The 6" woofer on the 32" wide baffle 17" tall....
pros and cons? With a 8" wide baffle it will create significantly less direct energy...in certain areas?
its down 2db or more in sensitivity in areas...how many watts is that?
@tmuikku Ty for reminding me that I can model the front baffle as a whole thing....Next on my list I guess.
@Juhazi Ty for the response, I think I answered your question in the post above this one.
pros and cons? With a 8" wide baffle it will create significantly less direct energy...in certain areas?
its down 2db or more in sensitivity in areas...how many watts is that?
@tmuikku Ty for reminding me that I can model the front baffle as a whole thing....Next on my list I guess.
@Juhazi Ty for the response, I think I answered your question in the post above this one.
Last edited:
You need to describe how you look at the direct energy as Juhazi asked. If you integrate whole 20-20kHz on-axis frequency response yeah narrow baffle with smaller transducer has less area under the line. But if you don't want the secondary sound source of the edge messing around in your passpand then use minimal baffle with appropriately sized transducer and crossover below the hump. Or just add slants / roundovers if you need to crossover above the hump, where there would be diffraction without, have the support but without the edge.
Btw. checkout what mr. Geddes has said about diffraction, there is material on his website. It is linear distortion and shows for example in group delay plot. You should avoid it if you want top performance, use less flat baffle to avoid it.
edit. saw your edited post now, yeah bigger baffle supports lower in frequency, this is right. As per the ridiculous example on my previous post with 6" and 32" drivers on 32" baffle you see the relation between baffle size and transducer size, baffle step stays the same as the baffle is equally big but there is huge difference in diffraction.
In this case, as you can crossover below the main diffraction hump and you already have 15" driver for it, you'd just scale the baffle some bigger to get some support, just be careful not to make it too big and introduce diffraction into the passband, or prepare to do some huge slants. If you made it sphere there would be no diffraction issues while still having "baffle support" to low frequency, you could make it as big as you want and the bigger the less diffraction ripple shows in the plots.
Btw. checkout what mr. Geddes has said about diffraction, there is material on his website. It is linear distortion and shows for example in group delay plot. You should avoid it if you want top performance, use less flat baffle to avoid it.
edit. saw your edited post now, yeah bigger baffle supports lower in frequency, this is right. As per the ridiculous example on my previous post with 6" and 32" drivers on 32" baffle you see the relation between baffle size and transducer size, baffle step stays the same as the baffle is equally big but there is huge difference in diffraction.
In this case, as you can crossover below the main diffraction hump and you already have 15" driver for it, you'd just scale the baffle some bigger to get some support, just be careful not to make it too big and introduce diffraction into the passband, or prepare to do some huge slants. If you made it sphere there would be no diffraction issues while still having "baffle support" to low frequency, you could make it as big as you want and the bigger the less diffraction ripple shows in the plots.
Last edited:
To me looks like it does. GD is in milliseconds and phase is in degrees. At high frequency 180 deg phase shift is only fractions of millisecond in time.
Regarding you low-mid woofer, it's passband looks like this. So, something like 2dB difference in on-axis spl and a small difference is dispersion.
When equalized, I don't think you could hear a difference in sound, until the driver starts distorting.
In your real world speaker, the 15" driver in 32x18" box is just fine. Narrower box gives no benefit, on the contrary!
Regarding you low-mid woofer, it's passband looks like this. So, something like 2dB difference in on-axis spl and a small difference is dispersion.
When equalized, I don't think you could hear a difference in sound, until the driver starts distorting.
In your real world speaker, the 15" driver in 32x18" box is just fine. Narrower box gives no benefit, on the contrary!
Last edited:
Given how many of these simulations I have run and posted the results of this comment is either disingenuous or poorly worded
This seems like a semantic discussion. If this reply is not enough it may be better to move to your own thread as this does not have much to do with the thread topic and I don't want to distract from it.
Most of the descriptions are accurate except that direct sound generally means sound that arrives at the listener without a reflection in the way. A speaker that narrows directivity sprays less sound off axis and creates a higher ratio of direct to indirect sound.
A wider baffle creates more frontal hemisphere energy but much of this ends up as indirect sound as it goes through a reflection before being heard. If you consider that "direct energy" then the mystery is solved.
