I completely agree. The whole concept of directionality fails when there are dominate modes because then the sound waves can only travel along very precise prescribed directions. Arbitrary directions are not possible so the whole concept of "source directivity" fails.
The classic audiophile catch-all response. "I don;t care about facts, I know what I believe!"
And this is a single monopole (I hope nobody believes that is the way to go anymore) without any EQ. The Dipole is obviously EQ'd. Hardly a conclusive test result.
The DBA is essentially absorbing all the bass energy after the wave has travelled the length of the room once. This is similar to extreme room treatment for bass absorption....and no doubt is equally expensive and/or cumbersome.
However the idea is the same: reduce the bass energy in the indirect sound field...which is what the dipole is also doing to a certain degree, without the added expense and effort. Delayed out of phase cancellation of the low frequency sound waves.... is the central theme of both the DBA and damped U-Frames....with a slightly different implementation.
I have significant experince with both and my opinion is that both sound comparable when both are done correctly. Why is that so hard to accept?
I do find monopoles much easier to impliment however.
Huh? A DBA doesn't work in each and every room but have you calculated what it costs to get equal absorption from a passive solution?
As a physicist this bothers me since the physics says that this should not happen. Once the source is turned off the decay rate is room dependent ONLY, and if the room is unchanged then the decay rate cannot change.
The only way that you could get two different decay rates is if there are two different systems decaying. If the one is EQ'd and the other isn't then you caould be looking at the decay of the EQ.
At any rate those results are not comparable and no conclusions can be drawn from them. But of course these are just facts so they can easily be ignored but any competent audiophile.
It would be helpful to know the the approximate dimensions of what you call a "small room". What if a small room had multiple openings/doors/windows and the walls are not brick but the usual drywall. Will such a small room acoustically behave like a large one ?
What has been the reason(s) of success of cardioids in concert settings...is it just to prevent feedback to the mics on the stage behind the speaker...or is it actually better bass reproduction than monopoles ??
My guess as stated in a previous post, is that even in an open arena, a dipole/cardioid will sound cleaner with better transients due to minimal stored energy issues.
I am fortunate to have just such a room- my floorplan has largely connected (no doors) kitchen, living room, dining room, and main hallway.
Distributed bass is still very helpful but I've definitely not had the challenges with bass I've had in other rooms.
There are a lot of mistakes here:
1)"However if less sound waves reach the room boundaries"- except that in the steady state, which is true of all small rooms at LFs, this is just not true. Both sources "see" all boundaries.
2)"The only other way to reduce room colourations is to increase wall absorption."- If you mean this exclusively for LFs it is partially true, but above the modal region it is not. There are many ways to do this and IMO large HF wall absorption is exactly the wrong way to do it.
3)"the only way to reduce the room colourations in the bass region is to get narrow directivity in the bass region"- LF directivity in a small room is a falicy.
4)"as the backwave from a sealed monopole can never be completely absorbed (unless it is a rare long stuffed sealed TL) the unaccounted and un-simulatable stored energy from a monopole eventually comes out" This is not true on several accounts. The back wave IS accounted for, it is the load impedance that is presented to the driver. It does not "come back out", it simply stiffens the cone. We are talking here about wavelengths that are an order of magnitude greater than the source dimensions. These are not short wavelengths that "travel" and "reflect" as sound rays. This is the modal region and "sound rays" are not applicable.
It is not just the cost but the cumbersome nature of setting up a DBA. Forget room treatments if it is more expensive than DBA...here we are talking about the cost/effort to make dipoles being significantly lower than DBA and providing similar bass reproduction.
I have not heard a DBA...you might have. What i do know from listening, is that dipole sounds more natural and closer to a real musical instrument like a kickdrum, compared to a sealed monopole, forget about ported.
As I mentioned before, I paid attention on this phenomena. But explanation is, better blending (in terms of perception) of reverberation of the room with reverberation reproduced by stereo speakers, and better stereo image, that works on frequencies above what you are talking about. Try to compare dipole and monopole for bass only, both properly equalized, and you will never hear such an effect.
No not really. The dominate effect is still "small". Any room in any home is small (That I have been in that is - my circle are just "normal people", you know the 99%.) A large classroom (100+ seats) is just leaving what I would call the "small" domain where directionality of LFs even begins to be a factor.
I don't buy the "stored energy" argument. There is no such thing in the context that you are using it. Cardiods work in large venues because they direct the energy where it needs to be. The complete lack of a modal region in these spaces allows this to happen. In a small room no source can be "dierctional" at LFs.
