The acoustics of dipoles is well known and characterized; I didn't invent them. Same with wavelength relationships and constructive/destructive interference. If you want to ignore basic physics, feel free. All I can do is explain it to those who want to understand.
Bass quality and bass extension are different matters and all the goal-post moving gets me dizzy. To prefer a good experiment is correct- but one must have a good experiment!
Bass quality and bass extension are different matters and all the goal-post moving gets me dizzy. To prefer a good experiment is correct- but one must have a good experiment!
Another thing that has been overlooked with doing a nearfield measurement, with the mic 3cm from the front of the diaphragm, is that the SPL encountered by the mic, from being energised by that diaphragm, is substantially higher then that of any reflected sound, be it from the room itself or from the rear wave being diffracted around the panel.
With the long gate time necessary to allow the accurate capture and analysis of the long wavelengths, secondary sources of sound are inevitably going to come into contact with the mic, but because they are significantly lower down in magnitude vs the original sound, the effects are effectively pushed under the carpet. This is the way I've understood how nearfield measurements work and the reason why you don't measure any cabinet issues in the near field is because measured up next to the cone, the sound waves have yet to encounter the baffle so its effects are not measured.
Regardless of the room size or type you're going to get 6dB/octave OB cancellation happening because that's surely related to the long, out-of-phase wavelengths, from the rear of the cone, diffracting around the relatively narrow baffle and interfering destructively with the waves from the front of the cone?
With the long gate time necessary to allow the accurate capture and analysis of the long wavelengths, secondary sources of sound are inevitably going to come into contact with the mic, but because they are significantly lower down in magnitude vs the original sound, the effects are effectively pushed under the carpet. This is the way I've understood how nearfield measurements work and the reason why you don't measure any cabinet issues in the near field is because measured up next to the cone, the sound waves have yet to encounter the baffle so its effects are not measured.
Regardless of the room size or type you're going to get 6dB/octave OB cancellation happening because that's surely related to the long, out-of-phase wavelengths, from the rear of the cone, diffracting around the relatively narrow baffle and interfering destructively with the waves from the front of the cone?
Hi,
No wonder that this Lacroix guy's planar gives lots of output even working as dipole, even working at low freqs. Just consider the sheer size of the thing and the claimed possible excursions. One m2 of panel surface equals 12 pcs of 15"ers! Per channel! The same output below say 30Hz would be gained with just two or three 15"ers in a equalized CB not only because of dipole cancellation but also due to missing pressure chamber effect of the dipole.
After my experience a dipole bass is well suited for freqs above the lowest room mode, but there's no sonic advantage anymore below that frequency (you can't even measure output here under nearfield conditions), but just serious air volume displacement demand. Dimensions of the dipole quickly become impractical.
jauu
Calvin
No wonder that this Lacroix guy's planar gives lots of output even working as dipole, even working at low freqs. Just consider the sheer size of the thing and the claimed possible excursions. One m2 of panel surface equals 12 pcs of 15"ers! Per channel! The same output below say 30Hz would be gained with just two or three 15"ers in a equalized CB not only because of dipole cancellation but also due to missing pressure chamber effect of the dipole.
After my experience a dipole bass is well suited for freqs above the lowest room mode, but there's no sonic advantage anymore below that frequency (you can't even measure output here under nearfield conditions), but just serious air volume displacement demand. Dimensions of the dipole quickly become impractical.
jauu
Calvin
Since I re purposed the four 10" XLS drivers, out of their H baffles and into 4 small subs and configured them as 'multiple subs' I've never looked back. I get better bass with the multiple subs AND 105dB @ 20Hz in room. Movies are a lot of fun like that let me tell you, something the OBs could never do.
1m squared is a lot of panel area, but do bare in mind that the maximum displacement is probably only a couple of mm. A 15" driver will have around 850cm squared, which as you pointed out is ~12 times less that of the 1m sq panel, but it can have up to around 10 times the total displacement. Two good 15" will give you greater volume displacement I bet.
1m squared is a lot of panel area, but do bare in mind that the maximum displacement is probably only a couple of mm. A 15" driver will have around 850cm squared, which as you pointed out is ~12 times less that of the 1m sq panel, but it can have up to around 10 times the total displacement. Two good 15" will give you greater volume displacement I bet.
Big difference between waving your arms in the air and invoking the eternal principles of acoustics and another thing to apply these principles to a real-world shape in a real-world room, as designers are expected to do.
