Hello all,
Recently I have taken an interest in studying diffraction's effects on loudspeaker response, after reviewing Gedlee's whitepapers. Gedlee's research shows that linear distortion artifacts of diffraction are highly audible, more so than non-linear distortion. On the full-range thread I studied the correlation between full range polar response in relation to speaker diameter. This was an effort to use driver beaming to reduce the frequency bandwidth effected by diffraction loss. In summary, you can not practically eliminate all diffraction related distortions by using high directivity fullrange drivers on maximally small baffles. ...but what about creating a speaker that effectively has no baffle step, because the baffle is so large that it loads all frequencies equally?
I've recently taken interest in the design behind Stereophile's A rated Devore Fidelity Orangutan O/96. It's a pretty unusual speaker in that it uses a wide baffle, against the current grain of "tall and skinny" loudspeaker design. Going over reviews on the internet, I came across an interesting tidbit behind the engineering of the O/96. John DeVore states,
"... It eliminates what a lot of people call the step response. Which is where the speaker goes from broadcasting its sound[wave] forward in a hemisphere because the baffle has a certain width, to broadcasting into the room as a sphere because the sound wave becomes double the width of the front baffle, and generally speaking, at that point, whatever frequency that is, the power response in the room will drop...
Hide the fact that there is a baffle by moving the speaker close to a wall boundary so that the wall boundary essentially catches what’s left and everything is coming forward in a hemisphere no matter what the frequency. The other way to do it is to move the frequency so that the step response happens by changing the dimensions of the front baffle, and that’s what we did with the ’96."
Some simple math can show exactly what he is talking about here.
"When a loudspeaker produces a sound, this sound is in the form of a pressure wave trying to expand equally in all directions spherically (like the balloon analogy). The first obstacle that this wave encounters is the baffle face itself. For higher frequencies with shorter wavelengths where the baffle is acoustically large, the baffle causes a doubling of axial pressure into the forward hemisphere (since the pressure can’t expand spherically), much like a perfect reflector. This doubling of acoustic pressure produces a +6dB gain on axis in the forward hemisphere. A baffle with a width of about 9” would correspond to one wavelength at about 1500Hz, this +6dB gain would then be seen at frequencies above 750Hz (that half-wavelength rule)." - Salk Sound
The O/96 utilizes a 18" baffle. 18" would create a baffle step at 376Hz. The wavelength of 376Hz is exactly 3 feet.
Thus, if a speaker with an 18" baffle was 3 feet or closer to the wall behind the speaker, the baffle will appear acoustically infinite - no baffle step.
\
Recently I have taken an interest in studying diffraction's effects on loudspeaker response, after reviewing Gedlee's whitepapers. Gedlee's research shows that linear distortion artifacts of diffraction are highly audible, more so than non-linear distortion. On the full-range thread I studied the correlation between full range polar response in relation to speaker diameter. This was an effort to use driver beaming to reduce the frequency bandwidth effected by diffraction loss. In summary, you can not practically eliminate all diffraction related distortions by using high directivity fullrange drivers on maximally small baffles. ...but what about creating a speaker that effectively has no baffle step, because the baffle is so large that it loads all frequencies equally?
I've recently taken interest in the design behind Stereophile's A rated Devore Fidelity Orangutan O/96. It's a pretty unusual speaker in that it uses a wide baffle, against the current grain of "tall and skinny" loudspeaker design. Going over reviews on the internet, I came across an interesting tidbit behind the engineering of the O/96. John DeVore states,
"... It eliminates what a lot of people call the step response. Which is where the speaker goes from broadcasting its sound[wave] forward in a hemisphere because the baffle has a certain width, to broadcasting into the room as a sphere because the sound wave becomes double the width of the front baffle, and generally speaking, at that point, whatever frequency that is, the power response in the room will drop...
Hide the fact that there is a baffle by moving the speaker close to a wall boundary so that the wall boundary essentially catches what’s left and everything is coming forward in a hemisphere no matter what the frequency. The other way to do it is to move the frequency so that the step response happens by changing the dimensions of the front baffle, and that’s what we did with the ’96."
Some simple math can show exactly what he is talking about here.
