First of all, if I understand what you mean, what you call CMP is just a name for a response that is composed of a summation of impulses which when taken individually represent a MP response. In other words the system frequency response is the sum of a series of minimum phase frequency responses. You should recognize that such a summation need not be MP. That is, a system which is CMP does not necessarily have to be MP. Similarly, a system can be MP even though the individual components in the summation are not.
With a dipole, I have said numerous times that the dipole response is not usable above the dipole peak. Being MP is one thing, but the nulls, while exhibiting MP behavior are not suitable for equalization. The point I have made repeatedly is that when you start with a MP response and equalize it to a specific MP target (like a band pass response) then the impulse of the eq MP response will identical to the impulse of the target.
You know we should really define a dipole a little better. What we really have, in the simplest case of two sources separated by some distance, is not a dipole. It is an acoustic doublet. The polar response of an acoustic double approximates a dipole response as the frequency decreases. The difference between a dipole response and the polar response of an acoustic doublet becomes acceptably insignificant about 1 octave below the "doublet" peak. And above that frequency the doublet response really should not be referred to as a dipole as it has no relation to a dipole at all, though it does remain MP.
Now if you want to know if the response in MP or not, make a measurement and then compare the phase of the measured response to the phase generated form the amplitude response using the Hilbert Bode transformation. If, by elimination of only excess delay the measured phase reduces to the HBT phase, then you can conclude that the measured system behaves as a MP system. If that is true, then any MP equalization applied to the system will also result in a MP response. I never said how this eq should be constructed and I would not limit that discussion to caps and coils.
Perhaps credit should be given to:
THOMAS J. HOLMES, The "Acoustic Resistance Box" A Fresh Look at an Old Principle, J. Audio Eng. Soc., Vol. 34, No. 12, 1986.
Juha Backman, Theory of acoustical resistance enclosures, Presented at the 106th Convention, 1999 May 8-11, Munich, Germany.
Gary's system, as well and my NaO woofer system are just applications of past develoments.
Wow, now that makes sense. Thank you for taking an acoustic phenomenon and explaining it in terms an EE can understand. After thousands of posts and understanding very little of it, posts #6716 and #6721 have really helped me understand the basics of combing and baffle size.
True. Hope more people will realize this.
After looking at the illustrations of the resistive box, Those would be nice in a very limited bandwidth, cross them too high, and the sound will be sound as though part of the music is compressed. Tricky enclosure to design.
I absolutely agree with John. The resistive boxes I've setup are just re-hashes of ideas used in the past using modern drivers and damping materials. Pleasing results, fun to build, measure, and listen to, but nothing revolutionary.
The discussions in this thread are great. Lots of great information and view points to observe and absorb.
Thanks for responding Gary. I think it is important to the general audience here that they realize many of the topics being discussed have been well researched over the years. What was old is new again. As I point out on the web page I referred to, this type of enclosure goes back tot he 50's. Of course, what we have today is better materials with which to damp the back wave and better drivers capable of the necessary excursion with resulting better execution of the concept.
True. Hope more people will realize this.
After looking at the illustrations of the resistive box, Those would be nice in a very limited bandwidth, cross them too high, and the sound will be sound as though part of the music is compressed. Tricky enclosure to design.
From some quick calculations (could be in error) it appears to me that if the rear wave were uniformly attenuated by 6dB the nulls would still be on the order of 12dB deep (as opposed to true nulls), the peak would come out around 7 dB, the roll off below the peak would be initially steeper but then plateau at -12dB. This assumes point source behavior.
But a resistively damped box will never have uniform 6dB attenuation. Resistive damping of the rear wave always results in an LP type response.
John, thats partly an elegant summation - in other words - on the perspective I brought up.
Its obvious that OB is not "true" dipole but just an approximation that works pretty well.
You missed to clearly point out that its even an approximation *below* dipole peak, and you also missed to point out that this is so because OB simply is a CMP system (if you have an other term from "ancient" researchers in that topic - no problem for me ).
Now - again - with the discussion about CMP, I'm neither *specifically* focusing on OB - and even less on OB above dipole peak (I gave the reasons why I'm referring to OB nevertheless in demonstration and description and of course above dipole peak is valid for that as well as below dipole peak).
I'd like to stress though, that CMP systems do *not* depend on frequency at first hand - hence I asked for your proposal to correct CMP (be it in OB or elsewhere on constructive / destructive interference in audio) be it as low in frequency you wish.
On the pure theoretical level, I guess, you hardly will come up with anything different than I've shown - at least *if* you correct as 100% for the CMP behaviour as already shown above.
I would not necessarily have restricted you to caps and coils - though I was under the impression that "min phase" correction *always* could be done by caps and coils as well.
Michael
Its obvious that OB is not "true" dipole but just an approximation that works pretty well.
You missed to clearly point out that its even an approximation *below* dipole peak, and you also missed to point out that this is so because OB simply is a CMP system (if you have an other term from "ancient" researchers in that topic - no problem for me
).Now - again - with the discussion about CMP, I'm neither *specifically* focusing on OB - and even less on OB above dipole peak (I gave the reasons why I'm referring to OB nevertheless in demonstration and description and of course above dipole peak is valid for that as well as below dipole peak).
I'd like to stress though, that CMP systems do *not* depend on frequency at first hand - hence I asked for your proposal to correct CMP (be it in OB or elsewhere on constructive / destructive interference in audio) be it as low in frequency you wish.
