• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Is baffle shape really important? What about these ideas?

GM3

Active Member
Joined
Nov 26, 2022
Messages
140
Likes
159
Just thinking, would no baffle be the best baffle? Ex; if you could just have woofers floating in the air, without any baffle, would that be best?

If so, a sort of easy way to achieve it would be to stick woofers at the end of cylinders, and maybe fix them to a sort of thin blade like stand (similar to Anthony Gallo Reference 3), so series of stacked cylinders... They would interact with each other, so I guess maybe you could try to minimize the effect using absorbent material between the woofers... Even have some sort of 'spikes', like '>' sticking out between woofers, as to absorb some of the sound waves which would bounce of other drivers... Sort of absorbing cone, but just horizontal as isolate each driver from the other..

Guess Cabasse-like design, but again looking at Cabasse, they went with spheres/spheroids, maybe they have better properties than plain cylinders? An in-between idea, something you don't see much, but sort of see with some high end speakers; is some sort of sculpting; rounding of the cabinet as to make it less square... Again, thinking out loud, couldn't you achieve this with, strategically adding small blocks in construction to allow removing some material by sculpting and to have more rounded baffle? It would add cost and work... I guess that since it's pretty rare, maybe the baffle diffraction isn't that significant? maybe a bit like perlisten, but more round... (if round helps?!)

You could even hang the aforementioned cylindrical enclosures with some wire or elastics, as to isolate vibrations from the floor.. Like drivers hanging between two towers / cranes.

Other than that, sticking to conventional box design, would sticking the woofer on top of the cabinet, and making the top part of the cabinet like an inverted U, following the woofer top plate, effectively have no baffle at all for the top part, wouldn't that be good? I'm guessing that since I've never seen such a speaker, it wouldn't...

Intuitively, wouldn't something like a really curved cabinet, like the rear of the 802 be a superior cabinet to just slim flat box with rounded edges? Like if you stuck a tweeter and a couple of 6 inch woofer in the rear of a 800d, would that be better for diffraction than a slim box cabinet with rounded edges?
Pair-edit.jpg



Basically, guess what I'm asking, what would be the optimal cabinets / baffles? I'm also wondering if diffraction is just overblown... I mean, looking at the BW 802, the top woofer is in a sort of spheroid, but, it's just sitting on top of the cabinet, which can't be good..... Can't see that spheroid top cabinet improvement not being hurt by just being 'stuck' on top of the cabinet like that... It's like a self-contradictory design, where diffraction is critical, but it isn't, at the same time...!
 
Last edited:

fpitas

Master Contributor
Forum Donor
Joined
Jul 7, 2022
Messages
9,885
Likes
14,206
Location
Northern Virginia, USA
Take a look at the Genelec line. If you're after accuracy they're hard to beat. Regular "high-end" speakers have to conform to the aesthetics de jure.
 

digitalfrost

Major Contributor
Joined
Jul 22, 2018
Messages
1,535
Likes
3,140
Location
Palatinate, Germany
This tool can simulate baffle shapes:


I think Boxsim can also do this:



If you remove the corners in Edge you will indeed find that there is no influence anymore on the sound, however realisticly no baffle is not possible yes?

1685634491332.png


Here I simulated a 180mm driver in an 160mm x 160mm baffle (I know that's impossible) and I still get some influence.
 
OP
GM3

GM3

Active Member
Joined
Nov 26, 2022
Messages
140
Likes
159
Thanks! Yeah that answers my question for the more cylindrical cabinet (rear 800d). But what about like I was saying, no baffle at all? The cylinder idea also would be that it would effectively have no baffle, so say a 6 inch woofer has an outer diameter of 6.3 inch, the cylinder outer diameter would be 6.3 inch, not a 10 inch cylinder as seems to be demonstrated in that image. Or is that what is represented here? the 'driver' seems tiny in that image...
 
OP
GM3

GM3

Active Member
Joined
Nov 26, 2022
Messages
140
Likes
159
This tool can simulate baffle shapes:


I think Boxsim can also do this:



If you remove the corners in Edge you will indeed find that there is no influence anymore on the sound, however realisticly no baffle is not possible yes?

View attachment 289628

Here I simulated a 180mm driver in an 160mm x 160mm baffle (I know that's impossible) and I still get some influence.
That's pretty interesting, thanks! "realistically no baffle is not possible", well I guess like I said, woofer 6.3 inch wide, cylinder 6.3 inch wide, and you have 'no' baffle; the baffle is just the driver casing itself... If you wanted to improve, and go fancy, the tear-like shape on the top of B&W or early Kefs, example, might also be better, but yeah building such a curved cabinet instead of just plain pipe would be another story...

So yeah it seems like the truly ideal shape is a sphere, and with the shape of the sphere, concentric drivers are where it would be at... Bit weird that there's no so much interest or enthusiasm about them? I'm guessing a sort of capsule shape for a 2 or 3 way might also work, but yeah looking at the graphs I have no idea how that would end up... (though results don't seem very instinctual...!)

Looking at the image from voodooless, it would be really hard to predict how just a top round section would behave.. Looking at the Kef Reference Model 201 I posted just above, looks like they sort had the same idea, but instead of rounding off the top of the speaker with a cylindrical inverted U, they had a spheroid shape, which strangely didn't go down at the center of woofer, instead just covering the top...?! Wonder why they did that, probably because the cylinder thing just doesn't work...? Aesthetics?

If the baffle-less cylinder would still act like a baffle and screw things up... What about using felt or some other absorbent, and putting it on the front of the driver, and even wrapping the cylinder itself with it? That would surely reduce any diffraction? Like a foam ball, that you'd put in the front of the cylinder; like this shape: ===O (where the driver would be at the center of the sphere, on the right; would stick out at the center, exactly like a Cabasse sphere driver), that would likely definitely help in having no baffle shape diffraction?
 
Last edited:

fpitas

Master Contributor
Forum Donor
Joined
Jul 7, 2022
Messages
9,885
Likes
14,206
Location
Northern Virginia, USA
Wow... That's pretty interesting, in the case of Active / DSP speakers, you could just apply the inverse FR response and .... how would that work out?!

So yeah it seems like the truly ideal shape is a sphere, And with the shape of the sphere, concentric drivers are where it would be at... Bit weird that there's no so much interest or enthusiasm about them?
You can EQ to correct the on-axis response, and that is done. But at the cost of also affecting the off-axis response.
 

digitalfrost

Major Contributor
Joined
Jul 22, 2018
Messages
1,535
Likes
3,140
Location
Palatinate, Germany
So yeah it seems like the truly ideal shape is a sphere, And with the shape of the sphere, concentric drivers are where it would be at...
With typical boxes, you can also move the chassis off-center to distribute the peaks somewhat. This is rarely done in commercial speakers, but we can see it in DIY.


sonics-argenta.jpg

(Sonics Argenta)
 

MaxwellsEq

Major Contributor
Joined
Aug 18, 2020
Messages
1,738
Likes
2,613
Just thinking, would no baffle be the best baffle? Ex; if you could just have woofers floating in the air, without any baffle, would that be best?
Just to be clear - the sound from the rear would join with the sound from the front. Depending on the diameter and the shape of the cone, at some frequencies it would reinforce the sound and at others cancel it. Infinite baffle speakers eliminate this cancellation, ported speakers enhance some of the lower frequencies.
 

gab

Member
Joined
Mar 20, 2018
Messages
45
Likes
40

"The LX521 evolved from experimentation with minimal width baffles, which can provide a more uniform dipolar radiation pattern at higher frequencies, if also suitable drivers are available. Inevitably this leads to a 4-way design, which I mostly tried to avoid in the past. The shape of the midrange/tweeter baffle was arrived at empirically for the chosen SEAS drivers and after many acoustic free-field measurements."
 

sigbergaudio

Major Contributor
Audio Company
Forum Donor
Joined
Aug 21, 2020
Messages
2,703
Likes
5,701
Location
Norway

Where is this from again? Not sure how accurate it is, based on just looking at that one would assume you get crazy response errors of +/-5dB in the 200-2khz area with a cube, which is just not the case at all.