"When a loudspeaker produces a sound, this sound is in the form of a pressure wave trying to expand equally in all directions spherically (like the balloon analogy). The first obstacle that this wave encounters is the baffle face itself. For higher frequencies with shorter wavelengths where the baffle is acoustically large, the baffle causes a doubling of axial pressure into the forward hemisphere (since the pressure can’t expand spherically), much like a perfect reflector. This doubling of acoustic pressure produces a +6dB gain on axis in the forward hemisphere. A baffle with a width of about 9” would correspond to one wavelength at about 1500Hz, this +6dB gain would then be seen at frequencies above 750Hz (that half-wavelength rule)." - Salk Sound
The O/96 utilizes a 18" baffle. 18" would create a baffle step at 376Hz. The wavelength of 376Hz is exactly 3 feet.
Thus, if a speaker with an 18" baffle was 3 feet or closer to the wall behind the speaker, the baffle will appear acoustically infinite - no baffle step.
\
Last edited by a moderator:
Somewhat simplistic - have a look at the origins of the IEC standard baffle - 50 something " wide x about 63" high and non-central driver position ... (for a 6dB drop at 100Hz) This is calculated by the various distances between driver centre and baffle edges - I think Wiki has a more detailed explanation or maybe the Archives here.
And then you compound this by curving the baffle using a parabolic function then adding some rear surface damping plus a rear absorbent wedge behind the driver, etc, etc
Then the fun starts as you investigate the properties of absorbtion, diffusion and the room's acoustics for freq response, dynamics response, phase response, intelligibility, etc, etc - the baffles are the easy part!
And then you compound this by curving the baffle using a parabolic function then adding some rear surface damping plus a rear absorbent wedge behind the driver, etc, etc
Then the fun starts as you investigate the properties of absorbtion, diffusion and the room's acoustics for freq response, dynamics response, phase response, intelligibility, etc, etc - the baffles are the easy part!
A copy of one of my previous replies on this forum:
The wide baffle with sloped back vertical edges, has several benefits over a narrow baffle, though not very often mentioned. It's basically a portable infinite baffle design. Low edge diffraction, and much more even power response as the baffle step has been pushed down into the region of room wall gain. Several listeners have mentioned it sounds similar to a dipole. Commercially, Sonus Faber builds the Stradivari and Elipsa wide baffle speakers. The Pioneer TAD Reference One is essentially a wide baffle loudspeaker also.
Wide Baffles
http://www.troelsgravesen.dk/download/IBL.pdf
Poor Man'
http://www.sonusfaber.com/en-us/prod...adivari-homage
? Elliptical speaker Elipsa Red - 3 way floor standing speaker with wood finish
Sonus Faber Cremona Elipsa loudspeaker | Stereophile.com
Sonus Faber Stradivari Homage loudspeaker | Stereophile.com
http://www.theabsolutesound.com/arti...ker-tas-218-1/
http://www.pioneerelectronics.com/PU.../Reference+One
--------------------------------------------------------------------------------
The wide baffle with sloped back vertical edges, has several benefits over a narrow baffle, though not very often mentioned. It's basically a portable infinite baffle design. Low edge diffraction, and much more even power response as the baffle step has been pushed down into the region of room wall gain. Several listeners have mentioned it sounds similar to a dipole. Commercially, Sonus Faber builds the Stradivari and Elipsa wide baffle speakers. The Pioneer TAD Reference One is essentially a wide baffle loudspeaker also.
Wide Baffles
http://www.troelsgravesen.dk/download/IBL.pdf
Poor Man'
http://www.sonusfaber.com/en-us/prod...adivari-homage
? Elliptical speaker Elipsa Red - 3 way floor standing speaker with wood finish
Sonus Faber Cremona Elipsa loudspeaker | Stereophile.com
Sonus Faber Stradivari Homage loudspeaker | Stereophile.com
http://www.theabsolutesound.com/arti...ker-tas-218-1/
http://www.pioneerelectronics.com/PU.../Reference+One
--------------------------------------------------------------------------------
Wide baffles have been around for many years. Early Snells (now AudioNote in UK) and Some Boston Acoustics speakers come to mind.
I think the move to narrow baffles came with the advent of home theater and WAF of the speakers now situated either side of the TV. Of course, imaging improvement efforts also contributed to the move to narrow baffles. Baffle step compensation in the XO seemed to satisfy most narrow cab. builders. Diffraction has also been dealt with to a large degree via cabinet edge rounding, and chamfering.
I think the move to narrow baffles came with the advent of home theater and WAF of the speakers now situated either side of the TV. Of course, imaging improvement efforts also contributed to the move to narrow baffles. Baffle step compensation in the XO seemed to satisfy most narrow cab. builders. Diffraction has also been dealt with to a large degree via cabinet edge rounding, and chamfering.
Exactly my point twinter!
Doc, I wholeheartedly agree that the narrow baffle trend was due to WAF. Although, I honestly think that the O/96 is one of the most beautiful speakers I've ever seen. Maybe just because it's different? Possibly because it has a whole lot of golden ratios in it's design- intentional or not.
I agree that various efforts have reduced linear distortion on narrow baffles pretty well, although I personally think that it is still audible and sub-optimal compared to a wide-baffle "portable infinite baffle" as twinter put it.
Honestly in a way it kind of makes me think of Harbeth's thin wall enclosure approach. Lower linear distortion products in frequency (where they are less audible due to equal loudness contour) and mask them (room gain for diffraction, dampening for thin-wall enclosures).
Doc, I wholeheartedly agree that the narrow baffle trend was due to WAF. Although, I honestly think that the O/96 is one of the most beautiful speakers I've ever seen. Maybe just because it's different? Possibly because it has a whole lot of golden ratios in it's design- intentional or not.
I agree that various efforts have reduced linear distortion on narrow baffles pretty well, although I personally think that it is still audible and sub-optimal compared to a wide-baffle "portable infinite baffle" as twinter put it.
Honestly in a way it kind of makes me think of Harbeth's thin wall enclosure approach. Lower linear distortion products in frequency (where they are less audible due to equal loudness contour) and mask them (room gain for diffraction, dampening for thin-wall enclosures).
Baffle step causes linear distortion if not adequately engineered for (rounding corners, spherical enclosures, wide baffles, etc...).
Either you use a wide baffle and maximally effect all frequencies by moving the enclosure near the back wall to where the baffle is virtually infinite, or use a slim baffle which decreases output at lower frequencies, and use a BSC filter to reduce high frequency output.
Wide baffles will therefore have less crossover circutry, and higher efficiency. It's just a way to deal with an acoustical physics problem with geometric engineering, versus electrical engineering.
Scientific studies of different shaped baffles have shown baffle step can reduce bass output below 200-300 hz up to 6 dB relative to the output above the step.
Compensating for that requires reducing efficiency above the step.
I encourage Boden to Google baffle step. There is a wealth of information available.
Here is but one example...Baffle Diffraction Step
Check out graph #3 and associated baffle step discussion at this link...
http://www.harbeth.co.uk/blog/posts/What-the-'crossover-network'-does...
Last edited:
Sorry WBS, the original post was a couple of years ago. The links worked then. Use the important words in the links of the titles for Google searches if still interested.
Here's one working link to get you started.
http://www.troelsgravesen.dk/Acapella_WB.htm
Last edited:
Baffle step compensation is an adjustment to blend direct and indirect sound from the high and lower frequencies from a narrow baffle. This adjustment results in an acceptable representative sound as if an infinite baffle had been used. A 3 to 4 dB compensation from around 200 to 1,500 or 2,000 Hz is frequently used.
Now the sound power of the upper frequencies to the room, starting above 200 Hz, gradually increasing to say 1,500 Hz, is still being reduced when baffle step compensation is used. So, a narrow baffle speaker with BSC is putting relatively less high frequency energy into the room than a wide baffle speaker without BSC compensation. When in the room listening, this difference is masked by your hearing integrating direct and indirect sound. If you were to listen from a neighboring room, the reflected sound that you hear from a narrow baffle speaker would have less high frequency energy, or sound more base heavy. So the sound presentation of the two different configurations when in the room is probably a little different also.
Now the sound power of the upper frequencies to the room, starting above 200 Hz, gradually increasing to say 1,500 Hz, is still being reduced when baffle step compensation is used. So, a narrow baffle speaker with BSC is putting relatively less high frequency energy into the room than a wide baffle speaker without BSC compensation. When in the room listening, this difference is masked by your hearing integrating direct and indirect sound. If you were to listen from a neighboring room, the reflected sound that you hear from a narrow baffle speaker would have less high frequency energy, or sound more base heavy. So the sound presentation of the two different configurations when in the room is probably a little different also.
Last edited:
Frankly I think it would be much better if you stop posting as if you have some new revelation that has been revealed to you and take the approach that you have an observation and would like input on it. I find that many of your "findings" are simply half-truths or outright falsehoods. This is true IMO of this thread and others that you have written.
Take your assertion that drivers that are very near the baffle width (you call "larger" drivers) have LESS DIFFRACTION than a "smaller" driver on the same baffle. Actually this is not the case. The diffraction process is exactly the same, and so the amount of it is the same, because it is generated by the EDGES of the baffle and these have not changed. What has changed is the number of dx*dy source elements, that is to say the area of surface that is radiating sound waves has increased. This new larger distribution of radiating surface leads to a wider range of pathlengths between sources and edges. This tends to "smear out" in the time domain the reflections produced by the edge, which cause the "baffle step". As a result, the baffle step is "smoother" but this is NOT LESS DIFFRACTION. Only the cumulative effect of the edge diffraction has changed, as viewed from the listening location. You probably have noticed that the limits at high and low frequency of a given baffle are the same no matter what radiator size is used. It's only the transition region that will be influenced by baffle and radiator dimensions, and the location of the radiator on the baffle. Saying something like "the diffraction signature is smoother" would be much less misleading in this case.
How about your assertion that "large baffles" are Genus [sic]. Why is that? This seems to contradict your earlier assertion that a driver having width as close to the baffle width provides the minimum amount of objectionable edge diffraction. I don't really know what you are calling "wide" in this case. 1 meter? If you have a very wide baffle, the diffraction that causes the "baffle step" will simply be pushed lower in frequency and, since your driver is likely now small compared to the baffle width, you will get a strong, sharp diffraction signature from its edges (see below on edge rounding). The only way that larger baffles could reduce the PERCEIVED effect of edge diffraction is if the ear/brain is less sensitive to diffracted sounds at lower frequencies. But I have not seen you make that assertion nor have I ever seen that kind of study done by a reputable source (or any source for that matter).
Then there is your assertion that you should round over your cabinet or baffle edges to "fix" diffraction effects. Your typical rounded edge is only influencing the diffraction at relatively high frequencies because a rounded edge only looks rounded to wavelengths that are small with respect to the radius of the curve. Even if you round over the edge with a 4 inch radius (that is an 8" diameter curve) it will still look "sharp" at low frequencies. So your wide baffle with a rounded edge, where the baffle step diffraction effects have been pushed down to lower frequencies, will derive NO BENEFIT from any practical edge rounding because, compared to the wavelength of sound at the frequencies in which the baffle step is occurring, the radius of the round is small and thus the edge is relatively "sharp". Frankly speaking, I find edge rounding enthusiasts to be poorly informed of what is actually going on and the real benefit of rounding is very often exaggerated. In order to really get a tangible benefit from edge rounding you need a narrow baffle and a large edge rounding radius. Guess what that shape is - it's a sphere. It has been shown that a sphere has the smoothest baffle step transition (by Olson). Your typical 10" wide rectangular box with a 1" rounded edge just isn't doing all that much for diffraction, except at 5kHz and above, and that's not where the baffle step is taking place. By the way, I don't see any rounded edges on that "Orangutan" speaker you mentioned...
So, if you are convinced that edge diffraction is acoustically objectionable there a practical way to minimize or eliminate edge diffraction: use an open baffle. In an OB system when the wavefronts radiated to the front and rear are the same (this is more true at low frequencies than at high frequencies) the acoustic pressure of the front and rear wavefronts pretty much completely cancels out at the baffle edge. No acoustic pressure means no excitation of any diffraction sources along the edge, meaning no "baffle step". Instead you get a different type of acoustic cancellation that leads to the well known OB frequency response signature. This still has peaks and dips and a rolloff starting at a frequency related to the baffle dimensions, but it's not due to diffraction like with a closed box.
Finally, I do not know what Mr. John DeVore is smoking, but moving a speaker close to the wall is just moving the signature of the reflection from the back wall to a higher frequency. The picture that is pained by the quoted passage is overly simplistic and misleading. Only in the limit of a wide shallow speaker placed essentially on the wall will this help. In that case the frequency at which the wavefront reflected off the rear wall causes interference with the direct sound will be relatively high, while most of the wavefront at "high" frequency that are produced by the drivers are projected forward (not backwards towards the wall) because of the wide baffle. This is essentially an "in wall" design, and these are known to have worse imaging compared to a loudspeaker placed out in the room away from room boundaries (if you care about that kind of thing). The DeVore speaker is NOT wide to begin with, so I really can't see how moving it near the wall is going to help unless the frequency response of the loudspeaker is flawed (not flat on or off axis) and this is a marketing band-aid that offsets that flaw. I did not see it pictured near the wall in the photos that I saw. Honestly, I read that Sterephile review just now and I had the same visceral reaction that I always do - I wanted to puke. Please throw that Stereophile crap in the trash and don't use it as a supporting argument for loudspeaker design! It's just part of a slick marketing cabal... just like the Devore web site.
Take your assertion that drivers that are very near the baffle width (you call "larger" drivers) have LESS DIFFRACTION than a "smaller" driver on the same baffle. Actually this is not the case. The diffraction process is exactly the same, and so the amount of it is the same, because it is generated by the EDGES of the baffle and these have not changed. What has changed is the number of dx*dy source elements, that is to say the area of surface that is radiating sound waves has increased. This new larger distribution of radiating surface leads to a wider range of pathlengths between sources and edges. This tends to "smear out" in the time domain the reflections produced by the edge, which cause the "baffle step". As a result, the baffle step is "smoother" but this is NOT LESS DIFFRACTION. Only the cumulative effect of the edge diffraction has changed, as viewed from the listening location. You probably have noticed that the limits at high and low frequency of a given baffle are the same no matter what radiator size is used. It's only the transition region that will be influenced by baffle and radiator dimensions, and the location of the radiator on the baffle. Saying something like "the diffraction signature is smoother" would be much less misleading in this case.
How about your assertion that "large baffles" are Genus [sic]. Why is that? This seems to contradict your earlier assertion that a driver having width as close to the baffle width provides the minimum amount of objectionable edge diffraction. I don't really know what you are calling "wide" in this case. 1 meter? If you have a very wide baffle, the diffraction that causes the "baffle step" will simply be pushed lower in frequency and, since your driver is likely now small compared to the baffle width, you will get a strong, sharp diffraction signature from its edges (see below on edge rounding). The only way that larger baffles could reduce the PERCEIVED effect of edge diffraction is if the ear/brain is less sensitive to diffracted sounds at lower frequencies. But I have not seen you make that assertion nor have I ever seen that kind of study done by a reputable source (or any source for that matter).
Then there is your assertion that you should round over your cabinet or baffle edges to "fix" diffraction effects. Your typical rounded edge is only influencing the diffraction at relatively high frequencies because a rounded edge only looks rounded to wavelengths that are small with respect to the radius of the curve. Even if you round over the edge with a 4 inch radius (that is an 8" diameter curve) it will still look "sharp" at low frequencies. So your wide baffle with a rounded edge, where the baffle step diffraction effects have been pushed down to lower frequencies, will derive NO BENEFIT from any practical edge rounding because, compared to the wavelength of sound at the frequencies in which the baffle step is occurring, the radius of the round is small and thus the edge is relatively "sharp". Frankly speaking, I find edge rounding enthusiasts to be poorly informed of what is actually going on and the real benefit of rounding is very often exaggerated. In order to really get a tangible benefit from edge rounding you need a narrow baffle and a large edge rounding radius. Guess what that shape is - it's a sphere. It has been shown that a sphere has the smoothest baffle step transition (by Olson). Your typical 10" wide rectangular box with a 1" rounded edge just isn't doing all that much for diffraction, except at 5kHz and above, and that's not where the baffle step is taking place. By the way, I don't see any rounded edges on that "Orangutan" speaker you mentioned...
So, if you are convinced that edge diffraction is acoustically objectionable there a practical way to minimize or eliminate edge diffraction: use an open baffle. In an OB system when the wavefronts radiated to the front and rear are the same (this is more true at low frequencies than at high frequencies) the acoustic pressure of the front and rear wavefronts pretty much completely cancels out at the baffle edge. No acoustic pressure means no excitation of any diffraction sources along the edge, meaning no "baffle step". Instead you get a different type of acoustic cancellation that leads to the well known OB frequency response signature. This still has peaks and dips and a rolloff starting at a frequency related to the baffle dimensions, but it's not due to diffraction like with a closed box.
