This has been discussed ad nauseum. The directivity of the WGs and the midwoofer lead to off-axis loss, which is balanced with pathlength loss to create a larger listening area with less change in tonality based upon lateral movement than with a "straight ahead" alignment. Additionally it reduces both amplitude and audibility of sidewall reflections, via pathlength loss and delay.
"pathlength loss"
Don't you mean pathlength gain? After all my point was that as you move closer to the toed in speaker laterally the total output in that direction decreases. But if the speaker were entirely constant directivity at all frequencies it wouldn't matter, the only thing that would change is how loud it is, not its FR. And that was my point, it's flaw based on your argument is that it isn't. I'd like to hear Dr. Geddes answer.
This is stricktly non-pro here and I assume the same. We are strickly talking about HI-Fi in small rooms, nothing else.
Last edited:
Its been discussed here at length (I don't know about "ad nauseum") and is discussed in my white papers on my web site. Badman has it right. You have got it wrong.
I had a long talk with Dave Clark about this. We both agreed that people get acklimated (SP?) to what they have or what they think that they want and that becomes the "standard" by which everything else is measured. That this "standard" is horrible is not relavent, it's what is now expected.
This is precisely why I let measurements tell me what "sounds good". I have not been let down by this approach yet.
I don't know why a simple example in plane geometry has created such a problem. Doesn't anyone learn it anymore? When you are midway between two speakers you are at the intersection of two sides and two hypoteneuse of two right triangles. As you move laterally say from the center to the right you are getting closer to the right speaker because the path length of the leg, that is the distance to the speaker where the two legs interset at a right angle is shorter than the hypoteneuse. Therefore the sound is getting louder because you are closer to it. But if the speaker is toed inward you are moving at an angle away from its axis so the sound directly from the CD driver is getting softer. The path length gain due to getting closer compensates for the angular loss due to the falloff. At this point I give up. You're all right. I resign. It's hopeless.
Yep, we are all right. Your explanation above means that you are right too! But you're misinterpreting what everyone else is saying, hence you're not seeing the equivalence. The pathlength attenuation of reflections is because the highest intensity axial energy is being aimed at the further boundary, and is another advantage to CD horns crossed in front of the "prime" position.
The change in SPL because of the pathlength difference is much smaller than the change due to the polar response. Insignificant in fact. One is a few % and the other a few dB.
And yes, I did take geometry - and then some.
And yes, I did take geometry - and then some.
I wonder how the multi-cell horns compare in this regard. How "CD" are they?
Somewhere around here I used to have the Altec lit on this, but can't find it now.
But if CD means a fixed response and level within an angular range, and then nothing...I wonder whether that would have drawbacks. Like making the beam more distinct and increasing the time taken to build a diffuse reverberant field.
http://www.xlrtechs.com/dbkeele.com/PDF/Keele (1975-05 AES Preprint) - Whats So Sacred Exp Horns.pdf
Page 9 and 10.
Exactly as I described in post #5499. Directivity is constant while the horizontal and vertical beamwidths have complimentary errors.
Constant directivity alone is not a complete discriptor of adequate performance.
David S.
CD does not mean that.
CD does not define a polar curve or even imply uniformity of polar curves, just that the area under the curves is constant (if you combine horizontal and vertical).
David S.
Speaker Dave's responses above and, to a large extent, many of the 5000+ posts on this thread, suggest to me there are those here who are seeking the 'single holy grail' of loudspeaker design that will summon some sort of audio nirvana.
Music is largely an art form; just like painting. It's accurate reproduction is largely, an engineering challenge. However, in the end, the appreciation of that art form again reverts to art appreciation wherein, the sonic beauty will be in the 'ear of the beholder'.
Music is largely an art form; just like painting. It's accurate reproduction is largely, an engineering challenge. However, in the end, the appreciation of that art form again reverts to art appreciation wherein, the sonic beauty will be in the 'ear of the beholder'.
Thanks David, I had not seen that paper. The DI looks similar to what I remember from the Altec lit.
Well, that's the long explanation, anyway! 🙂 If you've ever heard them, they simply sound too bright directly on axis. They are not designed to be used that way. A little toe-in does a world of good for the tonal balance. Used as directed, they are very good.
Yep. But you know there are people who will be so accustomed to having to fire tweeters directly at their ears due to the top octave beaming in cone'n'domes, who will try to set up properly balanced CD speakers that way.
Silly kids!
Cornucopia Revisited
Originally Posted by whgeiger
While reading the recent posts in this thread, some general comments regarding horn design come to mind:
1) Use of a horn, implies that the designer is willing to trade driver bandwidth for efficiency. To achieve this mission a horn constrains the driver’s radiation pattern only above the lower limit of its pass-band.
2) Here at this lower limit, driver unloading occurs while diaphragm excursion limits are rapidly approached as well. To avoid driver destruction and the introduction of excessive distortion in driver output, aggressive high pass filtering of the drive signal is required.
Facts:
2a) Operation of a compression driver at and below system resonance is contraindicated, due to the attendant increase in distortion products in this region.
2b) To reduce driver diaphragm displacement with a decline in frequency requires a high pass filter slope > 12 dB per octave.
2c) Driver loading (or the absence thereof) provided by a loudspeaker enclosure (horn included) is not a matter of mythology. In conical horns, as well as those asymptotically conical, the early decline of driver loading and the absence of a material [Fc] are well known and understood.
3) Under these conditions, a minimum of a 3-way partition of the audible spectrum is needed; e.g., 20-200, 200-2000, 2000-20,000 Hz. This fact is particularly relevant for an all-horn system.
3a) Bifurcation of the frequency range, where speech and most instrument fundamentals exist, is not particularly desirable.
3b) Below 100 Hz it is also know that signal direction cannot be detected by human ears. But in the crossover region above this point, that is not the case. So how the outputs of a satellite loudspeakers and sub-woofers combine is of equal importance as well.
3c) Above 10k Hz, signals from cymbals, triangles and other like instruments are audible and important. In this region, of course, is where signal beaming becomes most acute.
4) To mitigate beaming at the upper limit (determined almost entirely by the perimeter size and geometry of the effective piston radiating area), dispersive elements must be introduced such as an acoustic lens, passage bifurcation, or other beam spreading mechanisms. All drivers, horn loaded or not, exhibit high frequency beaming. This occurs when signal wave length approaches or becomes smaller than radiator dimensions.
Facts:
4a) The laws of acoustics dictate the beaming; unless of course, you mitigate it. In your case, this is achieved by introducing a foam acoustic lens and a geometric discontinuity in the horn neck. In the later case flare curvature increases rapidly from the compression driver exit and then steadily declines as a cone asymptote is approached. To mitigate the inevitable reflectance at the mouth, a toroidal surface of arbitrary dimensions is introduced. Here curvature abruptly increases again.
There are many ways to design a loudspeaker system. We differ in approach and as to what is important when doing that.
Regards,
WHG
Originally Posted by whgeiger
While reading the recent posts in this thread, some general comments regarding horn design come to mind:
1) Use of a horn, implies that the designer is willing to trade driver bandwidth for efficiency. To achieve this mission a horn constrains the driver’s radiation pattern only above the lower limit of its pass-band.
2) Here at this lower limit, driver unloading occurs while diaphragm excursion limits are rapidly approached as well. To avoid driver destruction and the introduction of excessive distortion in driver output, aggressive high pass filtering of the drive signal is required.
Facts:
2a) Operation of a compression driver at and below system resonance is contraindicated, due to the attendant increase in distortion products in this region.
2b) To reduce driver diaphragm displacement with a decline in frequency requires a high pass filter slope > 12 dB per octave.
2c) Driver loading (or the absence thereof) provided by a loudspeaker enclosure (horn included) is not a matter of mythology. In conical horns, as well as those asymptotically conical, the early decline of driver loading and the absence of a material [Fc] are well known and understood.
3) Under these conditions, a minimum of a 3-way partition of the audible spectrum is needed; e.g., 20-200, 200-2000, 2000-20,000 Hz. This fact is particularly relevant for an all-horn system.
Facts:
3a) Bifurcation of the frequency range, where speech and most instrument fundamentals exist, is not particularly desirable.
3b) Below 100 Hz it is also know that signal direction cannot be detected by human ears. But in the crossover region above this point, that is not the case. So how the outputs of a satellite loudspeakers and sub-woofers combine is of equal importance as well.
3c) Above 10k Hz, signals from cymbals, triangles and other like instruments are audible and important. In this region, of course, is where signal beaming becomes most acute.
4) To mitigate beaming at the upper limit (determined almost entirely by the perimeter size and geometry of the effective piston radiating area), dispersive elements must be introduced such as an acoustic lens, passage bifurcation, or other beam spreading mechanisms. All drivers, horn loaded or not, exhibit high frequency beaming. This occurs when signal wave length approaches or becomes smaller than radiator dimensions.
Facts:
4a) The laws of acoustics dictate the beaming; unless of course, you mitigate it. In your case, this is achieved by introducing a foam acoustic lens and a geometric discontinuity in the horn neck. In the later case flare curvature increases rapidly from the compression driver exit and then steadily declines as a cone asymptote is approached. To mitigate the inevitable reflectance at the mouth, a toroidal surface of arbitrary dimensions is introduced. Here curvature abruptly increases again.
There are many ways to design a loudspeaker system. We differ in approach and as to what is important when doing that.
Regards,
WHG
Last edited:
Fortunately I don't buy that. If you do, then my analytical approach to deisgn will not be saitisfying at all and I doubt that you will get much out of it.
Reproduction is completely independent of "art". A recording of a car should sound like a car - no art. A bad recording should sound bad - the art is bad. Judgement of the "art" is subjective, but reproduction is not.
Nope, I don't buy it.