Hello Noah,
Polars at various distances can be considered as subsets of the set of pressure field maps obtained at various frequencies. (For a given distance to the origin, the polar can be derived from the set of pressure fields at various frequencies ).
For waveguides delivering a pressure field looking like "leopard skin", polars defined at various distances will look different one of the other.
Best regards from Paris, France
Jean-Michel Le Cléac'h
Polars at various distances can be considered as subsets of the set of pressure field maps obtained at various frequencies. (For a given distance to the origin, the polar can be derived from the set of pressure fields at various frequencies ).
For waveguides delivering a pressure field looking like "leopard skin", polars defined at various distances will look different one of the other.
Best regards from Paris, France
Jean-Michel Le Cléac'h
noah katz said:
Hello Earl,
Many "non Le Cléac'h" horns using smooth profiles exist. Not all of them deliver smooth response, smooth pulse response, smooth polar, smooth pressure fields.
Smoothness of the profiles is not sufficient by itself in delivering smoothness of the results. Others characteristics are required one being a correct expansion law of the wavefronts area in order to obtain a very low reflection coefficient at all frequencies in the usable frequency range of the horn.
Best regards from Paris, France
Jean-Michel Le Cléac'h
Many "non Le Cléac'h" horns using smooth profiles exist. Not all of them deliver smooth response, smooth pulse response, smooth polar, smooth pressure fields.
Smoothness of the profiles is not sufficient by itself in delivering smoothness of the results. Others characteristics are required one being a correct expansion law of the wavefronts area in order to obtain a very low reflection coefficient at all frequencies in the usable frequency range of the horn.
Best regards from Paris, France
Jean-Michel Le Cléac'h
gedlee said:
Jmmlc said:
Jean-Michel
Nothing new that we don't agree on the physics of horns and waveguides. You state things as fact that I find completely untenable.
And for what its worth I have no idea what you were talking about in your polar field response answer to Noah. Maybe he understood, but I don't.
Hello Earl,
I cannot believe one second that you could have problems in deriving polars from soundfields maps. ().
For Noah and the others:
consider a set of 4 pressure fields maps like on:
http://www.diyaudio.com/forums/attachment.php?s=&postid=1762155&stamp=1236070764
(eventually you may want to CD equalize in the HF)
we can draw on those pressure maps circles at radius from the mouth for which we would want to have polars.
Angles ticks may be drawn on those circles symetrically to the axis.By example 5°, 10°, 15°, 20°, etc. (and their negative symetrical)
For those angles and for those 4 frequency, 1000Hz, 2400Hz, 5600Hz and 8000Hz we can retrieve the acoustic SPL.
Then we are able to derive a polar (SPL for coordinates frequency/angle) for a given radius.
Here we have only 4 frequencies but we can have more.
I don't have myself pressure fields covering large area around the OS waveguide mouth, so I used the previous graphs and traced 2 circles with radius R1 an R2 (I know this doesn't cover large radius but it is just for the purpose of the explanation why 2 polars may differ if taken at 2 different distances).
See attached file.
The variation of SPL along those circles differ a lot between R1 and R2. We can see that at 2400Hz and 5600 Hz the variations of SPL are very large for the radius R2 and smaller for radius R1.
This means that the polars at R1 and at R2 will differ a lot too in their smoothness / roughness.
As you can see as radius enlarges, the number of lobes reduces but the SPL variation for each one is larger. I don't know how it will be at 3 meters. I doubt that the roughness will be totally supressed.
Best regards from Paris, France
Jean-Michel Le Cléac'h
I cannot believe one second that you could have problems in deriving polars from soundfields maps. ().
For Noah and the others:
consider a set of 4 pressure fields maps like on:
http://www.diyaudio.com/forums/attachment.php?s=&postid=1762155&stamp=1236070764
(eventually you may want to CD equalize in the HF)
we can draw on those pressure maps circles at radius from the mouth for which we would want to have polars.
Angles ticks may be drawn on those circles symetrically to the axis.By example 5°, 10°, 15°, 20°, etc. (and their negative symetrical)
For those angles and for those 4 frequency, 1000Hz, 2400Hz, 5600Hz and 8000Hz we can retrieve the acoustic SPL.
Then we are able to derive a polar (SPL for coordinates frequency/angle) for a given radius.
Here we have only 4 frequencies but we can have more.