Just what would be an example where these well-known principles were meaningfully applied in any way beyond rules of thumb or billiard-ball geometry (excepting simple round baffles, eh)?
I recognize it may seem heretical to talk this way about one of the oldest chapters in the oldest acoustics books. But it is so simple, everybody takes OB "design" for granted - but shouldn't.
I'm not denying these principles, just suggesting, again, that you need Monte-Carlo methods or real-world measurements, not rules of thumb to characterize what a big OB will sound like. But because the shortcomings of OBs are so obvious from theory, I doubt many people would find it interesting to research them.
About experimentation, as opposed to Golden-Eared declarations of faith, couldn't agree with you more. Anybody who boasts about trusting their judgment about their own speakers is woefully under-read in psychology.
(BTW, nobody would expect too much bass from an H-shaped OB, once you take a tape measure to one.)
Ben
Last edited:
Concerning sound pressure, i just remind my aim is to get a good infrabass effect in a classic 30 sqmeters living-room. It's not to shake a whole hundred sqmeters home movie saloon in the basement for doomsday films.
Anyway, with the power admittance of a LS150 planar (400 watts), it's not difficult to slightly equalize a 12dB/oct roll-off, even completely if absolutely needed.
Anyway, with the power admittance of a LS150 planar (400 watts), it's not difficult to slightly equalize a 12dB/oct roll-off, even completely if absolutely needed.
What SY says though about open baffle theory is just simple well understood physics. The same principles are necessary in designing concert halls, recording studios and in the room treatments necessary to make them work as 'intended'.
If you want to take a crash course in a lot of open baffle theory head on over the linkwitzlab.com and do some reading.
Linkwitz has also produced a spread sheet that will calculate the maximum SPL that you can achieve by entering some very simple parameters...
1) The surface area of the radiator.
2) The width of the baffle it is on and
3) Its linear excursion.
I suggest you have a play with it as a way of understanding what is and what is not possible.
Google Linkwitz displacement to get a hold of it.
WIth the spreadsheet you don't actually need to understand the physics behind it, it just shows you how it is.
If you want to take a crash course in a lot of open baffle theory head on over the linkwitzlab.com and do some reading.
Linkwitz has also produced a spread sheet that will calculate the maximum SPL that you can achieve by entering some very simple parameters...
1) The surface area of the radiator.
2) The width of the baffle it is on and
3) Its linear excursion.
I suggest you have a play with it as a way of understanding what is and what is not possible.
Google Linkwitz displacement to get a hold of it.
WIth the spreadsheet you don't actually need to understand the physics behind it, it just shows you how it is.
I've done one or two dipoles over the years.
I did point that out in an earlier post, but it doesn't hurt to say it again.
"Another thing that has been overlooked with doing a nearfield measurement, with the mic 3cm from the front of the diaphragm, is that the SPL encountered by the mic, from being energised by that diaphragm, is substantially higher then that of any reflected sound, be it from the room itself or from the rear wave being diffracted around the panel."
I understood that it was precisely to be not influenced by reflexion and cancellation. What matters if you only want to see the regularity of the response? The classic measure (1m ) has for main quality to be a standard for comparisons.
The only good one for a simple user is the measure at the listening point, to find the less bad position of speakers, and then get an automatic equalization by your numeric processor.
I understood that it was precisely to be not influenced by reflexion and cancellation. What matters if you only want to see the regularity of the response? The classic measure (1m ) has for main quality to be a standard for comparisons.
The only good one for a simple user is the measure at the listening point, to find the less bad position of speakers, and then get an automatic equalization by your numeric processor.
Hi,
maybe some overread it, so I just repeat it.
Below the lowest mode the dipole loses on output since it is not able to pressurize the room.
Besides the -6dB/oct drop due to phase cancellation that already asks for +14dB of equalization between 20Hz and 100Hz, the natural -12dB/oct rolloff below the Fs needs to be added, giving -18dB/oct.
As if that weren´t enough a steep drop adds below the lowest mode where the room becomes a pressure chamber.
So efficiency in the lowest octave or even more at infrasonic frequencies is not only terribly bad, it also doesn´t bear any sonic advantage whatsoever. If a dipole bass sounds superior than it is in the upper bass range, above the lowest room mode (except my ears are always situated in the nearfield).
as Sy said, the physics and room interaction of dipoles is no mystery. It has been the topic of several papers. A good read for background is John Kreskovkys site at Tech
Another source (only part in english, mostly German) are the papers of Ferekidis/Kempe at Downloads
jauu
Calvin
maybe some overread it, so I just repeat it
.Below the lowest mode the dipole loses on output since it is not able to pressurize the room.
Besides the -6dB/oct drop due to phase cancellation that already asks for +14dB of equalization between 20Hz and 100Hz, the natural -12dB/oct rolloff below the Fs needs to be added, giving -18dB/oct.
As if that weren´t enough a steep drop adds below the lowest mode where the room becomes a pressure chamber.
So efficiency in the lowest octave or even more at infrasonic frequencies is not only terribly bad, it also doesn´t bear any sonic advantage whatsoever. If a dipole bass sounds superior than it is in the upper bass range, above the lowest room mode (except my ears are always situated in the nearfield).
as Sy said, the physics and room interaction of dipoles is no mystery. It has been the topic of several papers. A good read for background is John Kreskovkys site at Tech
Another source (only part in english, mostly German) are the papers of Ferekidis/Kempe at Downloads
jauu
Calvin
So... I linked to some authors and looked for acoustic measurements relating the immutable laws of acoustics to OB design and didn't see any in my cursory look. Maybe I am just being lazy and the Linkwitz crowd have many threads, but could somebody link me to a chart comparing "theory" and measurement, please?
But I did try to understand a very mathematical treatment that QED had all the truth reduced to some geometrical simplicities. It seems that when you have a complex baffle shape, you merely do a large number of calculations. Kind of like I've been saying about a Monte-Carlo simulation. Of course, that is still a very simplified picture, before you put the baffle in a room with, like most rooms, walls.
My basic point is this. Easy to talk about half-wave distances, Euclid, and cancellation. No argument from me about physics. You can even be confident that Linkwitz' bass boost will sound better than no bass boost. But as soon as you think real-world, with waves bouncing hither and yon, the simple mental picture goes out the window and an OB is a whole lot more bass than intuition would suggest.
Ben
But I did try to understand a very mathematical treatment that QED had all the truth reduced to some geometrical simplicities. It seems that when you have a complex baffle shape, you merely do a large number of calculations. Kind of like I've been saying about a Monte-Carlo simulation. Of course, that is still a very simplified picture, before you put the baffle in a room with, like most rooms, walls.
My basic point is this. Easy to talk about half-wave distances, Euclid, and cancellation. No argument from me about physics. You can even be confident that Linkwitz' bass boost will sound better than no bass boost. But as soon as you think real-world, with waves bouncing hither and yon, the simple mental picture goes out the window and an OB is a whole lot more bass than intuition would suggest.
Ben
Last edited:
I don't even understand why people are trying to counter the laws of wave physics in this thread as that makes no sense. The laws that govern how sound waves interact and behave with loudspeakers are the same laws that govern how electromagnetic waves behave in all other areas of physics. If they were wrong then our large radio telescopes would fail to operate and the complex arrangement that combines the light from the four individual telescopes in the VLT would also fail.
It might be so that there are complexities that one needs to add in to make a simulation more realistic, rather then taking a simplistic view of it - something that only modern simulation software can do. But the fundamental science describing how waves interact with objects and boundaries is something that is very well understood, it is taught to you first at GCSE level here in the UK and then retaught in a more complex fashion in first year A level physics.
It might be so that there are complexities that one needs to add in to make a simulation more realistic, rather then taking a simplistic view of it - something that only modern simulation software can do. But the fundamental science describing how waves interact with objects and boundaries is something that is very well understood, it is taught to you first at GCSE level here in the UK and then retaught in a more complex fashion in first year A level physics.
Where have I gone wrong to give 5th Element the impression that I am incompetent with high school physics? Is my writing too simple or is his reading too simple?
I think I can boil down my POV to this. No "theory" does a good job of predicting what a mic picks up at my ears in a room. But OB "theory" - perhaps because it looks so very simple but in practice isn't* - does an esp. poor job.
And marking me as something of a partisan, as much as I like true horns and sealed boxes, the sound resulting from a giant OB** is surprisingly nice.
Ben
*compare the relative simplicity of the acoustics of sound coming from a driver in a sealed box to the complexity of a funny-shaped dipole at a two angles to two rear walls
**I'm also quite a partisan for motional feedback and I believe an OB meets the assumptions for that app (uniquely along with sealed boxes)
I think I can boil down my POV to this. No "theory" does a good job of predicting what a mic picks up at my ears in a room. But OB "theory" - perhaps because it looks so very simple but in practice isn't* - does an esp. poor job.
And marking me as something of a partisan, as much as I like true horns and sealed boxes, the sound resulting from a giant OB** is surprisingly nice.
Ben
*compare the relative simplicity of the acoustics of sound coming from a driver in a sealed box to the complexity of a funny-shaped dipole at a two angles to two rear walls
**I'm also quite a partisan for motional feedback and I believe an OB meets the assumptions for that app (uniquely along with sealed boxes)
Last edited:
Actually, low frequency in a room is very well accounted for by standard simple physics. FRD Consortium used to have some excellent software for predicting response and, experimentally, it gave very good predictions. Don't know if it's still around.
Hi,
@Jys: no reason to become offensive. I´m in fact a true fan of dipole basses. But I also see and respect the disadvantages and limitations of these systems.
The flesh of the cite You quoted from me can also be read at John Kreskovskys site with all the academic terms You asked for. The second links papers also supply for theory and measurements. If You think there hasn´t been an explanation till now, I actually don´t know what else could help ...maybe an inroom farfield measurement of dipoles on Your own and seeing what You get as output from the dipole below the lowest room mode might convince.
Since You claim for considerable output of dipoles in the lowest octave and below, could You please be so kind and answer the following Q: If the room behaves as a pressure chamber due to the associated wavelengths beeing larger than the room dimensions. How do You explain a sound pressure buildup if there´s no change in room volume?
jauu
Calvin
@Jys: no reason to become offensive. I´m in fact a true fan of dipole basses. But I also see and respect the disadvantages and limitations of these systems.
The flesh of the cite You quoted from me can also be read at John Kreskovskys site with all the academic terms You asked for. The second links papers also supply for theory and measurements. If You think there hasn´t been an explanation till now, I actually don´t know what else could help ...maybe an inroom farfield measurement of dipoles on Your own and seeing what You get as output from the dipole below the lowest room mode might convince.
Since You claim for considerable output of dipoles in the lowest octave and below, could You please be so kind and answer the following Q: If the room behaves as a pressure chamber due to the associated wavelengths beeing larger than the room dimensions. How do You explain a sound pressure buildup if there´s no change in room volume?
jauu
Calvin
Indeed and there's also a piece of software called 'the edge' This is a diffraction simulator that predicts how a baffle + driver arrangement will behave as either a monopole or a dipole and again, that also gives extremely accurate results.
I wasn't trying to imply that you didn't know, or weren't taught the basics of wave physics, more that it is simple enough that it can be taught at GSCE and then expanded on as a first year A level topic. There's no magic involved here and to try and say that the theory isn't good enough to explain how a dipole will work in a room is wrong. Yes, to design a piece of software that takes every conceivable variable into consideration would be a rather large undertaking, but that isn't to say that it can't be done.
At a very basic level the dipole will roll off at 6dB/octave below a certain frequency, until you hit the drivers resonance. To understand and compensate for this much doesn't even require that you acknowledge that the room exists, because regardless of its environment, that much will always happen - that is of course if you are in the farfield, which you almost always are.
Then if you wish, you can start to add in the complex arrangement of wave interactions that occur within the room and how it will modify the response above, but this is seldom taken into consideration. This is usually because there isn't a whole lot you can do about it (the room is fixed, as are most likely the range of positions that you can place your loudspeakers in), nor is it in any way going to necessitate that you remove or significantly alter your 6dB/octave, OB compensation.
The fact remains though, if you take a nearfield meaurement of a dipole loudspeaker you will not see the 6dB/octave roll off that they will show in the farfield.
Here is an image of LW doing just that and are we going to say that this configuration has extension down to ~20Hz? Of course not, because it does not, unless of course you're going to listen to it with you ear pressed up to the opening of the baffle.
Once you start measuring in the farfield you end up seeing the expected roll off as LW does here.
There's no magic to this, except that to get accurate data down this low needs a complicated outdoor arrangement where you end up lifting the loudspeaker far off the floor to remove any unwanted reflections. Note that you still see the roll off when there is no room interaction, reflections are not needed, nor are they the cause of the roll off, but when added in, just like if you start letting reflections in with a normal gated farfield measurement, they end up marring the accuracy of your results.
In other words here, if the planar loudspeaker that this thread was created around, was measured using a nearfield measurement, then this cannot in any way be used as a reference for what will happen when you actually listen to it in the farfield. Now nearfield measurements are the norm when wanting to see what the overall extension of a monopole loudspeaker will be like, but this does not translate to a dipole.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.