more than 1/4 wavelength away, and your midrange quite a bit further up. NHT had a good approach in the 90s as well.
Or go with a 2.5 way, where you have several smaller woofers in an array down to the floor and each is a different distance. That also mitigates floor bounce.
There's no real elegant way to deal with it when you have a 2-way, other than a 2-way with a full range driver covering 200hz-20KHz at some distance away from a woofer that's near the floor.
I would start by reading this hoary classic paper by Roy Allison
"The Influence of Room Boundaries on
Loudspeaker Power Output"
https://audioroundtable.com/misc/Influence_of_Room_Boundaries.pdf
The way I recommend you deal with floor bounce is the way Allison Acoustics and Boston Acoustics did it back in the 80s which is a 3 way with a woofer crossed over below the frequency where the floor is more than 1/4 wavelength away, and your midrange quite a bit further up. NHT had a good approach in the 90s as well.
Or go with a 2.5 way, where you have several smaller woofers in an array down to the floor and each is a different distance. That also mitigates floor bounce.
There's no real elegant way to deal with it when you have a 2-way, other than a 2-way with a full range driver covering 200hz-20KHz at some distance away from a woofer that's near the floor.
There is a big difference between something that has to sit on a desk and something that can stand in a room. The design he proposed would not fit within the print area of the 3D printer.
The other design he proposed would have covered 1/3 of my upper two monitors or, I would have had to move my monitors so high up that I would strain my neck to see them.
It is for an office. I have to do actual engineering work in there. The kind of engineering work that actually pays my bills.
This was the available square area possible to be used practically which, in the engineering land I live in, is everything. Pragmatism is king.
Attachments
I will give this a read. I did make a post asking about some multiple woofer questions. One of those is how do multiple woofers hit as low as one big one. When I sim this, it never works out so maybe the low end simulators are not capable of this?
I also ask about the floor bounce thing. But I will give that paper a read.
Because of the flatter frequency response:
I don't know which planar are you using - PT6825 or PT6816?
Frequency response above is PT6816 mounted flat on a small dipole baffle. (B&G NEO8 PDR is much smoother)
Trial and error method (20+25 physical prototypes!?) is not the best way for designing waveguides/horns.
I'm not a super big fan of lots of small woofers because if you go that route you have to choose low efficiency small woofers to get a low F3, and it's hard for small drivers to move a lot of air.
But sometimes it's easier to get them to blend with a same-size midrange than getting a big woofer to blend.
I don't think the problem is the sims. Hoffman's iron law applies the same regardless of driver size.
... Back & forth, forth and back. Is too… is not…
Speaker (& what drives them and what connects them all), is a very comprosied system, which we stick into a very large et of potential domestic spaces, and listened to by a wide range of listeners, with a wide range of <taste, requirements, training, budget, domestic partner, more>
There is no best. Just disadvantages and advantages… Perry Masrshall’s audio journey — and what drove it — is an illustration of <things> change, morph, grow, shrink …
dave
Speaker (& what drives them and what connects them all), is a very comprosied system, which we stick into a very large et of potential domestic spaces, and listened to by a wide range of listeners, with a wide range of <taste, requirements, training, budget, domestic partner, more>
There is no best. Just disadvantages and advantages… Perry Masrshall’s audio journey — and what drove it — is an illustration of <things> change, morph, grow, shrink …
dave
58 entries was the total number entered in all of the categories, not just the Unlimited category in which these were placed. This means they did not directly compete with over 40 of those in attendance. Different songs and such also means they were not judged under the same sonic characteristics even if the same criteria. They were also run on a different day than the rest of the classes.
The only way this could have resulted as you say is if they had the highest judged score all weekend.
You did not state it this way in the thread about your Live Edge Dipoles. It is closer to first of 10-12 entries, not first of 59.
You are proud of your design, and have the right to be, but let's call things what they are please, okay?
They took #1 in the top category, Open Unlimited. It’s not hyperbole to say that Open Unlimited is the apex. So it’s more than fair to say they also placed better than under $300 and Dayton-only.
Per this thread, some folks would set aside the passive category. (I personally would still include it.)
So…. Top out of at least 40 entries.
https://techtalk.parts-express.com/forum/tech-talk-forum/1500242-2023-speaker-design-competition
Last edited:
By DSP, for example.