I don't have myself pressure fields covering large area around the OS waveguide mouth, so I used the previous graphs and traced 2 circles with radius R1 an R2 (I know this doesn't cover large radius but it is just for the purpose of the explanation why 2 polars may differ if taken at 2 different distances).
See attached file.
The variation of SPL along those circles differ a lot between R1 and R2. We can see that at 2400Hz and 5600 Hz the variations of SPL are very large for the radius R2 and smaller for radius R1.
This means that the polars at R1 and at R2 will differ a lot too in their smoothness / roughness.
As you can see as radius enlarges, the number of lobes reduces but the SPL variation for each one is larger. I don't know how it will be at 3 meters. I doubt that the roughness will be totally supressed.
Best regards from Paris, France
Jean-Michel Le Cléac'h
gedlee said:
Attachments
Jean-Michel
Again, you don't listen very well. Those are all nearfield responses and completely irrelavent as no one sits that close to the sources. I made this point the first time that you showed them. The further point that was made was that as one gets into the far field, and 3 meters is certainly in the far field, then there will no longer be any variation on the polar response with distance. You seemed to disagree with that and I could never figure out what it was that you were saying. Now you are once again back to the same pointless nearfield plots and stating again the same misrepresentation that you stated before.
Again, you don't listen very well. Those are all nearfield responses and completely irrelavent as no one sits that close to the sources. I made this point the first time that you showed them. The further point that was made was that as one gets into the far field, and 3 meters is certainly in the far field, then there will no longer be any variation on the polar response with distance. You seemed to disagree with that and I could never figure out what it was that you were saying. Now you are once again back to the same pointless nearfield plots and stating again the same misrepresentation that you stated before.
Hi,
why don't you all post detailed measurements from different distances? Let's say 50cm, 1m and 2m (+/- 90º)?
Best, Markus
why don't you all post detailed measurements from different distances? Let's say 50cm, 1m and 2m (+/- 90º)?
Best, Markus
This kind of data is useful when you look at all three sets of data at the same time. E.g., it would make it extremely easy to determine a speakers appropriate application.
Best, Markus
Best, Markus
Markus
Sorry, but I don't see that. Nobody is going to sit at .5 m from speakers with horns or waveguides and even 1 m is not feasible. By 2 m you are pretty much in the far field so let's just stick with that.
Sorry, but I don't see that. Nobody is going to sit at .5 m from speakers with horns or waveguides and even 1 m is not feasible. By 2 m you are pretty much in the far field so let's just stick with that.
I'm talking about a more universal approach to comparable data for all kind of loudspeakers and not only horns. Having data from 0.5 and/or 1m would make it very clear what the application of a certain speaker is.
By the way, what do you think–how many people here in this forum do know that your speakers don't "work" closer than ≈2m? What's wrong with educating people?
Best, Markus
By the way, what do you think–how many people here in this forum do know that your speakers don't "work" closer than ≈2m? What's wrong with educating people?
Best, Markus
Markus
As I explained to you before, its not a limitation of "my" speakers but toed in speakers in general. If the speakers are about 6 feet apart and they are toed in at 45 degrees and one sits behind the crossing point, then this is going to be at back at least 4 feet or more - no matter who's speakers we are talking about. So there is no point in looking at data any closer than this.
"Having data from 0.5 and/or 1m would make it very clear what the application of a certain speaker is."
I don't see any validity to this statement. Its the far field that matters, not the near field. And the near field does not tell you a thing about the speakers "application". They are not headphones!
As I explained to you before, its not a limitation of "my" speakers but toed in speakers in general. If the speakers are about 6 feet apart and they are toed in at 45 degrees and one sits behind the crossing point, then this is going to be at back at least 4 feet or more - no matter who's speakers we are talking about. So there is no point in looking at data any closer than this.
"Having data from 0.5 and/or 1m would make it very clear what the application of a certain speaker is."
I don't see any validity to this statement. Its the far field that matters, not the near field. And the near field does not tell you a thing about the speakers "application". They are not headphones!
For you the near field doesn't matter but there are people that (need to) put their speakers on a meter bridge as close as 1m or even closer. Again, having balloon data (http://www.four-audio.de/en/products/elf.html) would only help people in choosing the right speaker for their application.
And why do you assume that everybody has their speakers 6 feet apart–wishful thinking? From my point of view everything from 1m to 3m distance is most interesting. I don't understand why anybody would argue against having data from exact that range.
Best, Markus
And why do you assume that everybody has their speakers 6 feet apart–wishful thinking? From my point of view everything from 1m to 3m distance is most interesting. I don't understand why anybody would argue against having data from exact that range.
Best, Markus
Huh?
I don't get this argument. Dr. Geddes has clearly stated that this sort of data is irrelevant for the applications he deems desirable for his speaker designs.
If you want data for non standard and/or sub-par listening locations, maybe generate it and present it, if that's important to you. But to argue with the designer as to why his design limitations aren't valid in your mind... I just don't get it. Sounds a bit pedantic.
That is, unless you like listening to unavoidable artifacts and skewed presentations from speakers never meant for the kind of use you envision, ...and need some sort of ammo to defend this usage
John L.
markus76 said:
I don't get this argument. Dr. Geddes has clearly stated that this sort of data is irrelevant for the applications he deems desirable for his speaker designs.
If you want data for non standard and/or sub-par listening locations, maybe generate it and present it, if that's important to you. But to argue with the designer as to why his design limitations aren't valid in your mind... I just don't get it. Sounds a bit pedantic.
That is, unless you like listening to unavoidable artifacts and skewed presentations from speakers never meant for the kind of use you envision, ...and need some sort of ammo to defend this usage
John L.
John
Looking at the wearisome discussion between Earl and Jean-Michel I just asked for meaningful data that might help bringing the discussion back to a more objective level. One can argue how that data might look like. But you can't argue that there's no need for that kind of data.
Best, Markus
P.S. Obviously you missed it but I've set up my Nathan's exactly the way recommended by Earl.
Looking at the wearisome discussion between Earl and Jean-Michel I just asked for meaningful data that might help bringing the discussion back to a more objective level. One can argue how that data might look like. But you can't argue that there's no need for that kind of data.
Best, Markus
P.S. Obviously you missed it but I've set up my Nathan's exactly the way recommended by Earl.
Markus
I would just like to see any real data on one of Jean-Michels horns instead of simulations. To the best of my knowledge we haven't seen any, let alone at .5 m, 1 m AND 2 m! Some axial stuff and in the mouth, but nothing "meaningful".
I would just like to see any real data on one of Jean-Michels horns instead of simulations. To the best of my knowledge we haven't seen any, let alone at .5 m, 1 m AND 2 m! Some axial stuff and in the mouth, but nothing "meaningful".
Earl
That's a reasonable claim. Hope you both can agree on a comparable set of data.
Best, Markus
That's a reasonable claim. Hope you both can agree on a comparable set of data.
Best, Markus
I think the sims that Jean-Michel presented about the OS wave guide would probably be questionable if the driver section did not include phase plug model. How this and the wave guide matches seems very critical from what I have seen in my sims. What is really necessary is that compression driver needs to be modelled with the diaphragm and the phase plug dimensions in place.
Ever since the debate started, there have been many people asking for others to present data without spending the effort to produce some of there own additional data. If this were the the Vendors related section where sales is involved, I can understand this. But this is not that kind of section. Maybe we should split this into Technical and Listening threads like the EnABL thread?😀
Ever since the debate started, there have been many people asking for others to present data without spending the effort to produce some of there own additional data. If this were the the Vendors related section where sales is involved, I can understand this. But this is not that kind of section. Maybe we should split this into Technical and Listening threads like the EnABL thread?😀
soongsc
Every data from simulations is questionable as long as it's not verified by real world data. I hope Earl and Jean-Michel post comparable data. Otherwise the whole discussion is academic and of no other use than intensifying beliefs instead of knowledge.
Best, Markus
Every data from simulations is questionable as long as it's not verified by real world data. I hope Earl and Jean-Michel post comparable data. Otherwise the whole discussion is academic and of no other use than intensifying beliefs instead of knowledge.
Best, Markus
Marcus
As a comparison you are correct, but I have compared my data - which I DO post - with the sims and it IS mostly correct. There are the expected deviations due to the non-ideal driver, but on the whole my theory and the measurements are completely consistant. So one of us has done the job right.
As a comparison you are correct, but I have compared my data - which I DO post - with the sims and it IS mostly correct. There are the expected deviations due to the non-ideal driver, but on the whole my theory and the measurements are completely consistant. So one of us has done the job right.