Sorry if I don't answer that "in detail" - It may be interpreted as "personal painting" as it has not so much to do with the "physics" of HOM but with how and when and for what reasons it pops up here...
Michael
To actually see "performance improvement" versus detrimental effects of damping materials - first hand those detrimental effects have to be characterized.
This, to my knowledge has not been done so in the past, though "everybody knows" and there is lots and lots of "anecdotal evidence" available
Michael
There isn't much to say about it. Until yesterday I was ignorant about the supposed non-linear properties of dampening material. Assuming that dampening material doesn't have non-linear properties and that all sound radiated into the box is undesirable, I think you should turn as much of it as possible into heat, before it finds a way to your ears. Stuffing the box to the brim does that best, as long as you don't restrict air-flow too much.
I'll come back to you on that topic...
Michael
Much appreciated!
Not to encourage this topic, but this clears it up for me. Filling the box but not compressing the damping material too much. Before this was clarified, there was potential for misunderstanding.
Hello speaker dave,
You can eventually study the impulses reponses given in the posts:
http://www.diyaudio.com/forums/multi-way/103872-geddes-waveguides-514.html#post2555612
http://www.diyaudio.com/forums/multi-way/103872-geddes-waveguides-514.html#post2555957
Best regards from Paris, France
Jean-Michel Le Cléac'h
You can eventually study the impulses reponses given in the posts:
http://www.diyaudio.com/forums/multi-way/103872-geddes-waveguides-514.html#post2555612
http://www.diyaudio.com/forums/multi-way/103872-geddes-waveguides-514.html#post2555957
Best regards from Paris, France
Jean-Michel Le Cléac'h
Could somebody point me to a curve comparison of a horn with and without the internal damping foam? I think I've seen it somewhere in the past (maybe here a couple hundred pages back
I would think a good set of curves would settle the argument, at least for those that value measurements.
David
Thanks Jean-Michel.
I was looking for simple before and after response curves showing the effect of the foam insert.
I found this: http://www.diyaudio.com/forums/multi-way/145096-how-hom-measured-5.html
Is it representative?
David S.
I was looking for simple before and after response curves showing the effect of the foam insert.
I found this: http://www.diyaudio.com/forums/multi-way/145096-how-hom-measured-5.html
Is it representative?
David S.
Hi Dave
Let me give a little historical background since, from my perspective, I differ in my beliefs from most others posing here.
Way back in early 2000's I worked for B&C to study what consititutes "sound quality" for compression drivers. To make a long story short we found "not much" - they did not appear to be a major factor in sound quality. (This work was all published in the AES.) A position that I now hold and promote. But there certainly IS a difference in sound quality for different horn/driver combinations - what IS it?
Looking to the waveguide device itself (which I had actually done some ten years earlier) I began to suspect that the HOMs that I had identified in my work, and others had verified the existance of (i.e. Makarski), might be an audible factor. So I set about to design a system with minimal HOMs to test this concept. That design is what I use in my products today and all the data that I know of indicates that my hypothesis was correct. HOMs are a serious audible problem that when reduced (they can never be elliminated) the devices sound quality is dramatically improved.
How to "prove" this became an issue (really the key issue here). Lidia and I did show that an HOM type of signal could be highly audible and that its audibility increased with increasing SPL - the exact opposite of nonlinear effects, such as THD or IMD. This was strong supporting evidence. But the direct measurement and comparison between a high HOM and lowered HOM unit and its correlation to audibility was lacking. It still is.
I became too busy with my speaker business and all "research" related to this topic either ceassed or was done as proprietary. This is where the "argument" is today, as I see it.
That HOMs exist is IMO incontrovertable, but as to their audibility there is supporting evidence, but most of the subjective data is circumstantial. Many people are saying many things, but I agree with very few of them. Take for example the use of "wavelet analysis". This will not distinguish a resonance from an HOM very readily if at all. I have discussed how I believe this needs to be done (above), by using a spatial mode decomposition to find the modal content of a wavefront from which we can directly seperate the primary mode from the HOMs. I am currently working on this, but at a very slow pace.