Go Horbach-Keele. It offers a quite radical and elegant solution for the floor and ceiling reflection. If well implemented, Horbach-Keele solves the root of the problem by not radiating any energy into this vector. "Simply" make the angle of the lower array Null match the angle of the floor reflection. This will very well work for a distinct hearing aera in the 3d-room. In a more elaborate variant, the array can be slightly tilted away and backwards from a standard 90° vertical geometry to have the angles of both symmetrical Nulls, e.g. the lower (floor) and the upper (ceiling) Null aiming into the respective reflection vectors. Uh ... did anyone say ceiling reflection at all? Sorry for introducing yet another complexity layer, then.
More generally, Horbach-Keele is a super example of an advanced DSP-only application. Be well aware then that Horbach-Keele is not a project for the faint of hearth, you have to really understand it's theoretical basics and you have to know what you are doing for correct results. Horbach-Keele is not suited as a beginner's quick and dirty weekend-project. And in the end, if you blame Horbach-Keele for bad results, then better blame yourself.
But why am I going on posting at all to this tread? I expect any such contribution of mine to be basically useless. Because this kind of DSP-vs-passive discussion seems to me as useless as any religion war anyway: Try to persuade a convinced croyant that he is wrong in doing so. The real truth will be revealed post mortem, anyway. In terms of "... And the DSP addicts to the right. And the passivist's section also, please. Both were wrong ...". As a conviced DSP applyer I certainly will end at the right. Or the left, maybe. I will see. See you there, then.
Last edited:
Definitely hyperbole. Open/unlimited is those that don't fit into the other, all passive categories. A mono Bluetooth ammo-box speaker would be in open/unlimited. Or a simple passive pair with an active sub.
But absolutely no need for hyperbole... Your speakers sounded great!!
The hours I've spent tinkering with those passive crossover components to achieve results that can be matched or improved in a few minutes with DSP does not mean I am done designing, making or using passive crossovers, it just means it is far more difficult and time consuming.
As a real life example, to simply compare the response of two passive second order electrical crossovers achieving the same acoustical slope with two different crossover frequencies would require substitution of a minimum of four components.
With DSP, that could be done in less time it took to write that sentence, with the passive components that could easily take 15 minutes without even doing a listening test.
"Never say never"- although it has been a while since I've built any speakers, that doesn't mean I won't again.
Most resolvable questions about multiple woofers have been asked and answered on this or the subwoofer forum.
Your tweeters could be vertically aligned with the midrange drivers using the same printer.
A restricted horizontal listening position is even more confining at a desk than in a room, as the direct to reflected ratio reduces with distance, masking off axis polar frequency response deviations.
As a guy who's sound engineering work "paid the bills" (before retirement..) the first thing I notice in a speaker design is that any vertically non-aligned driver arrangement creates unsolvable polar response problems, and require the crossover (yours look very nice) to be a compromise for one horizontal alignment axis at one specific distance.
Using stereo desktop monitors with three video monitors, my head would definitely move side to side and rotate enough where that arrangement would be unacceptable.
I can notice that problem even with a single 21" monitor, and my HF hearing is definitely far below "normal".
Art
They could not. Not with the necessary wave guide to tame the planar mid. I have hundreds of hours of CAD and design into those monitors. I tried just about every configuration in theory, and soooo many in reality. The only way would have been to offset them which would have made them 3 pieces and thoroughly screwed up the impulse alignment of the drivers.
Well this was a trade off. I'm a sloucher in my chair. I move more vertically in my chair than I do horizontally. I modelled it both ways and decided to vertical and hortizontal offset of the drivers was better to have a slight horizontal cancellation with less of a vertical hole. I am literally bobbing up and down in my chair right now as I write this so I believe I made the right decision. I almost never move side to side. I actually can't, because my desk is an L and I sit in the crux of the L
Were it up to me I'd throw an 18" in everything. Limited space though. The dual woofer design is for our small kitchen. So multiple woofers on a thin baffle is about all I can do without it looking odd. Plus, I just kind of want to try it. I have a bunch of leftover scrap plywood at the shop so I can test out all sorts of arrangements and see how they measure. That is, if I cannot just figure out how to sim it straight it away and avoid unncecessary wood work