The difference between a monopole and a dipole (at modal frequencies) can be stated quite simply: A dipole excites the modes velocity while the monopole excites the modes pressure. And therein lies the only difference between the two. They both excite ALL modes, but in different ways. A monpole can be on a pressure node just as a dipole can be on a velocity node, and visa-versa. One does not decay any faster than the other or anything like that (assuming comparable loudspeaker Q's, but these are insignificant when compared to room mode Q's). Get the physics right and you will get the understanding right.
Have you heard a room with multiple EQ'd subs?
A kickdrum is a fairly HF instrument. Are you sure its the bass difference that you are hearing. Even I see a justification for a dipole in the 125-500 Hz region.
In that region directionality is just begining to be effective and pistons are still not directional enough while the dipole is. Its very hard to get a directional source < 500 Hz. except with a dipole or a dipole derivative like a cardiod.
But < 125 Hz I just do not buy the dipole argument.
Last edited:
Thanks Dr. Geddes for the explanations
[/QUOTE]
There was a discussion on this forum debating if reaching "steady state" is at all necessary to appreciate bass transients. Dont remember if that reached a conclusion.
You may believe that directional bass from a dipole is a fallacy. But i can certainly appreciate a reduction in SPL when moving from the front to the side of a Hframe woofer. Dont know how else to explain this.
What you say could be true. In which case a long line closed TL with quarter wavelength absorption should sound exactly the same as the smallest sealed box equalised by Linkwitz transform, to the same LF extension. I will need experiments to prove this to myself and then ridicule everybody that makes a large cabinet !
[/QUOTE]
Last edited:
Connecting several bass drivers in reverse and applying a delay is cumbersome?
Dipoles with 16 21" drivers are more cost effective and less of an effort?
Steady state is perhaps not the right way to say it. The fact is that a sound wave has gone arround the room several times before even a single period has elapsed. This means that as far as any boundary effects are concerned, all of the bondaries contribute and the result is a sort of average of them. This is precisely "steady-state" but its hardly a transient natire either.
This is where you have to be careful to seperate your perceptions from the physics. That this effect occurs perceptually I have no doubt. I have doubts about it being an effect at frequencies below 125 Hz were there are discrete modes. What you are perceiving could easily be dominated by the sound > 125 Hz.
I did my PhD thesis on small rooms in the modal region. In that thesis I found that a dipole source could be modeled in either one of two ways and that they were complely equivalent. One was as a velocity source which excited the modes through their velocity contours and the other was as two monpoles, seperated by a distance and 180° out of phase. Both methods yielded the same results. Since superposition holds, each monopole in the dipole simulation would excite the modes as a monopole - proving that dipoles excit the same numbers of modes, just in different ways. The further appart the two monopoles became the greater the "dipole moment". If they are seperated by the room dimensions you have a close facimaly to what Marcus is talking about.
Doesn't it just seem logical then that using two monopole subs and allowing one of them to be set arbitrarily to the first would have to have a better chance at a smooth response than two monopoles with a fixed relationship?
This is precisely what I do if you think about it, only I add the room into the "Linkwitz transform" to get the best "net" response from this source in this room. Who cares what the response of the sub is in a free field - thats not where it is being used. Then add a second one, and a third if you like, and I claim that this result will equal any dipole configuration that you want to use. It won't necessarily be any better, but it won't be worse either. This is really a "multipole" - a linear combination of monopoles and dipoles set to achieve the best overall response in any given situation.
Markus
When I do EQ for my clients I often find that turning on one of the subs actually reduces the mean sound level in the room. Thus, this sub is working as a sink at that frequency, i.e. an active absorber. In this example then it would be very possible that a multi sub implimentation was reducing the modal decay.
I would add:
*barring thermal modulation or clipping resulting from the differences in actual efficiency.
Last edited:
A simple summary:
Monopole: can pressurize a room, For any given position what ever modes are excited are independent of orientation. Nominal power into room is 0dB reference.
Dipole: can not pressurize a room. Excited modes depend on orientation. Nominal power is - 4.8dB.
Cardioid: Can pressurize a room but the degree is frequency dependent, increasing at low frequency. Excited modes independent on orientation except for specific placement. Nominal power = -4.8dB.