This I agree with
This I don't. Just looking at the basic diffraction response doesn't tell the whole story. The previous horizontal and vertical graphs showed the difference more clearly. They are different and in the intended passband neither is bad but I would pick the narrow one. Someone else may pick the wider one, neither is wrong or right.
More than without the wider baffle?
Yes, I'm not a fan of narrow boxes on conventional speakers.
This needs more context. Narrow baffle relative to size of transducers on it will give less diffraction ripple, but it will have wider directivity in comparison to if the same transducers were on bigger baffle. Wide baffles with small drivers on it will have narrower directivity, higher DI, but also diffraction creeps in and roundovers are needed to help with that.
Just use big transducers on narrow baffle to get higher directivity and less diffraction issues. Throw in waveguide and roundovers (which would make the baffle wider) to get even more directivity and less diffraction.
The funny thing is that in this case, 65-400Hz (below Schröder) range the room pretty much sums up the on- and off-axis sound energy. Then add boundary gain and reflections messing it all up... So the difference will be hard to hear? And I suppose there will be some BS eq anyway in the design.
Yep, this is the main thing, what is relevant and what is not in an application, and common to all applications is the listener. Lets not forget that especially with shallow xo slopes the sound of a way is audible at least octave from the xo, perhaps two or more (what makes audible, 20 or 40db down? 1% distortion = 40db down). Setting xo at baffle width wavelength and response octave above that would probably be audible, not sure if diffraction on it would. It is from the system and application where the line of compromise is drawn, what is more important than something else yeah. We should always remember it is not the speaker alone that makes the sound but speakers as system with and in room and we hear sound in room. After nice brain processing there is finally perception of sound. More over we need to afford the system, and fit it into the decor in the first place!
Last edited:
Fluid, you shouldn't clip just one sentence from a message. I was discussing only camplo's midwoofer case
I agree with you and others, with upper mid and treble range narrow baffle and chamfers, roundovers are beneficial.
About BS compensation/eq - I have made several dsp-controlled multiway speakers and had them set up in different rooms. Woofer gain has to be changed up to 6-10dB to have same tonality! I have pretty much lost my interest in passive speakers...
Last edited:
I don't agree with you about that specific case, I prefer the overall shape of the directivity in the narrower baffle. Even though the large baffle doesn't do badly it tends towards having off axis that peaks up compared on on axis. This makes setting a good sound balance difficult and more SPL dependent than it has to be.
^What makes better for you in this case? Roundovers seem always help on small transducer diffraction ripple on simulator, where there is relatively much rim around the transducer. It might be 1" dome on a 4" flange for example, and it would need huge roundovers to make effect while if there was no rim not much roundover is needed to kill most if not all diffraction ripple. If you don't compensate for bafflestep then there might be reason to have it particular frequency depending on your fullrange driver depending on its response, and it perhaps wider baffle would make more sense than small difference in diffraction. Wider baffle (box) would also make narrower directivity lower in frequency, perhaps something that is a good thing. Simplified, to have better diffraction performance just keep the flat baffle area small, be it narrower enclosure or roundovers or anything else that reduces flat baffle area, but you could go great lengths on it like the thread shows
But thinking only the outside of the box don't forget to think what happens inside the box, dimensions scale there as well! Hope it helps
ps. I'm having hypothesis that making uniform diffraction across the whole bandwidth of the speaker, across all transducers, like your case with the single fullrange driver, would make the perceived sound good. Perhaps this is why people like big baffle speaker with small transducers on it even though in the simulator the diffraction performance is poor, but if diffraction approaches the same signature for all transducers on the baffle then hearing system might just filter it all out? Same with multiple entry horns and coaxial systems, even if the measurements don't look too good due to diffraction and other issues they still have good following and reportedly good enjoyable sound, just like single driver fullrange speakers. Any ideas on this, anyone know if there is studies on it? Even if it wasn't for diffraction, perhaps first room reflections are the worst and makes it for the point source systems regardless.
But thinking only the outside of the box don't forget to think what happens inside the box, dimensions scale there as well!