Finally, I do not know what Mr. John DeVore is smoking, but moving a speaker close to the wall is just moving the signature of the reflection from the back wall to a higher frequency. The picture that is pained by the quoted passage is overly simplistic and misleading. Only in the limit of a wide shallow speaker placed essentially on the wall will this help. In that case the frequency at which the wavefront reflected off the rear wall causes interference with the direct sound will be relatively high, while most of the wavefront at "high" frequency that are produced by the drivers are projected forward (not backwards towards the wall) because of the wide baffle. This is essentially an "in wall" design, and these are known to have worse imaging compared to a loudspeaker placed out in the room away from room boundaries (if you care about that kind of thing). The DeVore speaker is NOT wide to begin with, so I really can't see how moving it near the wall is going to help unless the frequency response of the loudspeaker is flawed (not flat on or off axis) and this is a marketing band-aid that offsets that flaw. I did not see it pictured near the wall in the photos that I saw. Honestly, I read that Sterephile review just now and I had the same visceral reaction that I always do - I wanted to puke. Please throw that Stereophile crap in the trash and don't use it as a supporting argument for loudspeaker design! It's just part of a slick marketing cabal... just like the Devore web site.
The more important issue of wide baffles, 24" and greater, is the baffle step characteristics and power response.
However, here is a link with tests showing the impact of a 3/4" round over on a small speaker cabinet.
MurphyBlaster Productions
However, here is a link with tests showing the impact of a 3/4" round over on a small speaker cabinet.
MurphyBlaster Productions
Wow someone's jimmies are rustled. Ad hominem to the max! So first of all lets go over a few things.
I have been studying acoustics pertaining to music reproduction on my own time, and here and there have ideas that I don't see talked about much. I like to share my ideas when they seem a little out of the mainstream. Nothing about this is illegal.
Speaker directivity does indeed effect linear distortion caused by baffle diffraction.
"Finally, a feature that helps to control diffraction is controlled directivity. Our example above treats the loudspeaker driver as a point source. In reality this is not correct. All drivers have an effective radiating width or diameter; the larger this diameter, the greater the directivity of the driver. As a result, less acoustic energy at higher frequencies is able to illuminate the edge of the cabinet that is usually 90 degrees off-axis. (Just look at the 60 and 90 degree off-axis frequency response curves for many drivers for a good example of what I mean). If less energy illuminates the edge, then the strength of the edge source is reduced and so is the diffraction."- Salk Sound
I clearly stated in the first post that the previous thread was an investigation into maximally minimizing the bandwidth effected by baffle diffraction related distortions; while this thread is looking to effectively maximally increase the bandwidth effected by baffle diffraction -or to make it clearer- lower the frequency which radiation would become spherical, and negating that by placing the speaker enclosure near a boundary which correlates to the wavelength of the shortest frequency beginning to wrap behind the speaker. Both are ways, although opposite approaches, to try and produce a frequency response curve that is minimally impacted by linear-distortion effects of baffle diffraction. Yes, obviously if we play semantics a larger baffle does not have less diffraction- it increases the bandwidth effected by diffraction-. The point is the resulting frequency response linearity.
Nowhere did I claim that a round over would effect low frequencies, nor do I see any proof claiming that it does or does not. I also disagree that modest roundovers do not have an effect. Here is your proof.
I clearly gave the example of the O/96's having an 18" baffle correlating to a wavelength of 3 feet- and moving the speaker 3 feet or closer to the rear wall will virtually lengthen the physical baffle. It's common sense that while the 376hz wavelength will wrap behind the speaker, the wavelength will be reflected off the rear boundary before completing a full cycle; effectively becoming part of the speaker baffle. As somebody who has a formal education on ray-tracing, I highly disagree that this is not possible. Additionally, nowhere did I state that this design negates the effects of SBIR. Additionally, if you want to talk about constructive and deconstruction interference, you have much more to worry about at low frequencies- room modes.
I have been studying acoustics pertaining to music reproduction on my own time, and here and there have ideas that I don't see talked about much. I like to share my ideas when they seem a little out of the mainstream. Nothing about this is illegal.
Speaker directivity does indeed effect linear distortion caused by baffle diffraction.