Now, I want to review some basic considerations about modes. Starting with a monopole, the pressure due to a room mode for a listener at position r and the source at position ro is proportional to the product of the eigenfunctions at both positions. It doesn't matter what the form of these eigenfunctions are, if one or both are zero, then there will be no contribution to the pressure at the listening position.
p(ω,r|ro) ~ ∑ Φn(r)Φn(ro)
When the source is a dipole composed of two monopoles separated by some distance, d, the pressure at the listener is proportional to
p(ω,r|ro) ~ ∑ Φn(r) [Φn(ro+d/2) - Φn(ro-d/2)]
This is interesting because it says a couple of things/ First, if the eigenfunction at the listening position is zero there is no contribution to the SPL at the listening position. This is similar to a monopole. The next thing it says is that if eigenfunction for the front source has the same value at that of the rear source there is no contribution to the SPL at the listening position. In an ideal, rectangular room this occurs for axial modes which are perpendicular to the dipole axis when the dipole axis is aligned with one of the room directions. I also happens for tangential and oblique modes which include a contribution from such axial modes. This is the often used argument to support the statement that dipoles excite fewer modes. But there is another part of the story that is not generally discussed. Since the two sources are at different positions, there is a possibility that, for example, Φn(ro+d/2) (for the front source) is zero while Φn(ro-d/2) (for the rear source) is not. In this case the dipole modal excitement is reduces to that of the monopole:
p(ω,r|ro) ~ ∑ Φn(r) Φn(ro+d/2)
but there is another part to this. The SPL is also proportional to the source strength. This is where it gets interesting because, as you know, for a dipole to have the same free field axial SPL the source strength must be equalized at 6dB/octave. That means the below th dipole = monopole frequency the dipole source will be greater than the monopole source with the consequence that the dipole may excite such a mode to a higher degree that a monopole. With a typical dipole separation of 18" the dipole = monopole frequency is at about 125 Hz so at 60 Hz there is 6db greater source strength and at 30 Hz 12dB greater strength.
Looking at the cardioid is is similar to the dipole but there is an added delay to the rear source:
p(ω,r|ro) ~ ∑ Φn(r) [ p(ω,r|ro) ~ ∑ Φn(r) [Φn(ro+d/2) - Φn(ro-d/2)exp(-iωTd)]
As a consequence of this delay those modes perpendicular to the cardioid axis do not cancel as they did for the dipole. The lower the frequency the closer they come to canceling, but they never cancel exactly unless both Φn(ro+d/2) and Φn(ro-d/2) are zero which requires a very specific position. Again, similar to a dipole if placement is such that either Φn(ro+d/2) or Φn(ro-d/2) is zero, the excitement of such a mode reduces to that of a monopole. Again, there is the 6dB/octave eq applied which makes the resulting source strength greater than a monopole so that in such a case the cardioid may excite such a mode to a greater extent than a monopole. However, there is a difference between the cardioid and dipole. Give the same separation of 1.5', the cardioid = monopole frequency is an octave below the dipole= monopole frequency, at 62 Hz. The result is that the source strength, while greater than that of the monopole, will be weaker than that of the dipole. Additionally, above the cardioid = monopole frequency the source strength will be weaker than that of a monopole. So above 62 Hz there is a possibility that modes will be excited more weakly than for the monopole. (The same applied to the dipole but with the dipole = monopole at 125 Hz we are probably above the woofer's low pass cut off frequency).
I hope I haven't confused the heck out of you all. The intent here is simply to show that these arguments about what source does what are really all 1/2 truths. It is not possible to make a blanket statement regarding how a source excites room modes. Anyone who states categorically that dipoles excite fewer modes is selling you a bill of goods.
The bottom line in my option is do what you need to get the bass to sound right at your listening position. If it is a set up for a single listener that is probably pretty easy to do with eq. If you are going for a wide area of good bass in an acoustically small room, good luck. But as far as I am concerned, the best bass is only going to happen at one spot. The key for music is to make that spot the same as that for the optimum stereo image through the midrange and up.
The only thing that can be said categorically about dipoles comparedf to other sources is that they can not pressurize a room, and this is what I think drives listeners to prefer a dipole woofer. The same type of behavior can be obtained with a monopole source but it requires correctly matching the monopole to the room.
Monopole: can pressurize a room, For any given position what ever modes are excited are independent of orientation. Nominal power into room is 0dB reference.
Dipole: can not pressurize a room. Excited modes depend on orientation. Nominal power is - 4.8dB.