Hope it helpsps. I'm having hypothesis that making uniform diffraction across the whole bandwidth of the speaker, across all transducers, like your case with the single fullrange driver, would make the perceived sound good. Perhaps this is why people like big baffle speaker with small transducers on it even though in the simulator the diffraction performance is poor, but if diffraction approaches the same signature for all transducers on the baffle then hearing system might just filter it all out? Same with multiple entry horns and coaxial systems, even if the measurements don't look too good due to diffraction and other issues they still have good following and reportedly good enjoyable sound, just like single driver fullrange speakers. Any ideas on this, anyone know if there is studies on it? Even if it wasn't for diffraction, perhaps first room reflections are the worst and makes it for the point source systems regardless.
Last edited:
1.1.2.4. Direct field The direct field of a sound source is defined as that part of the sound field which has not suffered any reflection from any room surfaces or obstacles.
My definition and intent to use the words Direct energy are dead on. The energy reflected from the baffle of a woofer enclosure are apart of the direct energy field. The same is true for a waveguide....because the front baffle is a waveguide....even if not a good one. So one may argue the quality of the direct field but the baffle reflections are apart of the Direct Field UNLESS we can call the enclosure Baffles a room source....which as I stated, seems unfair if we do not do the same for a horn and waveguide, which in there output, include reflections from there own baffles, which shape the waveguide/horn!
You've said that twice now! So you must agree....Now I need figure out how @fluid sees higher directivity is creating more "indirect" sound, as I am sure he knows what he means....but has not explained....If he means the sound reflecting from the baffle, then that mystery is solved. My guess is that the physical size increase of the "source" will find more objects in the room to interact with????
In my mind This is what will lead to more room interaction.....not a wider baffle with a larger radiation size.
@fluid I am talking about the measurements that come out of akabak where it shows polar radiation per frequency. I've seen a bunch from Docali reviewing horns but I don't recall seeing the same details of a box like mine vs the opposite...If you already shared this material I apologize
Last edited:
^you've got to remember first interaction that makes actual reflection happens with the baffle edge! The reflection that happens as sound wave travels along the baffle seems to sum zero so that the there is only the initial wavefront, but magnified some due to the baffle restricting propagation to half space. But as the wavefront reaches the baffle edge diffraction happens and a "back wave" scatters, just like a reflection.
If you want only direct sound you need to crossover low enough frequency which doesn't interact with the baffle, at least not before the direct sound interacts with objects and boundaries in the room like the floor or furniture you might have. Low enough frequency would be roughly at or below bafflestep unless the baffle is wide enough to make the delays into room reflection region, few milliseconds. Or you could just remove the edge by using big enough round over.
If you need to cross over at some particular frequency and want to stay away from reflections (diffraction), say at 500Hz then you need quite a small baffle in order to reach 500Hz diffraction free. This means baffle needs be lot smaller than 500Hz, which is roughly ~70cm long. Lets be a bit cautious though that it is not the mere width but also the height, or any other dimension on the baffle that diffracts so in the end we need to consider the longest length from driver to any point at baffle edge needs be half of this, < 35cm or less to be sure no diffraction happens in the intended pass band. With baffle dimension of 1 wavelength and driver being half way you'll get maximum interference of edge diffraction you'll see in the frequency response plots, the main diffraction hump top of the bafflestep slope. So in fact you'd need to crossover at least octave below 1 wavelength of longest baffle dimension in order to stay clean from the "back wave" from baffle edge diffraction. This would make baffle smaller than 35cm for 500Hz crossover. But you won't get the baffle support then, baffle step would be up in frequency and not at the pass band so it is a compromise.
You could just make huge roundovers to kill the "back wave" instead and make big baffle back into direct energy definition, while having bafflestep low enough in frequency. It is just that huge roundovers are pain to fabricate, they too need to be in relation to wavelength. Basically a sphere is optimum, imagine fabricating 32" diameter sphere yeah, lots of work, so do what ever is practical
The ripple you see on frequency response plots due to diffraction is diffraction "back wave" interfering with the direct sound (and sound from all along the baffle edge, all of the diffraction), it is reflected sound that makes the ripple appear in frequency response plot, this is not direct energy anymore by the definition.
If you want only direct sound you need to crossover low enough frequency which doesn't interact with the baffle, at least not before the direct sound interacts with objects and boundaries in the room like the floor or furniture you might have. Low enough frequency would be roughly at or below bafflestep unless the baffle is wide enough to make the delays into room reflection region, few milliseconds. Or you could just remove the edge by using big enough round over.
