In general, I'd agree that the minimum spacing possible would provide the most uniform off-axis vertical coverage.
The larger spacing may be a compromise to mitigate the floor bounce comb filtering problems at what the designer considered to be a typical listening distance for buyers of a pretty lathed wooden exponential horn 20 years ago.
Zingali's later 3-way Client Evo 3.15 using a 10" mid-bass on top of the 12" "Omniray" exponential horn uses minimum spacing.
Most purchasers of these speakers are not likely to mess around with the digital FIR filters required to "fix" any phase problems they may have.
@weltersys
is this a MTW ?? why is not the mid in between the HF and the woofer ? 🤔
Also i completely forgot that larger drivers start to beam at higher frequencies.
Do we have some simulator/calculator for that ?
@AllenB
" High ceilings? In any case, it depends what you are doing. Would you rather a hole? "
Not sure i understand there. Not like i'd like anything in particular, was only trying to note my impressions after looking at the graphs
and playing with the app. Please correct me where i missed.
"Was phase ever a problem? "
Lol and how would i know that ?
I thought before playing with the phase value that it would have a very large effect on the patterns even with low numbers,
which it obviously didn't at all.
Then what i meant is that, if i add let's say 20degree to this number, would it give a simulation that would be the same as if one driver was some corresponding distance from the other off the front plane ?
Although here degrees will be frequency dependent and not only linear distance 🤔
VituixCAD diffraction tool does this.
VituixCAD can show you what happens 😉
But think a second, 20 degree at what frequency? phase is sliding concept as frequency, wavelength, but drivers are fixed physical objects. 20 degrees at 100Hz is 20/360 * 3.41m = ~19cm in offset and at 1000Hz it would be ~1.9cm.
Then another, If one driver is behind the other delay to on-axis can be set but now imagine looking at the system from 90 off-axis and now the distance to each driver is again the same. If the delay was done in electronics instead then the delay would be also to this off-axis position.
Each could work for some situation. If you have DSP just put them sources into same baffle and its possible to have nice frequency response all around. Some passive speakers use the offset for "time alignment", which of course works only to on-axis.
Last edited:
that software seems quite nice ... .and quite complex
i'll need some time to figure out the interface and what is what there lol
@tmuikku " But think a second, 20 degree at what frequency? phase is sliding concept as frequency, .... "
that is why i said it would be frequency dependent so yeah gets complex fast
"hen another, If one driver is behind the other delay to on-axis can be set but now imagine looking at the system from 90 off-axis "
That is a good point, from the front axis they might have a different distance but not from 90deg ...good to keep in mind for sure ! thanks 🙂
i'll need some time to figure out the interface and what is what there lol
@tmuikku " But think a second, 20 degree at what frequency? phase is sliding concept as frequency, .... "
that is why i said it would be frequency dependent so yeah gets complex fast
"hen another, If one driver is behind the other delay to on-axis can be set but now imagine looking at the system from 90 off-axis "
That is a good point, from the front axis they might have a different distance but not from 90deg ...good to keep in mind for sure ! thanks 🙂
It's a big topic but in short, I wouldn't worry about how it looks on the surface. Sometimes what looks bad doesn't sound bad and what sounds bad doesn't look bad. Some sims are good to experiment when building on knowledge, but not as a teaching tool. So if you know why you need this tool for a design, then use it. First, maybe start with overall system directivity instead.
Exponential horns directivity index (DI) increases with frequency, they also progressively "beam" at higher frequencies, the MTW puts the horn on axis for a seated listener.
The ceiling and floor bounce of the horn's high frequency waves are further back in the room compared to the lower DI at the crossover frequency.
Calculators don't accurately simulate beaming of drivers in their upper (non-pistonic) range, the apparent size of the driver is variable.
Art
Isnt this misconception? angle at which first reflection occurs does not depend on directivity, the reflection point (bounce) happens at same angle and location as with any sound source in place of the horn in this case. Trigonometry from point A to B through C. The bounce frequency response changes though, as directivity varies with frequency.
Onset of beaming is guite accurate though, usually crossover is below breakups and ideal model is reasonably accurate up to somewhere the crossover would usually be. Edge diffraction and the breakup stuff just make ripple to the ideal model. Of course these ideal models wont suffice for actual loudspeaker design work but it is very helpful to compare various sized woofers and make some general observations, like about any woofer from 15" - 3" beams where, what an array or ribbon does etc. food for hungry mind when ever there is more general question circling around the head 🙂
I don't get what you're saying...how ccan angle of first reflection not depend on directivity?
To me, every freq has an -6dB angle, as directivity is never constant, just as you say..
Wider angles have to hit hit room boundaries first. Narrower later. No ?
I try to achieve as little floor or ceiling bounce as low in freq as possible, and the same for lateral l bounce (which is much easier to achieve)
Yep, sims are still only sims, huh? 😉
@Speedysteve7 had a proper example of a full conical set. Even Romy mentioned them at the time. It's a shame that the photobucket images from the past have since expired.
This was only one of the sets. (perhaps an earlier one)
Last edited:
Right, pistonic only up to the driver's VC diameter or dustcap, whichever is largest.
I remember that thread. Hard to believe that so many horns can integrate into one cohesive soundwave. In the end Speedysteve7 sold that system, or at least tried.
Why are there no measurements of OMA horns?
@tmuikku "...food for hungry mind..." well said !
@AllenB " First, maybe start with overall system directivity instead." could you please help direct me on that ?
I am still far from having time/resources to design and build my first real system
( still have to finish the living room and that requires that i'm done with the kitchen which has the main bathroom as a prerequesite and so on ... you get the picture 😉
That is why i'm simply trying to dip my brain into the audio stuff here and there to start learning basics and try to enjoy doing it, while i have time.
Also i'd like to be able to build enough knowledge to decide on a rough idea of what my fist system will be or try to achieve etc..
I went from floor to ceiling OB line arrays like 10 years ago, to CBT array a few years ago and now i'm kinda liking horns/waveguides compresh drivers and simpler 2 way setups.
I agree about " looks " ... from my little life experience, the best " pure " designs always automatically look the part
Look at F1 cars for example .
Record breaking velomobile designs.
Most of nature stuff also... hihi
----------------------
Driver beaming : you guys are giving me so much homework i can't keep up lol
@GM " Right, pistonic only up to the driver's VC diameter or dustcap, whichever is largest." is this something i need to learn about now ? 🤔
@weltersys : "The ceiling and floor bounce of the horn's high frequency waves are further back in the room compared to the lower DI at the crossover frequency. " Not sure i understand what you mean here. What do you mean by " further back in the room " ??
Tmuikku and mark100 : don't we want less floor and ceiling bounces ? specially in the first few ms ... So anything that lowers the power of the floor/ceiling directions is good right ?
@AllenB " First, maybe start with overall system directivity instead." could you please help direct me on that ?
I am still far from having time/resources to design and build my first real system
( still have to finish the living room and that requires that i'm done with the kitchen which has the main bathroom as a prerequesite and so on ... you get the picture 😉
That is why i'm simply trying to dip my brain into the audio stuff here and there to start learning basics and try to enjoy doing it, while i have time.
Also i'd like to be able to build enough knowledge to decide on a rough idea of what my fist system will be or try to achieve etc..
I went from floor to ceiling OB line arrays like 10 years ago, to CBT array a few years ago and now i'm kinda liking horns/waveguides compresh drivers and simpler 2 way setups.
I agree about " looks " ... from my little life experience, the best " pure " designs always automatically look the part
Look at F1 cars for example .
Record breaking velomobile designs.
Most of nature stuff also... hihi
----------------------
Driver beaming : you guys are giving me so much homework i can't keep up lol
@GM " Right, pistonic only up to the driver's VC diameter or dustcap, whichever is largest." is this something i need to learn about now ? 🤔
@weltersys : "The ceiling and floor bounce of the horn's high frequency waves are further back in the room compared to the lower DI at the crossover frequency. " Not sure i understand what you mean here. What do you mean by " further back in the room " ??
Tmuikku and mark100 : don't we want less floor and ceiling bounces ? specially in the first few ms ... So anything that lowers the power of the floor/ceiling directions is good right ?
Why not begin with the standard waveguide configuration. Two way, woofer crossed to a similar sized round waveguide, then add subs. It is sufficiently challenging but it is understood and it works.
The point of reflection is on the floor or ceiling or on a wall, and the location of the reflection, which is called specular reflection, doesnt move when the directivity changes only when the speaker or listener position changes. Take a mirror and put it on the floor so that you can see your speaker from it while at listening position. Now swap the speaker to another one with different directivity and you still see it from the mirror as long as the position didn't change. Only color changed if it was different on the new one.
Of course directivity affects how all the reflections sound, the first ones, second order and late reverberant field but locations of reflections are physically the same regardless off directivity as long as stuff stays in same place physically. Only frequency response towards that reflection point changes, depending on directivity. This is how one exploits the directivity, for example toe in or tilt speaker try attenuate sound towards the reflection points, which stay in place.
Yes, but they mimic reality, reality adds more complexity on top but the underlying phenomena is the same, or good approximation off. Wavefunction or what it is, I don't even know the math, but its nothing too complicated for computers.
Last edited:
Yes, and/or as similar sound as direct sound, big constant directivity devices, or at least nice directivity devices work well. If you check out at which angles the floor and ceiling reflections happen, for your transducer height, listening distance and listening height you can then figure out how to achieve some on this. Big system, as size is needed for control long wavelengths. Array(s) or big horn for example.
Certainly makes sense not to introduce peaks, or even severe dips to these first early reflections. There are usually six boundaries on a room and from everyone there is the first specular reflection which is not attenuated much at all and for typical smallish rooms all fall within the first 5ms and color sound as hearing system cannot discriminalize them apart as separate sources but combines into one. Seventh one is the direct sound of course, but its only one from the seven. Precedence effect makes probably some kind of dominanse for the direct sound though, if it was purely for summing the seven signals any two or more of them that have similar frequency response would dominate. Hence we would like to have the direct sound signature to as many of the reflections as well.
Hearing system is amazing how it can make any sense from the noise we have flying around in a room, weed out all the hash and locate the sound sources and what not 😀 I dont know how hearing system works but no matter how it works I want to help it do its thing, not make it harder. Attenuating and making first reflections similar to direct sound makes sense but I'm not sure what is or would be most effective trick.
Last edited:
The more fundamentals of acoustics you know, the more you can 'connect the dots' as you accumulate knowledge.