The foam acts to dampen all waves, but the fact that HOMs travel a longer distance in the device than the primary wave means that they get damped to a greater extent than the primary wave. This is the concept. The foam also damps standing waves, but in my devices there really aren't standing waves so this is clearly not the reason for the foam. This is also why you will not see much difference from the foam if you just look at standard frequency response curves. Thats why I said that I did not think that my measurements would settle the issue because they show very small difference due to the foam, while subjectively there is a much larger difference. Clearly these measurements are not showing the correct effect and I do not suspect that they ever will.
When looking at comparisons, you have to remember that I EQ to the foam in place which means that any bulk frequency response changes are EQ'd out.
Thats how I see the situation.
Last edited:
Hi Earl,
Thanks for the detailed response and bit of history on the subject.
I guess my assumptions on the topic were wrong in that I was thinking in terms of the general length and mouth and diffraction slot reflections that create the ripples in response in the typical horn. In fact, since you frequently comment on diffraction style horns and I know they can be rougher in response, I was assuming the foam was a solution to that effect.
While I know, in general, what a higher order mode is, can you explain the expected effect? Are we talking a linear distortion or a non-linear distortion? Would we see it in the impulse response and hence be able to convolve it with a signal, add it and subtract it at various levels and find thresholds of detection? Or is it nonlinear (as you imply with it being perceived as a function of level)?
I agree that the frequency response effects of the foam are nominal and could be easily EQed out, and I certainly have no fear of secondary subjective effects of foam presence. But I'm having a hard time visualizing something that isn't easily seen in the frequency response curve or traditional distortion measurements.
Sorry if this is common knowledge to others, I'm a little late to the party.
David
Thanks for the detailed response and bit of history on the subject.
I guess my assumptions on the topic were wrong in that I was thinking in terms of the general length and mouth and diffraction slot reflections that create the ripples in response in the typical horn. In fact, since you frequently comment on diffraction style horns and I know they can be rougher in response, I was assuming the foam was a solution to that effect.
While I know, in general, what a higher order mode is, can you explain the expected effect? Are we talking a linear distortion or a non-linear distortion? Would we see it in the impulse response and hence be able to convolve it with a signal, add it and subtract it at various levels and find thresholds of detection? Or is it nonlinear (as you imply with it being perceived as a function of level)?
I agree that the frequency response effects of the foam are nominal and could be easily EQed out, and I certainly have no fear of secondary subjective effects of foam presence. But I'm having a hard time visualizing something that isn't easily seen in the frequency response curve or traditional distortion measurements.
Sorry if this is common knowledge to others, I'm a little late to the party.
David
Earl,
Thanks for pointing out that a direct measurement and comparison between a high HOM and lowered HOM unit and its correlation to audibility has not been done.
For some time I thought there were some measurements, rather than just a subjective SQ difference, that I had missed somewhere in all the posts regarding HOMs.
Art Welter
I appreciate you repeating what actually was the end result of a "flaming" log term discussion - spread out over several threads - between you and me (and some others) quite a while ago.
To sum it up:
all you brought to the table - with respect to HOM relevance regarding horns / waveguides / diffraction alignment device - is some anecdotal evidence from you, your wife plus some other persons.
Not that I would not be interested if your effort to trace it down bear fruits - but until then I prefer to call your HOMania simply "bogus" - as usually is done for any other snake oil claims.
Besides my outspoken point of view, I'm always open to being convinced otherwise - but there certainly is needed more than some hand waving.
First off you would have to explain, as to why you think that those HOM - being a simple subset of diffraction effects - are any *significantly* stronger dampened by horn stuffing.
If we see the origin of those HOM called diffraction effects at the "throat" of a horn than anything that is radiated by the horn is affected equally - *as long* - as we do not account for the looped reflections.
If we see the origin of those HOM called diffraction effects happening from along the boundary (the horn contour plus the "mouth") then its part part (and even a little bit more complex) - but - in any case it then must be seen as a "part spectrum CMP effect" (as actually *any* diffraction has to be seen that way) not any different with respect to dampening as any simple transmission line.
Michael
Last edited:
From what I read, you seem to be doing the hand-waving and it's in an unseemly fashion I would say.