Here's 0-30-60 on a cube speakers with just slightly rounded edges:
1685639106071.png


Here's the same on another cube speaker with hard edges:
1685639189014.png
 

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,383
Likes
18,317
Location
Netherlands
Where is this from again?
No idea, I just knew the picture, google for it, and copied the first version I found ;)
Not sure how accurate it is, based on just looking at that one would assume you get crazy response errors of +/-5dB in the 200-2khz area with a cube, which is just not the case at all.

Here's 0-30-60 on a cube speakers with just slightly rounded edges:
View attachment 289635

Here's the same on another cube speaker with hard edges:
View attachment 289636
Is this with the woofer in the middle? How large is the cube? how much smoothing is applied? Also, I'm not exactly sure how they did the old measurements...

Use a tool like The Edge, it should be much more realistic. Sadly I can't get it running anymore on my Mac :(. There is also this one: https://www.tolvan.com/index.php?page=/basta/basta.php
 

fpitas

Master Contributor
Forum Donor
Joined
Jul 7, 2022
Messages
9,885
Likes
14,206
Location
Northern Virginia, USA
Where is this from again? Not sure how accurate it is, based on just looking at that one would assume you get crazy response errors of +/-5dB in the 200-2khz area with a cube, which is just not the case at all.
I think it originally came from Olson's Acoustical Engineering. But it may predate him.
 

MAB

Major Contributor
Joined
Nov 15, 2021
Messages
2,151
Likes
4,840
Location
Portland, OR, USA
Where is this from again? Not sure how accurate it is, based on just looking at that one would assume you get crazy response errors of +/-5dB in the 200-2khz area with a cube, which is just not the case at all.
It's Harry F. Olson, Direct Radiator Loudspeaker Enclosures, edit: AES from 1950, appears to have been republished JAES Vol. 17, No. 1, 1969
 

Hugo9000

Addicted to Fun and Learning
Joined
Jul 21, 2018
Messages
575
Likes
1,754
Location
U.S.A. | Слава Україні
Where is this from again? Not sure how accurate it is, based on just looking at that one would assume you get crazy response errors of +/-5dB in the 200-2khz area with a cube, which is just not the case at all.

Here's 0-30-60 on a cube speakers with just slightly rounded edges:
View attachment 289635

Here's the same on another cube speaker with hard edges:
View attachment 289636
Dr. Harry Olson
There is an AES paper he wrote in 1950 (Direct Radiator Loudspeaker Enclosures), and versions also appear in articles he published in technical magazines of the era, as well as in his books.

 

MAB

Major Contributor
Joined
Nov 15, 2021
Messages
2,151
Likes
4,840
Location
Portland, OR, USA
Is the sphere really 'best'?
What if your design would benefit from a taper with the more gradual baffle step, and don't care about the response oscillations so much?
I don't think baffle diffraction is the only thing of importance. What if directivity matching is more important than the details of the baffle-step?
B&W are notorious for the mismatch between the high-frequency drivers in the sphere, and the low-frequency drivers in the box. So while tweeter or woofer by themselves may benefit somewhat, they end up with odd room interactions and sensitive toe-in characteristics.
 

sigbergaudio

Major Contributor
Audio Company
Forum Donor
Joined
Aug 21, 2020
Messages
2,703
Likes
5,701
Location
Norway
No idea, I just knew the picture, google for it, and copied the first version I found ;)

Is this with the woofer in the middle? How large is the cube? how much smoothing is applied? Also, I'm not exactly sure how they did the old measurements...

Use a tool like The Edge, it should be much more realistic. Sadly I can't get it running anymore on my Mac :(. There is also this one: https://www.tolvan.com/index.php?page=/basta/basta.php

No, and more of a rectangle than a cube (as most speakers). 1/24 if I'm not mistaken. Anyway, the point is that diffraction on regular, traditionally shaped speakers is typically not a massive problem.
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,446
Likes
7,955
Location
Brussels, Belgium
Where is this from again? Not sure how accurate it is, based on just looking at that one would assume you get crazy response errors of +/-5dB in the 200-2khz area with a cube, which is just not the case at all.

Here's 0-30-60 on a cube speakers with just slightly rounded edges:
View attachment 289635

Here's the same on another cube speaker with hard edges:
View attachment 289636


this obviously has baffle step compensation right? you can't really compare it then. it doesn't make sense.
 
Top Bottom