Soongsc
Short of my own work in this area you will find only "talk". I have met very few who really even understand the concepts let alone done scientific work on them. I have even seen them described completely erroneously.
Its not "manufacturing tollerenaces" that are the issue at all. Its the design that determines the bulk of the HOMs.
Short of my own work in this area you will find only "talk". I have met very few who really even understand the concepts let alone done scientific work on them. I have even seen them described completely erroneously.
Its not "manufacturing tollerenaces" that are the issue at all. Its the design that determines the bulk of the HOMs.
gedlee,
It was your identification of HOMs that inspired me to even consider looking into horns and wave guides, and I appreciate it very much. Looking at the frequency span of audio signals, it certainly seems that HOMs are unavoidable, but it also seems that the closer to the throat, the more critical the manufacturing tolerances, and the effected frequency seems to be in the top end of the spectrum. Additionally, the compatibility between driver and horn/guide becomes critical. Lots of work is required to optimize things, yet, the financial returns may not interest most profit orientated organizations. This is probably why not so much work is done in this area. Hopefully we whom don't have to do this for a living can push the envelope of application of this.
It was your identification of HOMs that inspired me to even consider looking into horns and wave guides, and I appreciate it very much. Looking at the frequency span of audio signals, it certainly seems that HOMs are unavoidable, but it also seems that the closer to the throat, the more critical the manufacturing tolerances, and the effected frequency seems to be in the top end of the spectrum. Additionally, the compatibility between driver and horn/guide becomes critical. Lots of work is required to optimize things, yet, the financial returns may not interest most profit orientated organizations. This is probably why not so much work is done in this area. Hopefully we whom don't have to do this for a living can push the envelope of application of this.
Vertical coverage
Firstly I’d like to thank Dr. Geddes for sharing his invaluable knowledge with us all...
This may be going back many pages (just got to the end of this mammoth from start to finish) but I’d just like to say something on the subject of many posts from wayne (the ‘Pi’ speakers guy). He claimed many times that it would be desirable to limit the vertical coverage angle of the waveguide to within the nulls formed at the crossover with the woofer, however I would suggest that this would be undesirable. As Dr. Geddes points out many times, to get a good sounding system in a ‘small’ room there needs to be a good ratio of reverberant to direct sound at the listening position, particularly within the crucial 2-6khz region. This limiting of the vertical directivity above the crossover point would surely weaken the reverberant field a fair amount?
Maybe its because I listen in a high ceilinged room, but surely at least ceiling (if not floor) bounce is going to be >10ms so not an issue?
Firstly I’d like to thank Dr. Geddes for sharing his invaluable knowledge with us all...
This may be going back many pages (just got to the end of this mammoth from start to finish) but I’d just like to say something on the subject of many posts from wayne (the ‘Pi’ speakers guy). He claimed many times that it would be desirable to limit the vertical coverage angle of the waveguide to within the nulls formed at the crossover with the woofer, however I would suggest that this would be undesirable. As Dr. Geddes points out many times, to get a good sounding system in a ‘small’ room there needs to be a good ratio of reverberant to direct sound at the listening position, particularly within the crucial 2-6khz region. This limiting of the vertical directivity above the crossover point would surely weaken the reverberant field a fair amount?
Maybe its because I listen in a high ceilinged room, but surely at least ceiling (if not floor) bounce is going to be >10ms so not an issue?
How much room interaction is best is really subjective. If the original recording was done in a fairly dead environment, the room interaction is necessary. For live recordings, the original reverberation is also recorded, if there is too much room interaction in the playback environment, then it messes things up.
Re: Vertical coverage
It is correct that a tall room would not benefit as much from a narrower vertical directivity. I would prefer not to excite the vertical aspects of a room if possible because its mostly the horizontal reflections that add spaciousness.
My disagreement with Wayne is that it is not obvious that what he proposes would work in the simple way that he claims or that there would in fact be any improvement. I'd like to try it and find out but my suspiciian is that the difference will be negiligable, but the costs will be high. Not a good tradeoff.
Theo404 said:
It is correct that a tall room would not benefit as much from a narrower vertical directivity. I would prefer not to excite the vertical aspects of a room if possible because its mostly the horizontal reflections that add spaciousness.
My disagreement with Wayne is that it is not obvious that what he proposes would work in the simple way that he claims or that there would in fact be any improvement. I'd like to try it and find out but my suspiciian is that the difference will be negiligable, but the costs will be high. Not a good tradeoff.
I remember somewhere someone mentioned about the foam padded horn (WG) mouth by Peavey (do I remember it correctly? I can't find it now). The foam padded mouth is to "mimic" the effect of a real large radius round-over.
This is interesting, and should be easier (or cheaper) to do in some case.
So I tried this:
This is an 8" Dayton WG bought from PartsExpress. Its overall flare is quite smooth, but the mouth opening to the outer surface (baffle) is not very rounded over.
Here I cut some wool felt and stick them around the mouth, as the pictures above. This made it sound more open.
Originally this 8" one sounded more 'concentrated' and 'congested' than the 12" one of the same brand. I think it's because the different flare angle -- 12" one opens up more rapidly, thus wide and shallow in overall shape. In direct comparison, the 12" one sounds much more relaxed and open on axis and spreaded the treble in a wider listening area.
Now the padded 8" one got a similar performance as the larger 12" one -- very open sound and larger coverage. I can hear the treble coming from the opposite side of speaker smoothly and clearly when standing right in front of this side. (Of course it's getting some help from the large toe in angle, but it's still fresh to me.)
Sorry for all of the subjective descriptions and not providing any measurements. I have no proper instrument and software to do it right.
It's easy to do and cost almost nothing. So I hope someone with tools and skills of measurement would spare some time for this and share some data.
Any comments are welcome
This is interesting, and should be easier (or cheaper) to do in some case.
So I tried this:
This is an 8" Dayton WG bought from PartsExpress. Its overall flare is quite smooth, but the mouth opening to the outer surface (baffle) is not very rounded over.
Here I cut some wool felt and stick them around the mouth, as the pictures above. This made it sound more open.
Originally this 8" one sounded more 'concentrated' and 'congested' than the 12" one of the same brand. I think it's because the different flare angle -- 12" one opens up more rapidly, thus wide and shallow in overall shape. In direct comparison, the 12" one sounds much more relaxed and open on axis and spreaded the treble in a wider listening area.
Now the padded 8" one got a similar performance as the larger 12" one -- very open sound and larger coverage. I can hear the treble coming from the opposite side of speaker smoothly and clearly when standing right in front of this side. (Of course it's getting some help from the large toe in angle, but it's still fresh to me.)
Sorry for all of the subjective descriptions and not providing any measurements. I have no proper instrument and software to do it right.
It's easy to do and cost almost nothing. So I hope someone with tools and skills of measurement would spare some time for this and share some data.
Any comments are welcome
I you want, I could probably drop by and do some measurements.CLS said:
...
Sorry for all of the subjective descriptions and not providing any measurements. I have no proper instrument and software to do it right.
It's easy to do and cost almost nothing. So I hope someone with tools and skills of measurement would spare some time for this and share some data.
Any comments are welcome
So, you removed a "fair usage" quote and the URL.
Parts of copyright material are published all over this site including some threads in which moderators are constant participants. So, what's so exceptional here?
Dubious move, guys.
Methane! I smell methane!
SY, VARIAC, are you really into this?
gedlee said:
Earl, although this paper did not reference your work, FWIW Charles Hughes referenced your work in his original AES preprint. However, I think that Hughes' broadest patent claims may not be valid claims in view of at least your work. The broadest claims are not limited to the quadratic expression. They only refer to a curved section starting off perpendicular to the throat plane and ending up tangent to the straight wall.
Re: Vertical coverage
The floor is usually the closest boundary, and the ceiling is usually the next closest boundary. Most agree that refections from the floor and ceiling are not a good thing, and would also agree that limiting the vertical pattern helps in this regard.
Another issue is the position of the vertical nulls. The closer you can space the sound sources, the further apart the nulls are. I would suggest that any arrangement that brings vertical nulls too close together is a problem, because the angle between the forward listening axis and the vertical nulls becomes too narrow.
I think most would agree that the forward lobe is the only truly desirable sound field, and that side lobes and the nulls between them are unwanted artifacts. Ideally, a forward lobe can be generated that is tall enough to cover the listening area, and sound radiation limited at angles outside the vertical null angles.
Theo404 said:
The floor is usually the closest boundary, and the ceiling is usually the next closest boundary. Most agree that refections from the floor and ceiling are not a good thing, and would also agree that limiting the vertical pattern helps in this regard.
Another issue is the position of the vertical nulls. The closer you can space the sound sources, the further apart the nulls are. I would suggest that any arrangement that brings vertical nulls too close together is a problem, because the angle between the forward listening axis and the vertical nulls becomes too narrow.
I think most would agree that the forward lobe is the only truly desirable sound field, and that side lobes and the nulls between them are unwanted artifacts. Ideally, a forward lobe can be generated that is tall enough to cover the listening area, and sound radiation limited at angles outside the vertical null angles.
pooge said:
Charley Hughes did not write that paper - I have no complaints with him. But when I talked to the author he actually tried to say that Charley did not reference my work because he didn't use it, which was bull since I was a consultant to Peavey and personally explained it all to Charley.
Yes only a quadratic curve could ever be patentable, and probably is valid with those limitations, but one does not actually want the waveguide to be perpendicular at the throat so non-infringement is trivial.
Re: Re: Vertical coverage
Don't know if I'd agree about most people agreeing about this. Have you done a study? I've seen at least one study that states how benign the floor bounce is. It is something we are quite naturally used to, as it occurs from any source in a room, and it actually helps localize the source.
Wayne Parham said:
Don't know if I'd agree about most people agreeing about this. Have you done a study? I've seen at least one study that states how benign the floor bounce is. It is something we are quite naturally used to, as it occurs from any source in a room, and it actually helps localize the source.
"I've seen at least one study that states how benign the floor bounce is. It is something we are quite naturally used to, as it occurs from any source in a room, and it actually helps localize the source."
I was agreeing at first, but in the end I don't see how it's different than any other early reflection, i.e., what if the floor bounce from the original event doesn't arrive at the same as in your room?
Though it does seem that it would be more benign, as the range of floor bounce delays is much narrower than for reflections from other directions.
I was agreeing at first, but in the end I don't see how it's different than any other early reflection, i.e., what if the floor bounce from the original event doesn't arrive at the same as in your room?
Though it does seem that it would be more benign, as the range of floor bounce delays is much narrower than for reflections from other directions.
gedlee said:
Sorry, I can't put my finger on the actual paper, but if I was to start looking for it, my first guess would be some BBC research in an old Wireless World article. Maybe Queen or Katz? I'm thinking it was an article on localization. If I recall correctly, the floor bounce was said to enhance localization, which makes sense as in nature it would seem beneficial to use this. Think about being out in a canoe in the middle of a lake and hearing a sound on the shore. Yes, there's less wall reflection there, but the floor bounce off the water certainly doesn't detract from localizing the sound. We are accustomed to it. I don't know about anyone else, but I have NEVER heard anyone complain about floor bounce at a concert. I think the crux was that early horizontal reflection were more detrimental than early vertical reflections. Perhaps the shape of our ears is part of the factor.
In a similar article, if not the same one, I recall it was stated that we are not that attuned to comb filtering, which may be another issue with floor bounce. Again, the rational was that comb filtering happens all the time, and we ignore it as it serves no survival purpose to be attuned to it.
pooge said:
Vertical reflections have no affect on lateral localization and stereo can only do lateral localization. But the floor bounce being "helpful" seems implausible to me.
Also, there is no "floor bounce" is a concert in the same sense as with an audio system. There is an audience, which is highly absorptive, and the stage is a multitude of bounces, etc.
I would not put floor bounce at the top of my list of bad refelctions, but I have found that minimizing it is beneficial. Of all the reflections is is the most benign, but I would never describe it as "positive and should be accentuated". A good thick rug on the floor at the bounce works fine. The ceiling is less benign - we are not used to it.
There are two different issues, one a self-interference notch and the other an HF reflection. Both are undesirable, in my opinion, but the ways to reduce them are different because of the frequencies involved.
Whan I talk about "floor bounce", I'm not talking about an HF reflection. It's the self-cancellation notch found usually in the lower midrange. I do think ceiling reflections are a problem at HF, but usually carpeting and furniture attenuate the same sort of HF reflection off the floor. I've definitely heard it in houses with hardwood floors though, which can sound like a gymnasium until treated.
The self-interference "floor bounce" notch happens when speakers are mounted on stands or when towers are used. Whatever makes the midrange sound, if spaced off the ground a couple feet, there is a self-interference notch from the path length difference of the direct sound and the reflection off the floor. It is probably the biggest boundary related notch I see, and it almost always happens somewhere between 100Hz and 200Hz on tower speakers or monitors on stands. The frequency depends on the height of the speaker and the distance to the listener.
The "ceiling slap" at HF makes a sort of ringing, slapping sound. I notice it most on gabled ceilings but it can happen in any room. Naturally, the less HF energy directed upwards, the less reflected energy there will be.
Whan I talk about "floor bounce", I'm not talking about an HF reflection. It's the self-cancellation notch found usually in the lower midrange. I do think ceiling reflections are a problem at HF, but usually carpeting and furniture attenuate the same sort of HF reflection off the floor. I've definitely heard it in houses with hardwood floors though, which can sound like a gymnasium until treated.
The self-interference "floor bounce" notch happens when speakers are mounted on stands or when towers are used. Whatever makes the midrange sound, if spaced off the ground a couple feet, there is a self-interference notch from the path length difference of the direct sound and the reflection off the floor. It is probably the biggest boundary related notch I see, and it almost always happens somewhere between 100Hz and 200Hz on tower speakers or monitors on stands. The frequency depends on the height of the speaker and the distance to the listener.
The "ceiling slap" at HF makes a sort of ringing, slapping sound. I notice it most on gabled ceilings but it can happen in any room. Naturally, the less HF energy directed upwards, the less reflected energy there will be.