"Finally, a feature that helps to control diffraction is controlled directivity. Our example above treats the loudspeaker driver as a point source. In reality this is not correct. All drivers have an effective radiating width or diameter; the larger this diameter, the greater the directivity of the driver. As a result, less acoustic energy at higher frequencies is able to illuminate the edge of the cabinet that is usually 90 degrees off-axis. (Just look at the 60 and 90 degree off-axis frequency response curves for many drivers for a good example of what I mean). If less energy illuminates the edge, then the strength of the edge source is reduced and so is the diffraction."- Salk Sound
I clearly stated in the first post that the previous thread was an investigation into maximally minimizing the bandwidth effected by baffle diffraction related distortions; while this thread is looking to effectively maximally increase the bandwidth effected by baffle diffraction -or to make it clearer- lower the frequency which radiation would become spherical, and negating that by placing the speaker enclosure near a boundary which correlates to the wavelength of the shortest frequency beginning to wrap behind the speaker. Both are ways, although opposite approaches, to try and produce a frequency response curve that is minimally impacted by linear-distortion effects of baffle diffraction. Yes, obviously if we play semantics a larger baffle does not have less diffraction- it increases the bandwidth effected by diffraction-. The point is the resulting frequency response linearity.
Nowhere did I claim that a round over would effect low frequencies, nor do I see any proof claiming that it does or does not. I also disagree that modest roundovers do not have an effect. Here is your proof.
I clearly gave the example of the O/96's having an 18" baffle correlating to a wavelength of 3 feet- and moving the speaker 3 feet or closer to the rear wall will virtually lengthen the physical baffle. It's common sense that while the 376hz wavelength will wrap behind the speaker, the wavelength will be reflected off the rear boundary before completing a full cycle; effectively becoming part of the speaker baffle. As somebody who has a formal education on ray-tracing, I highly disagree that this is not possible. Additionally, nowhere did I state that this design negates the effects of SBIR. Additionally, if you want to talk about constructive and deconstruction interference, you have much more to worry about at low frequencies- room modes.
Last edited:
Here is the title of a relatively recent paper published by the Audio Engineering Society. Visit the AES.org web site to download.
"Accurate Calculation of Radiation and Diffraction
from Loudspeaker Enclosures at Low Frequency"
JEFF CANDY, Author
The AES library is a great source of technical knowledge. In particular, the AES journal offers peer reviewed papers, not one-off anecdotal experiments. A search on cabinet diffraction turned up numerous hits.
"Accurate Calculation of Radiation and Diffraction
from Loudspeaker Enclosures at Low Frequency"
JEFF CANDY, Author
The AES library is a great source of technical knowledge. In particular, the AES journal offers peer reviewed papers, not one-off anecdotal experiments. A search on cabinet diffraction turned up numerous hits.
Last edited:
Baffle Step is a no-issue if loudspeaker design guidelines are followed.
Measure raw driver in box, properly merge nearfield and 1 meter raw woofer-inbox measurements, define desired target acoustic curve and design crossover on that basis.Voila. On does not apply a BSC filter, it is merely an extra pole designed into the lowpass section. You can easily recognize that by the relatively large value of the series inductor, usually somewhere between 1.5 and 3.5 mH, dependant on the chosen crossover frequency.
And, yes you lose 4/5 dB with a narrow baffle. Nothing to get excited about.
Eelco
Measure raw driver in box, properly merge nearfield and 1 meter raw woofer-inbox measurements, define desired target acoustic curve and design crossover on that basis.Voila. On does not apply a BSC filter, it is merely an extra pole designed into the lowpass section. You can easily recognize that by the relatively large value of the series inductor, usually somewhere between 1.5 and 3.5 mH, dependant on the chosen crossover frequency.
And, yes you lose 4/5 dB with a narrow baffle. Nothing to get excited about.
Eelco
Boden, totally agree, if I may add:
"Baffle Step Compensation implementation is a no-issue if loudspeaker design guidelines are followed."
I typically increase the inductance of the inductor in the woofer crossover for BCS in two-way speakers.
I'm a mechanical engineer, and my controls and filter theory lectures are way in the past. Is BCS implementation actually another pole, or different magnitude of the pole? Implemented BCS (not theoretical) is actually about 1dB/octave for approximately 3 octaves. As you stated, an additional crossover filter component is not typically added. Is my question stated correctly or making sense?
Last edited:
I think most builders of small, narrow 2-ways use BSC compensation via bumped up woofer inductors as you noted