Hi Dave
Good questions.
Your points are basically revolve arround one central principle and that is that internal resonances act similar to or the same as HOMs. I will try and explain why I don't think that is the case, which is the crux of my position.
First, let me elaborate on my use of the term "group delay". I really should have said "excess group delay" as in non-minimum phase as this is what I am really talking about. Of course all systems have group delay, but for the most part they are minimum phase, i.e. they have a group delay that is explicitly tied to their frequency response.
That cleared up, internal resonances are minimum phase, they have no excess group delay, while HOMs are non-minimum phase, they do have excess group delay. This can be seen in different ways.
First an internal resonance of the primary wave (non HOM) does not change the the shape of the wavefront at the mouth. It is spherical in all cases, at a resonance or not. This is true to first order, and I will get to that later. This means that its effect will be seen to be the same at all radiated field points and it can in fact be exactly Eq'd by electronic means. This occurs because the wave reflected from the mouth travels back along the same path from wence it came, and as long as the wave velocity is constant (which it is) it is completely synchronous with the first primary wave and as such its effect can only be to modify the amplitude of the radiation. It is indistinguishable from say a diaphragm resonace.
This is not the case with HOMs. They are not minimum phase, because they do not travel along the same path as the primary wave. They do affect the wave shape at the mouth and hence their effect is different at every field point. They arrive delayed in time from the primary signal in a way that is not synchronous with the primary wave. I believe that this is essentially why they are so audible. Since they are not minimum phase and their effect is different at all field points, they cannot be corrected with electronic means. Once they exist, there effect is always there, somewhere. Acoustical problems like this cannot be corrected except acoustically.
I hope that it is clear that we cannot simply lump these two effects together and say that they would be equally audible or equally easy to detect.
Now, as I said before, it has been shown that excess group delay is audible with an increasing audibility with SPL level. When I contemplated your comments I had to admit that there is something about this situation that is relavent and yet I know of no studies that would apply. That is that IF the minimum phase effects are NOT more audible at increased SPL levels, then this is a key difference since the nonminimum phase ones have been shown to be more audible with SPL level (i,e, nonlinear in SPL). But the fact is that I know of no studies to the effect that a minimum phase resonace is or is not more audible with SPL level. If they are, well then, of course, all of Dr. Toole's work on resonances are moot, because they would apply only at a specific SPL and I doubt that this SPL was controlled. And, as I said, I don't think that the audibility of resonances has been studied with SPL level as a variable. If they are of constant audibility with level, as I suspect they are, then this is a huge difference between them and HOMs.
When we look at the impedance effect of a horn/waveguide only its "bulk" effect namely the average impedance across the wavefront will be seen. This will always be zero for HOMs because they all have an average value of zero across the wavefront. Only the primary wave does not. So when you look at a horns impedance you simply will not see any HOMs. This is key to understanding why a diffraction slot is such a problem. By its very nature a diffraction slot will create a multiplicity of HOMs, or lateral traveling waves after the slot. The reflection from the slot will create the standing wave and the well know resonance impedance response, but you will not see any of the effects of the HOMs in this impedance.
This brings me back to the point from above about the mouth. The reflection from a mouth will always generate some HOMs that travel back down the device and could, under the right circumstance, reflect off of the diaphragm and back out the device again. These too will not appear in any impedance measurements. (I can measure and have seen these with my Spherical polar measurement technique and maybe I will show that at some point. I can actually recontruct the velocity at it occurs at the mouth and I can see resonances across the mouth at certain frequencies indictaing very clear non-primary wave behavior.) This HOM mouth reflection is a "second order" effect, but I do not believe that it can be regarded as negligable. Mouth treatment in a good sounding device is critical and seldom done properly.
It is true that the foam acts to dampen both types of aberation, the minimum phase ones as well as the nonminimum phase ones. This of coarse makes sorting out one effect from the other very difficult. I do not have a good understanding of the relative values of the subjective contributions of these two things however. Its obvious that the foam is a major advantage, it is not obvious which of its effects is the more dominate. From my perspective, it doesn't really matter.
What you say about modeling the wavefront is quite true and precisely how I arived at the existance of HOMs. They have to be present or the math does not work out. It was only years later when I was trying to reconcile why some horns and drivers sounded so bad and others did not with the fact that it was NOT the driver (the B&C studies). What was it in the horn/waveguide device that resulted in this rather marked difference in sound quality. There are still questions, to be sure, but I have a lot more answers now than I had twenty years ago. The success of my designs is no small testimonial to the correctness of this work.
Dampening of sound radiation is a "time distributed effect" - hence some traveling distance is needed - simply put, you cant place some magic dampening material at a precise position.
besides that - at any spot in space where diffraction happens - meaning where the boundary changes angle along the propagation direction of the wave front - there would be a need to place some magic dampening material *if* you would like to suppress those second sources (according to what you have in mind)- but - those diffraction spots in space are there for a reason - and - if you even got such magic dampening material, it would be of no help as even a transition point into "acoustic non-existence" would create diffraction!
So - to repeat - diffraction can not be avoided - also - its always the same amount of diffraction thats actually happening from any piston sources (or whatever) radiating into 2Pi or 4PI (or whatever).
What *can* be selected though - is the *alignment* of those diffraction points in space - which is nothing else than selecting a different "horn / wave guide / diffraction alignment device" contour.
Michael
Last edited:
Or use a dedicated horn with asymmetric coverage in the vertical plain to even better attenuate ceiling reflections
Michael
first not clear - second not true..
wide parts of the rest of the post could have been almost "copy and paste" from my CMP paper
LOL
The "real" crux you are facing with establishing a "specific" behaviour of HOM versus any other diffraction caused CMP effects is that you are trying to state that a subgroup of specific spatial orientation should have a significant different perceptability (as to your ideas).
Quite a task I'd say...
Michael
Last edited:
Since my work predates yours by more than a decade, its really the other way arround isn't it?
well depends - that part of your work that does not make any sense definitely is that old...
Michael
My work has all been peer reviewed and exists in highly reputable professional journals. Where does your peer reviewed work reside?
Quite frankly Michael I am getting very tired of your nonsense. There are people here - in a tread entitled "Geddes on Waveguides" - who would like to actually learn about the subject. Your contributions, or lack thereof, are a real distraction from this.
Quite frankly Michael I am getting very tired of your nonsense. There are people here - in a tread entitled "Geddes on Waveguides" - who would like to actually learn about the subject. Your contributions, or lack thereof, are a real distraction from this.
Which shows that your beloved "CMP" "invention" is effectively nothing else than plain old and well known edge dffraction
Rudolf
good question...
seems I don't care for
what - on the other hand - seems to be all your pride ...
Learning is comparing...
And as far as I can see I completely stay within current (prolonged) HOM topic with my CMP contributions - be it of your taste or not.
You may own several patents - but - to my knowledge not any patent for being the one an only "time domain effects" spokesman in the world...
You always can prove CMP concept to be wrong though - if you can - but as far as I see it for now, you are rather on the course to claim if for yourself
- quite a funny 180° turn I'd say.
Michael
Last edited:
Which shows that your beloved "CMP" "invention" is effectively nothing else than plain old and well known edge dffraction
Rudolf
not exactly - but those "well known edge diffraction" definitely is covered by CMP concept
Michael
I'm not experienced with adding foam to horns, so I'm not a good one to ask. It seemed that the curves a few posts back showed that using foam that was open enough to minimize the HF attenuations gave fairly minimal effect on the classic horn problems. As to the effect on HOMs, I'm still new to that subject too.
I can speak a little about the biradial designs. Most of the shape is defined by the beamwidth targets. You specify a vertical coverage and that defines the initial slope of the top and bottom sides. You specify what frequency you want the hroizontal coverage to be good to and that defines the diffraction slot width. At the diffraction slot you apply a lateral cross section based on your target for the horizontal beamwidth.
From the starting point of either contour, Don defined a short power series that created the end flaring, so the starting angle and the end flaring defined both cross sections completely. The only thing left, and the only aspect that is exponential, is from the diffraction slot back, or from compression driver exit up to the slot. Since the back area is fixed (compression driver throat area) and the front area defined by a width you set to determine upper horizontal beamwidth frequency, you have two areas and a length of horn between them. You can define it as exponential but the cuttoff frequency is preordained by your other choices.
So there is a clear tradoff between polar curves and response smoothness. Choose a narrow slot for wide (horizontal) bandwidth and your horn cutoff frequency drops or may even go negative. A low flare rate with insufficient mouth area is a recipe for choppy response.
In the end the art is in finding the right balance of tradeoffs.
(Try the foam in the diffraction slot and send us before and after curves.)
David S.
Thanks a lot David (though I need to learn quite a few more things to understand half of what you said here, and my bad english does not help...)
What I have at hand are H9800 horns (clones, but should be quite close), the horns that goes into the JBL S9800. In this design the slot is used sideway.
By the way, what do you think of Greg Timbers' array line (Array 1400) with similar horns turned vertically to get the benefit of the slot on the horizontal plan?
Well, maybe I should open a new "Speaker Dave on biradial horns" and ask you these questions there. That would really be interesting for a lot of us that are using biradial horns!
Would you be okay for that?
What I have at hand are H9800 horns (clones, but should be quite close), the horns that goes into the JBL S9800. In this design the slot is used sideway.
By the way, what do you think of Greg Timbers' array line (Array 1400) with similar horns turned vertically to get the benefit of the slot on the horizontal plan?
Well, maybe I should open a new "Speaker Dave on biradial horns" and ask you these questions there. That would really be interesting for a lot of us that are using biradial horns!
Would you be okay for that?
One of the problems with "stuffing the throat" is that it is not a good idea to compress the foam in place. This makes it denser in unpredictable ways. My plugs are not compressed in any way, they just sit in the opening - glued in. Shaping the foam for a to fit like this in the throat would be a problem to cut since there is a lot of currvature there. Thats another reason why I avoid the throat.
good question...
seems I don't care for
that - on the other hand - seems to be all your pride ...
Michael
Well I guess if you aren't interested in people taking you seriously then it doesn't matter. I'm certainly not going to seriously consider any concept that has not been taken through an independent peer review process.
Yea, I'm proud of the work that I've done. Whats wrong with that?
Dave
Just for the record, thats true for diffaction horns, but not for waveguides. That's precisely why the waveguide was developed.
Well, I think I just revealed the sum total of my knowledge. If you are interested in the JBL horns you should read what Don Keele, the designer had to say:
AES Papers -- Official website of D.B.Keele
http://www.xlrtechs.com/dbkeele.com/PDF/Keele (1975-05 AES Preprint) - Whats So Sacred Exp Horns.pdf
Between those articles and Earl's articles you have a pretty good survey of horn theory.
David