Cardioid: Can pressurize a room but the degree is frequency dependent, increasing at low frequency. Excited modes independent on orientation except for specific placement. Nominal power = -4.8dB.
Now, I want to review some basic considerations about modes. Starting with a monopole, the pressure due to a room mode for a listener at position r and the source at position ro is proportional to the product of the eigenfunctions at both positions. It doesn't matter what the form of these eigenfunctions are, if one or both are zero, then there will be no contribution to the pressure at the listening position.
p(ω,r|ro) ~ ∑ Φn(r)Φn(ro)
When the source is a dipole composed of two monopoles separated by some distance, d, the pressure at the listener is proportional to
p(ω,r|ro) ~ ∑ Φn(r) [Φn(ro+d/2) - Φn(ro-d/2)]
This is interesting because it says a couple of things/ First, if the eigenfunction at the listening position is zero there is no contribution to the SPL at the listening position. This is similar to a monopole. The next thing it says is that if eigenfunction for the front source has the same value at that of the rear source there is no contribution to the SPL at the listening position. In an ideal, rectangular room this occurs for axial modes which are perpendicular to the dipole axis when the dipole axis is aligned with one of the room directions. I also happens for tangential and oblique modes which include a contribution from such axial modes. This is the often used argument to support the statement that dipoles excite fewer modes. But there is another part of the story that is not generally discussed. Since the two sources are at different positions, there is a possibility that, for example, Φn(ro+d/2) (for the front source) is zero while Φn(ro-d/2) (for the rear source) is not. In this case the dipole modal excitement is reduces to that of the monopole:
p(ω,r|ro) ~ ∑ Φn(r) Φn(ro+d/2)
but there is another part to this. The SPL is also proportional to the source strength. This is where it gets interesting because, as you know, for a dipole to have the same free field axial SPL the source strength must be equalized at 6dB/octave. That means the below th dipole = monopole frequency the dipole source will be greater than the monopole source with the consequence that the dipole may excite such a mode to a higher degree that a monopole. With a typical dipole separation of 18" the dipole = monopole frequency is at about 125 Hz so at 60 Hz there is 6db greater source strength and at 30 Hz 12dB greater strength.
Looking at the cardioid is is similar to the dipole but there is an added delay to the rear source:
p(ω,r|ro) ~ ∑ Φn(r) [ p(ω,r|ro) ~ ∑ Φn(r) [Φn(ro+d/2) - Φn(ro-d/2)exp(-iωTd)]
As a consequence of this delay those modes perpendicular to the cardioid axis do not cancel as they did for the dipole. The lower the frequency the closer they come to canceling, but they never cancel exactly unless both Φn(ro+d/2) and Φn(ro-d/2) are zero which requires a very specific position. Again, similar to a dipole if placement is such that either Φn(ro+d/2) or Φn(ro-d/2) is zero, the excitement of such a mode reduces to that of a monopole. Again, there is the 6dB/octave eq applied which makes the resulting source strength greater than a monopole so that in such a case the cardioid may excite such a mode to a greater extent than a monopole. However, there is a difference between the cardioid and dipole. Give the same separation of 1.5', the cardioid = monopole frequency is an octave below the dipole= monopole frequency, at 62 Hz. The result is that the source strength, while greater than that of the monopole, will be weaker than that of the dipole. Additionally, above the cardioid = monopole frequency the source strength will be weaker than that of a monopole. So above 62 Hz there is a possibility that modes will be excited more weakly than for the monopole. (The same applied to the dipole but with the dipole = monopole at 125 Hz we are probably above the woofer's low pass cut off frequency).
I hope I haven't confused the heck out of you all. The intent here is simply to show that these arguments about what source does what are really all 1/2 truths. It is not possible to make a blanket statement regarding how a source excites room modes. Anyone who states categorically that dipoles excite fewer modes is selling you a bill of goods.
The bottom line in my option is do what you need to get the bass to sound right at your listening position. If it is a set up for a single listener that is probably pretty easy to do with eq. If you are going for a wide area of good bass in an acoustically small room, good luck. But as far as I am concerned, the best bass is only going to happen at one spot. The key for music is to make that spot the same as that for the optimum stereo image through the midrange and up.
The only thing that can be said categorically about dipoles comparedf to other sources is that they can not pressurize a room, and this is what I think drives listeners to prefer a dipole woofer. The same type of behavior can be obtained with a monopole source but it requires correctly matching the monopole to the room.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.