If you need to cross over at some particular frequency and want to stay away from reflections (diffraction), say at 500Hz then you need quite a small baffle in order to reach 500Hz diffraction free. This means baffle needs be lot smaller than 500Hz, which is roughly ~70cm long. Lets be a bit cautious though that it is not the mere width but also the height, or any other dimension on the baffle that diffracts so in the end we need to consider the longest length from driver to any point at baffle edge needs be half of this, < 35cm or less to be sure no diffraction happens in the intended pass band. With baffle dimension of 1 wavelength and driver being half way you'll get maximum interference of edge diffraction you'll see in the frequency response plots, the main diffraction hump top of the bafflestep slope. So in fact you'd need to crossover at least octave below 1 wavelength of longest baffle dimension in order to stay clean from the "back wave" from baffle edge diffraction. This would make baffle smaller than 35cm for 500Hz crossover. But you won't get the baffle support then, baffle step would be up in frequency and not at the pass band so it is a compromise.
You could just make huge roundovers to kill the "back wave" instead and make big baffle back into direct energy definition, while having bafflestep low enough in frequency. It is just that huge roundovers are pain to fabricate, they too need to be in relation to wavelength. Basically a sphere is optimum, imagine fabricating 32" diameter sphere yeah, lots of work, so do what ever is practical
The ripple you see on frequency response plots due to diffraction is diffraction "back wave" interfering with the direct sound (and sound from all along the baffle edge, all of the diffraction), it is reflected sound that makes the ripple appear in frequency response plot, this is not direct energy anymore by the definition.
Last edited:
Found it!
1.1.2.5. Reverberant field
The reverberant field of a source is defined as that part of the sound field radiated by a source which has experienced at least one reflection from a boundary of the room or enclosure containing the source.
So I am wrong...but now I will say that horns and waveguides produce a mix of direct and indirect sound since theres at least one reflection unless it beams out of the horn/waveguide... A horn/waveguide is an enclosure containing a source
1.1.2.2. Near field
The near field of a source is the region close to a source where the sound pressure and acoustic particle velocity are not in phase. In this region the sound field does not decrease by 6 dB each time the distance from the source is increased (as it does in the far field). The near field is limited to a distance from the source equal to about a wavelength of sound or equal to three times the largest dimension of the sound source (whichever is the larger).
ok then....can I argue that a larger baffle extends the nearfield??? Since the "source" (always a mixture of direct and indirect sound, unless beaming or baffleless) is larger, when the baffle/horn/waveguide is larger....actually sounds legit
i feel like Ive learned things 😬
waveguides and horns (and woofers on baffles) direct reverberant energy...I mean I know that but I personally would not of thought to classify its output as a mix of direct field and reverberant field... I always thought the "reverberant field" came from the room only.... the more you know😅
1.1.2.5. Reverberant field
The reverberant field of a source is defined as that part of the sound field radiated by a source which has experienced at least one reflection from a boundary of the room or enclosure containing the source.
So I am wrong...but now I will say that horns and waveguides produce a mix of direct and indirect sound since theres at least one reflection unless it beams out of the horn/waveguide... A horn/waveguide is an enclosure containing a source
1.1.2.2. Near field
The near field of a source is the region close to a source where the sound pressure and acoustic particle velocity are not in phase. In this region the sound field does not decrease by 6 dB each time the distance from the source is increased (as it does in the far field). The near field is limited to a distance from the source equal to about a wavelength of sound or equal to three times the largest dimension of the sound source (whichever is the larger).
ok then....can I argue that a larger baffle extends the nearfield??? Since the "source" (always a mixture of direct and indirect sound, unless beaming or baffleless) is larger, when the baffle/horn/waveguide is larger....actually sounds legit
i feel like Ive learned things 😬
waveguides and horns (and woofers on baffles) direct reverberant energy...I mean I know that but I personally would not of thought to classify its output as a mix of direct field and reverberant field... I always thought the "reverberant field" came from the room only.... the more you know😅
Last edited:
Except when a horn/waveguide/baffle is used to create said directivity... in this case the output is already a mix of indirect and direct sound....yet the higher directivity will include less further induction of more indirect energy...in the path leading away from the horn/waveguide/baffle
So basically not all indirect energy is bad energy...and the only way to truly recieve pure direct energy is beaming👍
Last edited: