Constant Directivity Tractrix?

[PrestonTom]

I am reading up on things and noticed that a Tractrix flare is not considered a CD (constant directivity) horn. However Roy refers to "his" tractrixes as being CD but he is also careful to refer to those flares as being a "modified' tractrix.

How do they work the magic? I assume it is done near the throat end since (as I understand things) with a more rapid flare the smaller wavelengths will no longer "cling" to the walls of the horn and therefore make the dispersion more beamy (there are probably other factors about the loading also). So my assumption is that the flare is more gradual towards the throat end than the more "standard" tractrix. At the mouth end the flare is certainly not gradual (especially in a tractrix) and geometry would have more to do with the longer wavelengths

I have no idea whether this is common knowledge or whether this is propriatary to Klipsch Inc. I have seen no discussion of this when Bruce Edgar writes about CD flares. I also understand there may be a justified reluctance to discuss the particulars.

Thoughts?

-Tom

[sfogg]

Tom,

Can't answer the question... only to confirm what you have said.

Until Roy mentioned that the K402/K510 had CD like qualities to them I didn't know the combo of Tractrix/CD was possible either. From looking around I'm not sure too many others know it is possible so it may be something new Roy cooked up.

I had avoided Tractrix horns like Edgar's salad bowls as most say they are pretty beamy which I didn't want.

After I learned about the K402/K510 being Tractrix/CD it got me very interested in hearing them.

Shawn

[sfogg]

One additional thing.... I'd also be curious to know (if it isn't a trade secret) how the heck the K510 makes it down to 500hz. Most references to Tractrix (Edgar...etc...) talk about them needing to be used above their flare rate as the horn doesn't really 'turn on' to a higher frequency.

The K510 is a tiny little thing, yet it hits its 500hz rating dead on.

Magic pixie dust?

Shawn

[WMcD]

Like I always tell people. Google your way to Don Keele's website and read, "What's So Sacred about Exponential Horns." (He is has lots of other good stuff too.) Then you can look up his patents at www.uspto.gov. Put his name in the search field and search for inventor. You'll need the TIFF viewer.

Gil

[Coytee]

I certainly don't have anything constructive to add to this thread as it's all WAY over my head. With that said, I do however, seem to recall Roy saying something to me once about the K402 being the state of the art.

I might be in error with my memory though so Roy, if you read that and I'm wrong, don't send me the bones from your (bbq) ribs.

[Coytee]

I stand somewhat corrected...I went back to read what he said and it was NOT that the 402 was the SOTA. He made a comment that another horn was NOT the SOTA. my apologies

[ZAKO]

You can get the jist of Roys 402 horn design by blowing smoke rings through it.

[PrestonTom]

You can get the jist of Roys 402 horn design by blowing smoke rings through it.

Maron,

Your comment is cryptic, perhaps I am missing something. I do not have my thinking cap on today.

-Tom

[PrestonTom]

Like I always tell people. Google your way to Don Keele's website and read, "What's So Sacred about Exponential Horns." (He is has lots of other good stuff too.) Then you can look up his patents at www.uspto.gov. Put his name in the search field and search for inventor. You'll need the TIFF viewer.

Gil

I had a quick look at the patents for Keele. Basically, there seemed to be two strategies. One is to put some sort of vane or a slot, or sectorial divider well inside the mouth of the horn near the throat. The second was to divide the flare into two sections with distinct flare rates (the discontinuity is quite visible). Both these approaches seemed to be used on exponential flares and intersetingly PWK also had some comparable approaches.

In neither case, were the flares of a tractrix nature. They appeared to be exponential (although the initial portion of both a tractrix and exponential flare are similar of course). When I look at the pictures of the K-402 or the K-510, they have the extreme flare at the mouth, consistent with the Tractrix. However they do not have any sort of sectorial, vane or slot that I can see. However, I have not inspected these horns other than what has been photographed. Perhaps something is lurking in the shadows......

Certainly when looking at the MantaRay style of CD horn, the "additions" are quite evident.

What did I miss?

Thanks,

-Tom

[sfogg]

Tom,

Nope, nothing hiding in the shadows. Like you said one can clearly see the change in the Mantaray horn on the right. Don't see that in the K510.

Shawn

[Deang]

I thought a horn's lower cutoff was determined by the size of the mouth?

[ZAKO]

Not cryptic at all...Refering to a Audio magazine atticle of March 1991....It showing a cuyaway of two horns one Tractrix one not..And the pattern of smoke ..It gets into Tractrix phasing & coverage pattern graphs.

[sfogg]

"..It gets into Tractrix phasing & coverage pattern graphs."

Roy has posted coverage patterns of the K402. It is very much a CD horn.

Shawn

[sfogg]

Dean,

If that were true this horn could go deeper then the LaScala....

It only goes a couple of hundred hertz below the K510. The larger mouth lets it hold its coverage pattern to a lower frequency.

Shawn

[Deang]

O.K., that's certainly easy enough to understand. So, is what is written below just wrong, or am I not understanding it right?

"The dimensions of the mouth of any horn determine it's low frequency limit - its cutoff frequency. Regardless of the driver used, it will not work as intended below the horn's cutoff frequency. This is because the horn itself is an acoustic transformer matching the acoustic impedance at the throat of the driver with that at the mouth of the horn. As wavelengths become longer, the horn system simply stops working as intended."

[DrWho]

There are two factors involved....length of the horn and the size of the mouth. Ultimately the low frequency limit is dictated by the length of the horn (as a mouth infinitely large won't extend it further). However, the mouth needs to be large enough to provide a smooth frequency response - too small and you get nasty dips and peaks.

I think most of the magic of the latest Klipsch designs has to do with their awesome phase plug technology. Being able to adjust the shape of the wavefront at the throat has great benefits to how the horn loads. It shows up in the math too - exponential horns assume a plane wave travelling through the speaker where tractrix horns assume something else (I thought it was spherical but I think Roy mentioned that it was actually going from planar to spherical). Generally speaking, tractrix horns can be shorter than exponential ones. With a phase plug, I would imagine you could choose the shape at the mouth and then the expansion rate for the horn needs to change - and I suppose if you're creative you can pick an expansion rate that is condusive to achieving the geometry for constant directivity? Heck if I know []

Speculating is fun though - cuz either I'm right or I get to learn something when someone slaps me with a herring monty python style [H]

[PrestonTom]

Mike,

Back to the topic, the tractrix per se is not CD, as I understand things.

I believe, and I may be wrong, that the horn has been tweaked and it is not some magic with the phase plug. In fact (I think), the drivers are B&C in origin although certainly Klipsch Inc may have had options applied.

In glancing at the patents, I suspect the magic is toward the throat and not at the mouth of the horn. No one is arguing with some of the other points you have made about the tractrix flare. As I mentioned earlier, other folks who have made CD horns usually have started with something similar to an exponential flare and done things that were very clearly visible (either a visible discontinuity between two flare rates or adding "stuff" in the form of slots, vanes etc.

Here the magic is more subtle

-Tom

[bodcaw boy]

I am reading up on things and noticed that a Tractrix flare is not considered a CD (constant directivity) horn. However Roy refers to "his" tractrixes as being CD but he is also careful to refer to those flares as being a "modified' tractrix.

actually, i say that our tractrix horns are constant coverage. the reason i say that is because if a horn that has both a constant coverage horizontal and a constant coverage vertical will have a constant directivity value. but a horn that has a constant directivity value will not always have a constant coverage horizontal or a constant coverage vertical. i say modified tractrix because, one part (that i am able to talk about and the others i can't, sorry.) is that all area expansion equations that i know of dictate area expansion and not what x and y should be. my equation does state x and y.

How do they work the magic? I assume it is done near the throat end since (as I understand things) with a more rapid flare the smaller wavelengths will no longer "cling" to the walls of the horn and therefore make the dispersion more beamy (there are probably other factors about the loading also). So my assumption is that the flare is more gradual towards the throat end than the more "standard" tractrix. At the mouth end the flare is certainly not gradual (especially in a tractrix) and geometry would have more to do with the longer wavelengths

i would like to say that no magic is involved; just good old scientific, engineering......

I have no idea whether this is common knowledge or whether this is propriatary to Klipsch Inc. I have seen no discussion of this when Brice Edgar writes about CD flares. I also understand there may be a justified reluctance to discuss the particulars.

Thoughts?

-Tom

have a blessed day,

roy delgado

[bodcaw boy]

I thought a horn's lower cutoff was determined by the size of the mouth?

a good amount of it is. length is also a consideration and so is flare rate. you can take a 50 hz flare rate horn and truncate it to look like a 300 hz horn.

have a blessed day,

roy delgado

[PrestonTom]

I am reading up on things and noticed that a Tractrix flare is not considered a CD (constant directivity) horn. However Roy refers to "his" tractrixes as being CD but he is also careful to refer to those flares as being a "modified' tractrix.

actually, i say that our tractrix horns are constant coverage. the reason i say that is because if a horn that has both a constant coverage horizontal and a constant coverage vertical will have a constant directivity value. but a horn that has a constant directivity value will not always have a constant coverage horizontal or a constant coverage vertical. i say modified tractrix because, one part (that i am able to talk about and the others i can't, sorry.) is that all area expansion equations that i know of dictate area expansion and not what x and y should be. my equation does state x and y.

How do they work the magic? I assume it is done near the throat end since (as I understand things) with a more rapid flare the smaller wavelengths will no longer "cling" to the walls of the horn and therefore make the dispersion more beamy (there are probably other factors about the loading also). So my assumption is that the flare is more gradual towards the throat end than the more "standard" tractrix. At the mouth end the flare is certainly not gradual (especially in a tractrix) and geometry would have more to do with the longer wavelengths

i would like to say that no magic is involved; just good old scientific, engineering......

I have no idea whether this is common knowledge or whether this is propriatary to Klipsch Inc. I have seen no discussion of this when Brice Edgar writes about CD flares. I also understand there may be a justified reluctance to discuss the particulars.

Thoughts?

-Tom

have a blessed day,

roy delgado

Roy,

Thanks for getting back to me with some of the design perspective.

I wasn't being glib when I referred to it as magic and I also understand that this is intellectual property. I imagine you waded through a good number of math simulations and built many physical prototypes.

-Tom