Exactly! The deeper I go into this it seems that waveguides are really one of the only ways to go if you really wanted to control it. The whole "if you can't beat em' join em' " strategy. At least this way you get to control the narrative. But I still am looking for some very indepth information on this. I just came across the
Huygens Principle which is likely the origin of science on diffraction, though it was for light it is likely used for audio as well. Interesting to find this. Compression drivers likely use something like this to tune their drivers.
Yeah sound is wave like light. Impedance changes, part of wave transmit and part reflects back. Simplified, rounding of corners ease out the impedance change and more transmits, less reflects. Cant get rid of diffraction (wave going aroubd corner) but the reflection, aka back wave, is the problem and gets reduced in amplitude and bandwidth.
Also felt makes impedance change, part of sound transmits to it and part reflects.
Also felt makes impedance change, part of sound transmits to it and part reflects.
Last edited by a moderator:
What do you mean when you say impedance change. There is no impedance after the driver produces the sound.
Acoustic impedance changes at the edge. Simplified, sound is pressure wave propagating spherically to all directions, from the transducer in case of loudspeaker. Baffle restricts sound propagation to half sphere increasing the pressure. Now at the edge the baffle disappears and pressure drops, acoustic impedance drops.
At least this is how I've understood and imagined how it happens. You could read Olson paper about edge diffraction on loudspeakers https://www.aes.org/e-lib/browse.cfm?elib=17816
There is also few threads on it on the forum if you search, and much info scattered around. Google doesnt find too much in depth explanations, perhaps someone with a degree in acoustics could help provide more info / resources. There is paper by Tore Skogberg about loudspeaker cabinet diffraction that has some math to it.
At least this is how I've understood and imagined how it happens. You could read Olson paper about edge diffraction on loudspeakers https://www.aes.org/e-lib/browse.cfm?elib=17816
There is also few threads on it on the forum if you search, and much info scattered around. Google doesnt find too much in depth explanations, perhaps someone with a degree in acoustics could help provide more info / resources. There is paper by Tore Skogberg about loudspeaker cabinet diffraction that has some math to it.
Late night thought experiment, imagination to work as its fun 🙂
The baffle is kind of temporary pressure doubler for wavelenghts that are roughly its size or smaller. Think a sound "bubble" whose diameter is its wavelength. This bubble starts from a point source on a flat baffle and is thus a hemisphere. As long as it wholly fits on the baffle half of the full spherical bubble is reflected from baffle and doubles up the other half, we now have sphere folded into a hemisphere, with doubled sound pressure.
Longer wavelengths get less pressure increase as they dont fully fit on the baffle and only fraction of the whole sphere reflects and adds up pressure. Long enough wavelength the speaker is basically invisible, only very small portion of this big bubble is reflected from baffle (as baffle is small in size) and thus doesn't increase pressure mucho.
As our bubble keeps propagating past the edge this temporary pressure increase now off, and sound propagates to all directions with the original pressure, not doubled anymore. With the pressure drop part of the sound reflects, part transmits.
The baffle is kind of temporary pressure doubler for wavelenghts that are roughly its size or smaller. Think a sound "bubble" whose diameter is its wavelength. This bubble starts from a point source on a flat baffle and is thus a hemisphere. As long as it wholly fits on the baffle half of the full spherical bubble is reflected from baffle and doubles up the other half, we now have sphere folded into a hemisphere, with doubled sound pressure.
Longer wavelengths get less pressure increase as they dont fully fit on the baffle and only fraction of the whole sphere reflects and adds up pressure. Long enough wavelength the speaker is basically invisible, only very small portion of this big bubble is reflected from baffle (as baffle is small in size) and thus doesn't increase pressure mucho.
As our bubble keeps propagating past the edge this temporary pressure increase now off, and sound propagates to all directions with the original pressure, not doubled anymore. With the pressure drop part of the sound reflects, part transmits.
I'm a new poster, please bear with me. The issue is the baffle width and back wave distortion on the driver cone?
I suggest a much narrower baffle face with at least 1/2" round over on the corners, The correct phase plug to take care of the returning back waves on the driver face. The issue you're having, you're creating by design. Wide baffles can work but the flatter the front baffle the more prone it is to collecting vs a stream line approach of glancing the waves in the direction you want. AWAY from the driver and cabinet. You don't want to absorb anything on the outside of the cabinet, that is the way OUT not IN. Graphite for speed and WCF for almost zero cone flex and very light.