The descriptions already given are perfectly logical, but that does take a certain level of critical thinking and objectivity that seems to be lacking. I'm always open to being convinced otherwise.
I haven't taken a position on any of it, but your attitude and method of argument lead me to discount your comments. There's little to learn from them in any case. It seems to be nothing more than repetitive hand-waving and self-congratulatory nonsense.
Dave
Dave
Whilke I have gone over this many times, its still not "common knowledge" because so many people get it wrong.
First HOMs are linear, so it's not a nonlinear effect. But that does not mean, and in fact this is the case, that thee will be the same subjective effect at low levels as at higher levels. This is such a key point that gets missed all the time. Diffraction and HOM are linear effects with a nonlinear perception. Here's why.
Both of these effects create a signal which is displaced in time from the direct signal. In other words they exhibit group delay. Now the audibility of group delay has been studied, most notably by Moore and he concluded that for most "practical situations" it was not audible, but he clearly found situations where it was. He also noted that it appeared in his results, and he noted that this was known, that the audibility of group delay would increase with absolute SPL level. This is in his study as a statement of fact, but he did not test this. His tests were all at a constant SPL level.
Lidia and I tested the audibility of group delay and found a significant factor with both the level of the signal aberation AND the absolute SPL playback level. Because of OSHA we could not go above an in-ear level of about 85 dB, which is not very high, and yet we found a significant result. At higher SPLs I suspect that this effect would be quite pronounced.
Basically the ear tends to unmask sounds in time at higher SPL levels, while it tends to mask sounds in frequency at higher SPL levels. So as the SPL goes up nonlinear effects are becoming less audible while diffraction and HOMs (group delay effects) are becoming more audible.
This means that a linear effect can and does have a nonlinear perception in our hearing system. This has profound implications. Our paper has largely gone unnoticed because Lidia and I could not present it so it has only been published as a preprint but never seen. (Lidia's mother had died and we had to attend the funeral in China. The Papers Chairman would not allow the paper to be given by a non-author. I was annoyed by this.) Floyd does mention it in his book I believe.
Sound quality that we associate with worsening sound at higher SPL levels has always been presumed to be nonlinear distortion, but that, as I have discussed, is not necessarily the case. In fact it can be very difficult to determine if it is a nonlinear effect or a linear effect with nonlinear perception. The two things can get completely blurred together.
So none of this will show up on a distortion test, and in fact it has only a small effect on the measure frequency response, indeed its effect in frequency can almost be inperceptable to the eye.
I looked at using group delay to find this effect, but that turned out not to work so well. Thats why I am looking at modal radiation techniques.
You can see the effect on the impulse response and you could use convolution to manipulate it, but remember that this will be different at every field point so there is no way to globally "correct" it electronically.
What you suggest as a way to test is pretty much what Lidia and I did in our test.
For the information of the reader who may not understand the situation:
Michael and I did argue the points that I make here at some length in the past. As the argument proceeded he got more and more insulting - to wit his posts here - and I simply stopped responding to him. He took this as an admission of defeat and has been waving a flag of victory ever since. He simply doesn't get it.
I would love to have JBL engineers posting their view point on this. Their designs are very successful so they must sound good to someone and their K2s do sound amazing so maybe HOM simply isnt an issue for the average person or the majority??
Not only JBL but Tom Danley certainly does not see the HOM issue as a problem. Its hard to determine which viewpoint to take on the subject, for myself I didnt find the FOAM doing a great deal of anything on my QSC HPR-152i + BMS4550 design.
Not only JBL but Tom Danley certainly does not see the HOM issue as a problem. Its hard to determine which viewpoint to take on the subject, for myself I didnt find the FOAM doing a great deal of anything on my QSC HPR-152i + BMS4550 design.
Just to be clear, my reference is to the descriptions of the damping mechanism and distance through the damping material, foam in this case. Perfectly logical as I see it. Talk of so-called CMP is a side-show and should not be part of this thread IMO.if you think so ....
Michael
Dave
And they are going to admit that its a problem, when the solution is patented, because?
As I mentioned before, JBL was a licensee of my OS waveguide. They seemed to think something of that design and continued to use it.