On the pure theoretical level, I guess, you hardly will come up with anything different than I've shown - at least *if* you correct as 100% for the CMP behaviour as already shown above.
I would not necessarily have restricted you to caps and coils - though I was under the impression that "min phase" correction *always* could be done by caps and coils as well.
Michael
John
I don't follow this distinction. Are you saying that the "doublet" is two out of phase sources seperated by a distance and the "dipole" is two out of phase sources seperated by a baffle? Otherwise I don't see what you are getting at.
Actually after I posted I realized that I really didn't say what I wanted to say and it was to late to edit it. What I was trying to distinguish was the difference between the radiation pattern of an acoustic doublet (or acoustic dipole) which is a function of frequency and the classic dipole figure 8 pattern associated with things like magnetic and electric dipoles. See it even confused me.
My thinking was that when the term dipole is read people generally think of the classic figure 8 patten which only applies at low frequency in an acoustic dipole or doublet. So I was thinking that maybe using the term acoustic doublet would break the connection that dipole = figure 8 pattern.
And I am never sure when some one else talks about a dipole if they are talking about a figure 8 pattern or the frequency dependent characteristic that degrades from a figure 8 to a daisy peddle affair as the frequency rises above the first on axis peak.
All this talk about nulls and reducing nulls by damping the rear wave is a little disconcerting to me because once the frequency rises to the point where those nulls or dips are present the response no longer resembles any type of constant directivity and thus is useless from my (and your
) point of view. That is why I keep repeating the statement that an acoustic dipole (or doublet) has a useful frequency range to about 1 octave below the first on axis peak. Above that the polar response degenerates quickly form the figure 8 pattern. Below that the error is acceptable.
John
A question for you, or anyone else.
When I looked at the Orion measurements it was obvious how well it controlled the directivity in the 200 - 500 Hz range, where I am not able to effect much control. Hopefully it is not a coincidence that this is precisely the region where I think the Orion shines. In my book I showed how a rear facing driver could be used to control the directivity in precisely this range while going away at LFs to yield a more efficient monopole. It turns out that this can be done passively, albeit not simply (passive attracts me for obvious reasons). I am going to try this and wondered if you, or anyone else, has actually done this as well and what your experience was (technically please!). This would not be too hard to impliment and may be an audible improvement. (There is no point in not disclosing this since its already public domain anyways.)
A question for you, or anyone else.
When I looked at the Orion measurements it was obvious how well it controlled the directivity in the 200 - 500 Hz range, where I am not able to effect much control. Hopefully it is not a coincidence that this is precisely the region where I think the Orion shines. In my book I showed how a rear facing driver could be used to control the directivity in precisely this range while going away at LFs to yield a more efficient monopole. It turns out that this can be done passively, albeit not simply (passive attracts me for obvious reasons). I am going to try this and wondered if you, or anyone else, has actually done this as well and what your experience was (technically please!). This would not be too hard to impliment and may be an audible improvement. (There is no point in not disclosing this since its already public domain anyways.)
I got an impression that you seem to be quite interested in dipole and brought it up several times. However you just haven't tried it yet, have you?
I have a design, while originally not designed for the purpose, but would be interesting how it effects directivity. However, 200~500Hz range is still a sensitive region where tradeoff between a simple wavefront and controlled directivity is VERY room dependent.
I got an impression that you seem to be quite interested in dipole and brought it up several times. However you just haven't tried it yet, have you?
No, I haven't, and there is no guarantee that I will have the time to "try" it anytime soon either. It's not a major thing at all, it's a slight improvement in the 200 - 500 Hz region - not earth shattering, not a paradigm shift, nothing like that. And I have heard lots of dipoles before, so its not new from that perspective either. As I said, the technique is described in my book, which is almost ten years old now, so its nothing new.
Its above the Schroeder frequency and so it should not be room dependent at all.
First steps have been undertaken – and already presented - no ?
There is some hard work to do in establishing a minimum of common ground on the subject of CMP first – and thats I'm even more interested in …
All the rest will come along by its own then.
...See it even confused me.
I'd say you begin to see the power and implications of my concept.
As said – CMP is part and parcel to OB.
Its been you – IIRC - who stated that dipole can *always* be seen as two point sources separated by some distance.
In my understanding too this is not *exactly* the case – as I agree to your conclusion that we rather should keep dipole *directivety* at one hand and CMP (by OB) at the other hand clearly separated.
As said once in a pronounced manner – due to OB being a CMP system - we listen to a *monopole* for the first time interval and only after that time interval (given by the delay) we actually listen to a dipole.
and to repeat once more: with OB this is the case well below the dipole peak as well !
Coming back to "nude" speakers:
There we actually have dipole behaviour *and* we start out with (practically) no delay.
Sure – CMP is involved here too (later on) – but CMP is involved here in a very special form.
-------------
As a side note :
I'm deeply impressed by the lack of reaction to my statement that CMP *in genereal* flaws our most used tool in measurement - as clearly demonstrated.
for any uncorrected CMP system the frequency response becomes a matter of the time interval we look at!
Michael
Last edited:
As a side note :
I'm deeply impressed by the lack of reaction to my statement that CMP *in genereal* flaws our most used tool in measurement - as clearly demonstrated.
for any uncorrected CMP system the frequency response becomes a matter of the time interval we look at!
Michael
Above becomes even more impact with looped reflections as seen with horns and compression drivers.
Michael
I possibly should be more precise here in stating that with nude speakers we start out (practically) with
"no delay - meaning: no delay that *sharply* separates two CMP intervals"
Michael
Last edited: