Thanks all, I think I need to learn more about what i'm doing with my software. I think it will come with using t more and learning more of what it's really doing and how to interput it better..
Earl, your software sounds very interesting but is probably (surely) too complicated for me at this point.
Earl, your software sounds very interesting but is probably (surely) too complicated for me at this point.
Yes, unfortunately ease of use was never a design goal.
It uses MathCAD as it calculation engine which is very flexible, but you do have to have a feel for that program to use my techniques. But you would not have to actually write anything in MathCAD, it would run even without doing anything. I don't know if a run-only version of MathCAD is still available or not.
I also use some external plotting packages to get publication quality graphics because MathCADs are not so great.
At any rate, doesn't sound like much interest in measurements.
It uses MathCAD as it calculation engine which is very flexible, but you do have to have a feel for that program to use my techniques. But you would not have to actually write anything in MathCAD, it would run even without doing anything. I don't know if a run-only version of MathCAD is still available or not.
I also use some external plotting packages to get publication quality graphics because MathCADs are not so great.
At any rate, doesn't sound like much interest in measurements.
Just looked at ARTA - a superb setup. I really liked the manual where the precise details of the implimentation were shown. This is rare. Iw ill certainly have to get this package as it does what I need (I currently use SpectraLab, which is much more expensive).
Thanks for showing this program to me!
Thanks for showing this program to me!
I'm glad you can use it. It is a bit over whelming for me at this point.
Another useful tool is the ARTA Steps program and SynRTA Freeware from Liberty instruments. A fine RTA with high resolution and it's FREE!
Another useful tool is the ARTA Steps program and SynRTA Freeware from Liberty instruments. A fine RTA with high resolution and it's FREE!
I would caution on the use of Real-time measurements because interpretation can be so difficult. They are easy to use but even easier to misinterprete. Basically, I don't do anything in real time. We simply don't hear like an instrument and unless some attempt is made to make the data more representative it tends to be pretty meaningless.
The common trend of EQing to a real time display is a big mistake IMO. Except at very LF, where this does work if you move the mic around by hand - i.e. spatial averaging. But this procedure makes the higher frequencies completely invalid. Its unfortunate, but its just not that easy to get good measurements.
The common trend of EQing to a real time display is a big mistake IMO. Except at very LF, where this does work if you move the mic around by hand - i.e. spatial averaging. But this procedure makes the higher frequencies completely invalid. Its unfortunate, but its just not that easy to get good measurements.
What plotting tool do you use? I've still been looking for a good one for both personal and work use. I use Matlab (for analysis, but graphics too) and Excel (powerpoint friendly graphics, but the graphics suck, big time).
I used to use Axum - not available anymore. Now I use Origin. Its better than Axum, but a lot more expensive. MatLab was supposed to have good graphics, I hear that Mathematica is very good too. MathCAD, which I like alot, does not have very good graphics. But the algebraic nature of MathCAD makes it much more usable for me than MatLab.
Matlab is my preferred tool so far, but its only pretty ok in graphics IMO. Splus is a bit better and there is R which is an open source (free) version.
gedlee said:
Hi Dr. Geddes,
I have a fair amount of interest, so I went ahead and started a new topic. I hope you will chip in with a 'nugget' or two. I for one would much appreciate it.
Regards,
John
Thread on Measuring Loudspeakers
Wow I just read this whole thread from the beginning, my eyes are starting to hurt. . . haha-
I'm not sure if anyone is still reading this; but one thing I would love to know is this; is the ideal wavefront for an OS waveguide spherical, or planer? Or at the point where it enters the throat are these essentially interchangeable?
My other comment would be regarding musicians and playback systems. As a professional musician (and engineer/producer) I will argue that Dr. Geddes is correct (which seems to be a reoccurring theme in this thread). The musician side of me does not really care about the playback system, I'm too much distracted by the interesting sound of ending the song with a IVdom7#11. For me personally I can kinda switch from one mode to another, but only after investing a considerable amount if time WRT sound recording and reproduction. Most musicians have not done this.
For an accomplished musician a piece of sheet music is a sort of playback system (no speakers needed).
I also think it is interesting to see the perceptions and misunderstandings that many people here have regarding the (non-classical) recording process, but none of this has anything to do with waveguides so I'll stop there.
I'm not sure if anyone is still reading this; but one thing I would love to know is this; is the ideal wavefront for an OS waveguide spherical, or planer? Or at the point where it enters the throat are these essentially interchangeable?
My other comment would be regarding musicians and playback systems. As a professional musician (and engineer/producer) I will argue that Dr. Geddes is correct (which seems to be a reoccurring theme in this thread). The musician side of me does not really care about the playback system, I'm too much distracted by the interesting sound of ending the song with a IVdom7#11. For me personally I can kinda switch from one mode to another, but only after investing a considerable amount if time WRT sound recording and reproduction. Most musicians have not done this.
For an accomplished musician a piece of sheet music is a sort of playback system (no speakers needed).
I also think it is interesting to see the perceptions and misunderstandings that many people here have regarding the (non-classical) recording process, but none of this has anything to do with waveguides so I'll stop there.
RyanC said:
No the wavefronts are most certainly not interchangable. The OS waveguide admits any wavefront - of course everything does - but it generates the least HOM for a flat wavefront. I can find the waveguide that generates the least HOM for any wavefront shape, but they are all different. Since most compression drivers "claim" or are designed to provide a flat wavefront the OS is the optimum solution. But, as I said, there is a solution for any situation.
I deal with musicians all the time and your comment is very true. Producers often "get" the difference between making music and making a recording. Not everyone appreciates this.
OK-
I guess I'm having a difficult time understanding the difference between planar and flat. I gather though that there would be a difference between the wavefront from a compression driver and that from a ribbon driver? The problem being the flex in the diaphragm? Would this still be the case for a large ribbon (like a B&G RD50)?
One plus to this is that the horn is automatically non-axisymetric. The downside being the beaming at HF, which also would seem to have a positive implication of less refraction within the horn.
I suppose I should not ask you to speculate though. . . Thanks Earl
I guess I'm having a difficult time understanding the difference between planar and flat. I gather though that there would be a difference between the wavefront from a compression driver and that from a ribbon driver? The problem being the flex in the diaphragm? Would this still be the case for a large ribbon (like a B&G RD50)?
One plus to this is that the horn is automatically non-axisymetric. The downside being the beaming at HF, which also would seem to have a positive implication of less refraction within the horn.
I suppose I should not ask you to speculate though. . . Thanks Earl
RyanC said:
We tend to think of planar and flat as the same thing, so I guess they are. But I sometimes think of a plane as being a shhet of constant coordinate and so for me flatt is the more appropriate. But yes, in this discussion they are the same thing.
A ribbon is a flat source, but its not round, so OS doesn't apply. There are minimum diffraction waveguides which can be made for a flat rectangle. Why is non-axisymetric a plus? I always thought the opposite. I agree that the beaming is a downside, but in a proper waveguide even the ribbon would not beam.
Right I know that it isn't round. I'm thinking of using a line 32" tall in the WG.
I was thinking of following the OS profile in a square. I uderstand that it may not be ideal, but if it were oblong, like if you were to take your WG and stretch it out so that the throat was 2" wide and 32" tall, this should be close to optimal no? Obviously the throat would need to line up directly to the radiation holes on the driver, which are not square (they are rounded). This is however far beyond the ability of google sketchup. . haha
So for now I'll play with the concept of a square one.
Like this, this one shows only the horizontal axis, and is only a crude approximation of the OS profile, just for the purpose of entertaining the concept.
FWIW, I have built some test horns that are 50x30, one foot deep, and they are sounding very good with 2 B&G Neo 8s in each. I want a larger vertical coverage area than the 2 will offer so I'm looking at 4 per side.
Testing and listening to the ones that I have built, it seems that the loading (or guiding) is mostly on the horizontal axis, but I need to run some tests to confirm that.
Thanks-
I was thinking of following the OS profile in a square. I uderstand that it may not be ideal, but if it were oblong, like if you were to take your WG and stretch it out so that the throat was 2" wide and 32" tall, this should be close to optimal no? Obviously the throat would need to line up directly to the radiation holes on the driver, which are not square (they are rounded). This is however far beyond the ability of google sketchup. . haha
So for now I'll play with the concept of a square one.
Like this, this one shows only the horizontal axis, and is only a crude approximation of the OS profile, just for the purpose of entertaining the concept.
FWIW, I have built some test horns that are 50x30, one foot deep, and they are sounding very good with 2 B&G Neo 8s in each. I want a larger vertical coverage area than the 2 will offer so I'm looking at 4 per side.
Testing and listening to the ones that I have built, it seems that the loading (or guiding) is mostly on the horizontal axis, but I need to run some tests to confirm that.
Thanks-
Attachments
RyanC said:
Yes, what you have drawn will only control in the horizontal plane, the vertical polars will be completely uncontrolled. It will be very difficult to do any form of directivity control for a line source in the plane of its length. This is just fundamental physics.
You need to start with a fairly compact source to get good directivity control and once it starts to get large in any dimension then control in that dimension begins to degrade. The best control therefor comes from a elliptical aperature of the desired area. Circular is second best, best for axisymmetric. This is why I don't use a line source. This is also why I use a 1" throat instead of larger. The smaller the throat the higher the HF control can be maintained. A 1" can just get to 16 kHz or so, a 2" about 1/2 of this. The 2" necessitaes a crossover somewhere between 2 - 8 kHz. A very bad place to put a crossover IMO. There is simply no way to get one that is not highly audible in a critical range.
RyanC said:
Doug Kelly has met Dr Geddes, and was messing around with using B&G drivers. He gave up on the project and bought a Yorkville Unity.
I met Dr Geddes the same day, and messed around with OS waveguides in a Unity clone. I gave up on the project also. It sounded great but looked like hell. I just picked up my Summas from Fed Ex.
I hate to be a wet blanket, but seriously guys, stick with compression drivers.
Hey Patrick
Yeah I've been speaking with Doug, he is actually in my neck of the woods. I believe he was actually experimenting with some monsoon drivers.
I don't understand why some people here were so insistent on attempting to use a dome tweet, that offers no benefits and only drawbacks when compared to a compression driver (especially in a WG). But a ribbon is a different animal. Going 2+ octaves lower and having a near ideal CSD are both significant advantages, that I believe when properly dealt with in a WG have the potential to perform at a very high level. (I need to experiment more with tests though)
From what I understand, a compression driver is a dome, that has a compression chamber and a phase plug, no? My feeling, and analysis of CSDs and listening experience, is that ribbons (especially like the B&G ones) are simply a better transducer then a dome/voice coil. Furthermore I find the concept of a high functioning compression chamber/phase plug, to be made in low weights to be very problematic (sound is a bit more persistent than a couple small pieces of plastic). Not that they don't do what they are suppose to do to some extent, just that they have their issues as much as anything else does.
2 things I would say, with the caveat that I know much less than many people here and I am trying to learn as much as possible.
1)I'm not trying to load a waveguide or horn with the wrong type of wavefront. . .
2)The current test horns that I built are working surprisingly well, I believe this is because the horizontal axis is only loading at roughly the point where the driver becomes more directional, and the vertical axis is loading even lower, which is compensating for the tilted natural response of the drivers. Due to the directionality of the drivers, there is also much less HF bouncing around inside the horn. So they are only loading below ~4k, and then again more loading at about 700hz, I think if I move this up to 6k and down to 450hz, with smoother transition to the horn wall in a 4ft deep horn this could work very well, but that needs to be tested. But first I need to test the off axis HF roll off of the driver in an IB or large box so that I can know what I am trying to match it with.
The one thing that I'm still struggling to understand is acoustic overload of a horn, vs this not being a problem in a WG? Other than a depth relative to more than an oct above the quarter wave, where can I learn more about this? How can it be measured or what does it sound like (playing with low XO points and higher spls i think i know but. . .).
Thanks guys, and I hope the tone does not come across as aggressive here, I can assure you that it would not at all if this conversation was in person-
Yeah I've been speaking with Doug, he is actually in my neck of the woods. I believe he was actually experimenting with some monsoon drivers.
I don't understand why some people here were so insistent on attempting to use a dome tweet, that offers no benefits and only drawbacks when compared to a compression driver (especially in a WG). But a ribbon is a different animal. Going 2+ octaves lower and having a near ideal CSD are both significant advantages, that I believe when properly dealt with in a WG have the potential to perform at a very high level. (I need to experiment more with tests though)
From what I understand, a compression driver is a dome, that has a compression chamber and a phase plug, no? My feeling, and analysis of CSDs and listening experience, is that ribbons (especially like the B&G ones) are simply a better transducer then a dome/voice coil. Furthermore I find the concept of a high functioning compression chamber/phase plug, to be made in low weights to be very problematic (sound is a bit more persistent than a couple small pieces of plastic). Not that they don't do what they are suppose to do to some extent, just that they have their issues as much as anything else does.
2 things I would say, with the caveat that I know much less than many people here and I am trying to learn as much as possible.
1)I'm not trying to load a waveguide or horn with the wrong type of wavefront. . .
2)The current test horns that I built are working surprisingly well, I believe this is because the horizontal axis is only loading at roughly the point where the driver becomes more directional, and the vertical axis is loading even lower, which is compensating for the tilted natural response of the drivers. Due to the directionality of the drivers, there is also much less HF bouncing around inside the horn. So they are only loading below ~4k, and then again more loading at about 700hz, I think if I move this up to 6k and down to 450hz, with smoother transition to the horn wall in a 4ft deep horn this could work very well, but that needs to be tested. But first I need to test the off axis HF roll off of the driver in an IB or large box so that I can know what I am trying to match it with.
The one thing that I'm still struggling to understand is acoustic overload of a horn, vs this not being a problem in a WG? Other than a depth relative to more than an oct above the quarter wave, where can I learn more about this? How can it be measured or what does it sound like (playing with low XO points and higher spls i think i know but. . .).
Thanks guys, and I hope the tone does not come across as aggressive here, I can assure you that it would not at all if this conversation was in person-
RyanC said:
First, I don't agree that a ribbon is a better source than a compression driver. It is very low efficiency which is a major drawback for the ribbon. And as I said in another post a circular source is the better choice than anything with extension - a circle is the most compact source with a given area.
CSD is way overblown in terms of its telling us anything about audibility, and I don't see where a ribbon on a horn has any better CSD that a compression driver. You'd have to have some real data to back up your claims as they just aren't consistant with my experince in either measurements or sound quality.
The only studies of acoustic overload in a horn are the ones that Lidia and I did some years back. We shown how the diffraction in a orn was level dependent whose audibility increased with level making the diffraction the single most likely reason for the apparent "overload" phenomina in a horn. That study is posted on my web site.