Home » Audio » Speaker » Distortion mechanisms
I checked up Floyd Toole's work [message #17600 is a reply to message #17597] Mon, 28 February 2005 11:12 Go to previous messageGo to next message
akhilesh is currently offline  akhilesh
Messages: 1275
Registered: May 2009
Illuminati (3rd Degree)
and you are right: they found that listeners favor constant directivity as well.

-akhilesh


Hey, I just saw your website [message #17601 is a reply to message #17587] Mon, 28 February 2005 11:24 Go to previous messageGo to next message
akhilesh is currently offline  akhilesh
Messages: 1275
Registered: May 2009
Illuminati (3rd Degree)
Earl,
Very impressive work! I really hope you come for the Tulsa audio fest in a month or two, since I am very interested in hearing your speakers.
-akhilesh
(PS i myself am just a hobbyist who has been dabbling in this for about 2.5 years)

Re: So you are saying [message #17602 is a reply to message #17599] Mon, 28 February 2005 11:29 Go to previous messageGo to next message
Earl Geddes is currently offline  Earl Geddes
Messages: 220
Registered: May 2009
Master
I have done some tests on the BMS drivers and don't find that they actually do have much more HF content. In the coaxial drivers the response at the crossover is a mess, and this is in a critical band for audibility.

I have looked at just about every driver ot there and B&C is my choice. I am looking more seriuosly at Beyma, but BMS is not in the running.

Re: So you are saying [message #17603 is a reply to message #17598] Mon, 28 February 2005 11:36 Go to previous messageGo to next message
Earl Geddes is currently offline  Earl Geddes
Messages: 220
Registered: May 2009
Master
First I ballpark the design by calculating what a 15" piston directivity would be and what a waveguide should be to match at 90°. Then I set up the system and measure the complete polar response of both drivers in the actual enclosure. Finally a crossover is designed and optimized with a computer program that I wrote to give the flatest response over all angles in the forward 90° arc, with a slight preference for 22.5° (the direct angle).

Finally, of course, the crossovers and systems are tested at all polar angles (these results are shown on my web site). I have never seen a situation where the measurements and computer predictions were not the same, so this last step is a simple confirmation.


Re: I checked up Floyd Toole's work [message #17604 is a reply to message #17600] Mon, 28 February 2005 11:42 Go to previous messageGo to next message
Earl Geddes is currently offline  Earl Geddes
Messages: 220
Registered: May 2009
Master
So then you have to ask yourself why there are so few constant directivity designs available. Maybe its because they are hrd to do?

Please keep in mind that omni IS constant directivity. Its constant high directivity that is the key.

If you have not heard a good narrow constant directivity loudspeaker then do yourself a favor and hear them. You won't like anythiing else again.

I would suggest mine.

DI matching [message #17606 is a reply to message #17598] Mon, 28 February 2005 13:46 Go to previous messageGo to next message
Wayne Parham is currently offline  Wayne Parham
Messages: 18680
Registered: January 2001
Illuminati (33rd Degree)

This is called DI matching, and it is a good practice, in my opinion. You run the midwoofer up through the vocal range which does two things: It allows the voice fundamentals to be handled by a single driver and sets the crossover point in the overtone range, which is a natural sounding place to split, in my opinion. It also matches the directivity of the midwoofer to that of the horn tweeter. A 90x40 horn has DI of about 12, and that's what a 12" or 15" speaker has between 1kHz and 2kHz. It is collapsing past 10 approximately where wavlength equals diameter. See chapter 3 of Augspurger's JBL Sound System Design Manual for more information.


CD versus H/LD [message #17607 is a reply to message #17604] Mon, 28 February 2005 14:32 Go to previous messageGo to next message
akhilesh is currently offline  akhilesh
Messages: 1275
Registered: May 2009
Illuminati (3rd Degree)
Earl, I hope you do come to Tulsa for the spring Audio fest, becuase I would love to hear your designs!

I did some more thinking on this, and i agree: i was confusing constant dir (CD) (varies with frequency) and high versus low directivity (h/LD) (varies with degree) , as you said in your post.

On re-reading Floyd's white paper, it seems he is saying one wants CD, but says nothing about LD versus HD. Clearly, HD (along with CD) will mean that the primary sound will dominate the musical energy, as opposed to first order and second order reflections. In other words the waterfall plot will slope much more with HD.

Do you have any studies at GEDLEE that have established, in double blind tests, that HD is preferable to LD, CD being held constant? Intuitively I agree with you: HD will make the room sound "deader" than LD.

Hope to see you in tulsa
-akhilesh
PS: Do you offer your speakers for sale or on a trial basis?

Thanks Wayne & Earl [message #17608 is a reply to message #17603] Mon, 28 February 2005 14:34 Go to previous messageGo to next message
akhilesh is currently offline  akhilesh
Messages: 1275
Registered: May 2009
Illuminati (3rd Degree)
Very enlightening.
-akhilesh

Re: CD versus H/LD [message #17610 is a reply to message #17607] Mon, 28 February 2005 14:54 Go to previous messageGo to next message
Earl Geddes is currently offline  Earl Geddes
Messages: 220
Registered: May 2009
Master
No - we only did studies of HD CD comparisons with different speakers.

But here is the thing to consider with HD. With HD I can point the speakers such that the early reflections are minimizes and so that the first reflections come to the ear opposite the first arrival. This is very important since a first arrival and an early refelction at the same ear will cause coloration. This is point one.

Now also consider the energy time arrival of an HD versus a LD. The LD reflections arrive almsot instantaneously there is not gap between the first arrial and the early reflections. Not so the HD. There is a significant gap between the first arrival and the early reflections. I cannot overstate how important this later aspect is to imaging. Without a gap in the arrival times of the direct sound and the early reflectiosn the ear cannot resolve spatial location cues and the imaging will be poor.

The imaging on my speakers is as precise as headphones but without the "in the head" image problem. Pinpoint locailzation in a space beyond the actual location of the speakers. It really has to be heard to be believed.


Re: CD versus H/LD [message #17611 is a reply to message #17610] Mon, 28 February 2005 15:49 Go to previous messageGo to previous message
akhilesh is currently offline  akhilesh
Messages: 1275
Registered: May 2009
Illuminati (3rd Degree)
Earl,
Thanks for your reply. I agree with you...intuitively it makes total sense: HD will lead to better imaging, and a cleaner sound.
Since you have the setup & the skills, i was speculating that it would be interesting to measure, in a double blind test, if (educated) listeners (a la Floyd Toole's methodology) actually do prefer HD, and by how much (keeping CD & the rest of the room & system constant of course).
Speculating further, we'll have to figure out how to develop 2 speaker systems, one with HD & one with LD, with CD constant. ALternately, one could use the same speaker system, but in one experimental tresatment emphasize room reflections so they are stronger and reach the listener sooner. Should be interesting & fun!
Not to mention a good potential AES paper!
-akhilesh


Previous Topic: Vott 210 and slot cars
Next Topic: Pro Sound Driver Sensitivity
Goto Forum:
  


Current Time: Mon Apr 29 01:14:31 CDT 2024

Sponsoring Organizations

DIY Audio Projects
DIY Audio Projects
OddWatt Audio
OddWatt Audio
Pi Speakers
Pi Speakers
Prosound Shootout
Prosound Shootout
Smith & Larson Audio
Smith & Larson Audio
Tubes For Amps
TubesForAmps.com

Lone Star Audiofest