Edgar

Roy, I know that I am responding to quite an old post, but I've only just read it for the first time today when the thread was updated with a new response. Please understand that I'm an engineer, but my specialty is signal processing not acoustics, so my understanding of acoustic waveguide theory is not as complete as I'd like. That said, don't acoustic waves propagate according to Huygen's Principle? If so, then the curvature of the wave front can intuitively be explained by differences in path length. If the portion of the sound wave adjacent to the wall propagates along the wall, while the portion of the sound wave in the "center" of the horn propagates straight down the center of the horn, and both propagate at the speed of sound, then the wave front has to curve. Maybe the easiest way to visualize this is to think of a point source in a flat baffle as a degenerate horn with instantaneous 90° flare. The wavefront will be hemispherical, not planar. BTW; the path length "L" (along the wall) of a circular Tractrix horn is easily computed from the Tractrix defining equation as: L=a*ln(r2/r1), where a is the mouth radius and r1, r2 are radii between which you want to know the path length (this corresponds to the nomenclature used by Bruce Edgar in his 1981 Speaker Builder article). So the path length (along the wall) of an entire Tractrix horn is L=a*ln(a/r1), where r1in this case is the throat radius. From the defining equation for the Tractrix it can easily be shown that the path length is much longer than the horn itself, meaning that the portion of the sound wave that propagated down the "center" of the horn exits the mouth long before the portion that hugged the wall even reaches the mouth. Hence, the "bubble" model of horn propagation. Greg

The only CAD program with which I have any experience is TurboCAD. It handles 2D and 3D, it does everything I need for speaker design (and a lot of other things), and it's inexpensive at about US$150. Older versions are readily available on ebay for far less than that. (I'm currently using v7.1; I just checked ebay and the oldest that I found was v9.1 for under US$20, including shipping!) And it can import and export DXF files. Greg

No kidding about Brrrrmont. I'm measuring about 18" in the Williston-Bolton snow corridor this morning; fallen since Friday. Supposed to drop to -10°F tonight.

Just in case anyone's interested, here's an example of the "Elongated Tractrix" contour that I mentioned earlier. In this case I have set its length equal to that of an exponential horn with the same cutoff frequency. As you can see, the Elongated Tractrix acts like a hyperbolic horn over the majority of its length (good low frequency loading), and then expands like a Tractrix near the mouth. I have not done ANY analysis of this contour other than to modify the Tractrix equation to be able to generate it. Greg

My experience with trying to design a 25 Hz horn has been ... disappointing. I have concluded that one would be better off using the volume that such a horn would require as the rear enclosure in an infinite baffle, along with a multitude of 18" woofers and lots and lots of Watts. (Dang it; how do I insert a clickable URL?) http://ibsubwoofers.proboards51.com/index.cgi Greg

Got it. Looks like he revised the code without revising the revision number.

I suspect that the product number is the date of download.

Mine is missing the distortion field. Here's the version info:

I just tried that. It offers SPL, electrical input power, acoustical output power, and system efficiency. Unfortunately no distortion.

Most woofers won't go to 2K. Using Keele's equations, I find that the best 12" are lucky to get to 500Hz before mass-rolloff, and the best 10" only go to ~800Hz. For 2kHz you might need to use a dedicated midrange driver or some EQ.

Which version of HornResp does that? I'm running v17. Perhaps there's a newer version. If I can find some time, I might try the rubber throat distortion analysis. Greg

Oh, man; I really need to make friends with someone with your woodworking skills. I can crank out designs like doughnuts, but I can't saw a straight line with a laser beam. There may be a way around that problem. A couple of years ago I came up with a modification to the tractrix equations that led to what I call the "Elongated Tractrix" contour. It maintains the full 90° mouth of the tractrix, but the horn is longer, like an exponential, and in fact the overall contour is closer to a hyperbolic. I corresponded with Bruce Edgar (no relation!) about it, and he seemed interested at first. But then he simply disappeared, and I don't know whether it was because it was not a viable idea or he just got busy with other things. I'm at work right now, and the derivation is on my computer at home, but if anyone is interested I can post it later. Nice, nice work. I am envious. Greg

(Hopefully not a double-reply; the Website coughed while I was previewing my original response and it disappeared completely.) No "harsh" intent taken from your words. I'll defer to your knowledge on this subject. I'm still skeptical about the rubber throat. Though I know that PWK liked it, simulation models indicate that it leads to a very rough frequency response. If it really does result in audibly lower throat distortion, though, then perhaps that's a worthwhile trade. Thanks, Greg

Yes, now that you mention it I recall either reading that somewhere or concluding it myself at some point in the past. For the home, where the SPL is nowhere near what it is in a theater, that issue may very well disappear. That was exactly my objective in the design I called Alternate Jubilee 3; see previous posts. Thanks, Greg

It's a great way to kill a boring day ... at work. I agree. However, there may be ways to improve it for specific circumstances. I cannot say this with certainty, but it looks to me like the Jubilee was designed as a corner horn that doesn't HAVE to be placed in a corner. In a theater or club there is no guarantee that there will be a good corner available. If that is true, then redesigning it to both be more attractive and as a dedicated corner horn, for use in the home, may be a worthwhile pursuit that just might bring some improvement. Again, I agree. However, given that the Jubilee is an already proven design, it makes a good starting point. Electro-Voice EVM-10M and DL-10X are very good. But I don't know if they're still available. However, just scaling everything in a horn by a constant generally doesn't work very well. Thought experiments are fun. I wish that I could build some of the designs I've come up with, but time and skill prevent it. Thanks, Greg

Thank you, jc. Here is the HORNRESP modeled response using the Klipsch K31 (previous EVM-12L model in gray), and the HORNRESP.DAT parameters so that you can check for yourself. Greg 0.5 x Pi 2.83 0.00 14.99 580.00 816.00 23.53 0.00 816.00 1727.40 59.75 0.00 1727.40 2817.66 29.45 0.00 2817.66 5527.15 44.81 0.00 530.009.69E-04 71.00 5.70 15.90 1.96 1.34 2P 80.00 40.00 0.00 20.00 4.00 376.00Two Klipsch K31; Alternate Jubilee 3 ConConConCon

For comparison, here's the HORNRESP response for the standard Jubilee. Numbers directly from the Klipsch/Delgado JAES article. Again, woofers are EVM-12L; I don't have model parameters for the Klipsch drivers.

Here's the HORNRESP response for that design with two EVM-12L inside (which I used because I just happen to have four of sitting unused in boxes). Of course, HORNRESP cannot model the effects of the folds. And here's the HORNRESP.DAT file info: 0.5 x Pi 2.83 0.00 14.99 580.00 816.00 23.53 0.00 816.00 1727.40 59.75 0.00 1727.40 2817.66 29.45 0.00 2817.66 5527.15 44.81 0.00 500.002.12E-04 32.00 5.20 17.50 3.30 0.50 2P 80.00 40.00 0.00 20.00 4.00 376.00Two EVM12L; Alternate Jutilee 3 ConConConCon

Here's another one. This one is 28½" from corner to face, so the 500 Hz Altec horns will fit. Again, I'm just playing with my CAD program. I wouldn't dream of second-guessing Paul Klipsch or Roy Delgado. But I do like to think about what could be ...

The frequency response associated with Figures 4 and 6 is in Figure 5, not in Figure 9. Compare Figure 5 with Figure 8, and below 500 Hz they're pretty similar.

The 511A is 18½" deep but doesn't need a separate throat. Might work. http://www.voiceofthetheatre.com

Like I said, just fiddling with CAD. Sometimes it's good to stimulate the imagination.On edit: PS "Edgar" is just my forum name; I have no relation whatsoever to Bruce Edgar. I took the name from Edgar Montrose, a character on the Red Green show who likes to blow stuff up.

Won't fit without hanging-over the edge. The 805 and 1005 are 17+" deep, plus 4½" for throat, plus maybe 6" for driver. The bass horn is only 25¼" from the corner to the front face.

I've seen a lot of references here to "ugly" Jubilees. So I sat down this morning with a copy of the original Klipsch/Delgado JAES paper and TurboCAD, to see if I could change the external appearance while keeping the internals the same. The graphic shows what I came up with. It is now a true corner horn. The shape of the backchamber is different, but the volume is the same. The horn follows the taper described in the JAES paper almost exactly, in fact the "A", "B", "C", and "D" callouts correspond to the similarly labeled points in the paper. Note that I haven't built this, only fiddled with it in CAD. But maybe it will give some of the DIYers some inspiration.

I've had an APC UPS for several years; no problems whatsoever. I like it because it provides a real sinewave output. I'm not certain whether, by "power conditioner", you mean the same thing as "uninterruptible power supply". A few years ago I was in the local "Batteries R Us" store and saw a big pile of APC power supplies in the corner, $10/each (without batteries). I asked the clerk to explain, and he said that many people bring in their UPS when the batteries die, and when they find out the cost of replacement batteries they say, "No, thanks, I'll buy a new one," and leave the UPS behind. So you might be able to pick up a used one for a song. Then go someplace like "http://www.batterystation.com/ups.htm" for replacement batteries, and you're in business.