WCF Sharkskin you can see the holes in the fabric between the weave. Add a fixed phase plug you have made good, even better. It's the design. A silk purse. You have to start with silk.
Regards
I suggest a much narrower baffle face with at least 1/2" round over on the corners, The correct phase plug to take care of the returning back waves on the driver face. The issue you're having, you're creating by design. Wide baffles can work but the flatter the front baffle the more prone it is to collecting vs a stream line approach of glancing the waves in the direction you want. AWAY from the driver and cabinet. You don't want to absorb anything on the outside of the cabinet, that is the way OUT not IN. Graphite for speed and WCF for almost zero cone flex and very light.
WCF Sharkskin you can see the holes in the fabric between the weave. Add a fixed phase plug you have made good, even better. It's the design. A silk purse. You have to start with silk.
Regards
I'm building a HT / family room where in-wall speakers will be installed in infinite baffles fabricated out of "Tecktum" or similar rigid sound attenuation board, speaker fabric wrapped (& felt where appropriate). I'm thinking this eliminates edge diffraction while the also diminishing pitfalls of infinite baffle speakers by fabricating virtually all walls and ceiling surfaces out of material w/ sound reduction coefficient of .75 - .80 and most importantly contributing to WAF.
Do any contributors have direct experience with any of the above (other than WAF issues)?
Add some diffraction panels too?
Inquiring mind wants to know!
Do any contributors have direct experience with any of the above (other than WAF issues)?
Add some diffraction panels too?
Inquiring mind wants to know!
I use 12" sonotube on my back wall.
Not cut in half, but leave them fully round, floor to higher than my ears, 4" between tubes.
Doesn't cost much, doesn't time any time to make.
I think someone far more experienced on audioasylum (long ago) said, for diffusion of a cylinder, it really hauls at twice time diameter (so 2khz for my 12"), but works somewhat tapering off to half diameter (550hz). I couldn't tell you about "linking up" to go lower.
Not cut in half, but leave them fully round, floor to higher than my ears, 4" between tubes.
Doesn't cost much, doesn't time any time to make.
I think someone far more experienced on audioasylum (long ago) said, for diffusion of a cylinder, it really hauls at twice time diameter (so 2khz for my 12"), but works somewhat tapering off to half diameter (550hz). I couldn't tell you about "linking up" to go lower.
Attachments
Last edited:
I was looking at how Wilson Audio adds felt to the baffles of there speakers and it got me thinking about my Martin Logan's that have these "wings" on either side of the panels. Would there be any benefit to adding felt/absorbent material to these areas?
I figured this would be the place to ask instead of wasting space creating a new thread.
I figured this would be the place to ask instead of wasting space creating a new thread.
hmmm.............
a parallel surface sticking out on the back, you can measure a something there.
Is it audible sitting 10' away ?
No clue.
But, to me, it can't hurt.
Maybe try rolled up towel.
90% wool or f10 felt, think I paid $30 for a 1/2" thick,
Or even a foam "box" completely around the back side if the speaker is less than 5' from the front wall.
Your ears, your preferences as to what you think is good !!!!!!!!!!
a parallel surface sticking out on the back, you can measure a something there.
Is it audible sitting 10' away ?
No clue.
But, to me, it can't hurt.
Maybe try rolled up towel.
90% wool or f10 felt, think I paid $30 for a 1/2" thick,
Or even a foam "box" completely around the back side if the speaker is less than 5' from the front wall.
Your ears, your preferences as to what you think is good !!!!!!!!!!
Thanks for the info, I might give it a shot at some point for fun 🙂. I like tinkering, even though it typically makes no difference it's still enjoyable.
Last edited by a moderator:
np.
even if your brain believes it is better, who cares about specs !!!!!!!!!!!!
I had some accoustat spectra 11s, there were amazing, so long as I listened below 100db !!!!!!!!!
even if your brain believes it is better, who cares about specs !!!!!!!!!!!!
I had some accoustat spectra 11s, there were amazing, so long as I listened below 100db !!!!!!!!!
Last edited by a moderator: