Jump to content

Changing the throat area.....


Recommended Posts

The Klipschorn is basically a trihedral, bifurcated, multiple impedance, exponential horn. As such, it is composed of what could be considered a series of horns or horn sections, whereby the acoustic impedance at the mouth of one horn section matches the throat impedance of the next section, & so on, as the horn unfolds. If you change the size of the throat area, you are changing the acoustic impedance at the throat which will likely result in some amount of acoustic impedance mismatch between the horn sections. Impedance mismatch = non-linearity = frequency response anomalies = distortion.

Link to comment
Share on other sites

On top of all of the preceeding, the horn throat is not the opening the woofer fires through. The throat is the area infront of the first splitter. Since it goes 2 ways, the throat area is twice the apparent dimensions.

Link to comment
Share on other sites

The woofer of my project is 97dB in efficiency, while the midrange(110dB) and tweeter(107dB) are way more efficient. I read that as you increase the throat area, the efficiency goes up(to the maximum when the throat area is eaqual to the driver) while the band width goes down at the upper range.

Is that true or what?

Link to comment
Share on other sites

Your situation is normal. That is why most upper frequency drivers are attenuated.

I believe the Khorn throat is about half that of the woofer (must use effective woofer area in calculations, not the physical size). I have seen text that states that maximum efficiency occurs at half the woofer area, so the Khorn is already operating at max efficiency.

It is also one of those vicious circle things were you tighten the throat to improve upper frequencies of the horn, which in turn starts choaking the lower frequencies increasing distortion, etc, etc. Enlarging the throat will reduce the overall efficiency and perhaps improve parts of the bass slightly (less constriction and probably increased resonance) at the expense of the bass horn's upper end, which is lacking in the Khorn already.

In other words, the Khorn is probably at it's best already which explains why it has been so successful. Improvement will be starting over with an entirely new design, including new woofer.

At least all this is my understanding of bass horns. You may take it with a grain of salt, since I do not consider myself to be a horn guru. Paul got it right in the first place.

Good luck.

Link to comment
Share on other sites

  • 7 months later...

I have been experimenting with various throat widths and crossover values. I have found using the K33E that the wider the throat (up to 6x13 max), the more mid and upper bass comes out; this can be used to more accurately mate with the crossover/midrange setup (bear in mind that I have a slightly larger back chamber area than the standard Khorn specifically for experimentation and am not using a standard midrange horn). The unfortunate draw back is that with the bass horn and the midrange smoothly integrated, a noticable softening of the bass frequencies occurs (it sounds rather "tube-like"). I can only surmise that this is an effect caused by what I would call "smearing" of the mid-range with the slower (due to horn length differences)arrival time from the bass bin. The resistance or reactance/capacitance in front of the driver is also changed which of course alters the crossover electrically too. The resultant LF air pressure is definately increased, so I assume that this indicates greater efficiency as the midrange required a -10.2 db setting to match the bass horn output using an ALK type A crossover. As the midrange/crossover/woofer/throat/back chamber combination is inextricably interrelated, it is a complicated mess to play with - unless you really have the need to change drivers, etc., I would say that it is best avoided and left to PWK. Also I have experimented with other drivers, and found the 6x13 in throat to be the best for other 8 ohm drivers; they tend to be far more punchy and efficient than the K33 (something that you really don't want unless you want to change the mid and tweeter to keep up with the bass output), but the K33 has a transparent quality to its bass that I really cannot describe accurately. It is more like listening to an instrument rather than a speaker, if that makes any sense... the other drivers would beat the hell out of you, but you would never forget that they were there; unlike the K33, which lets you forget where the bass is coming from; its just "there" - natural-like.

Link to comment
Share on other sites

The present approach is to pick the throat size based on the parameters of the woofer. There are some specific equations for that. Some by Wente and Thuras, which Klipsch cites to. A eqivalent equation was developed by Don Keele. These maximize midband efficency.

I agree with John that the throat size is set by the area just at the wedge. It is 78 square inches. I.e. two 3 x 13 inch throats.

There was a theory that a smaller than optimum mouth will increase upper end effecency. Don Keele's paper on use of T-S parameters for throat size and high end cut off cited by a paper by Prof. Small. My recall on the latter is that he is saying: Yeah, you can optimize for midband effecency by picking the throat size, but it is could be beneficial to have some flexibility to experiment to optimize overall frequency response.

The smaller throat size also came up in a work by Harry Olsen. His point was that you could use a horn with a high frequency flare at first and then later down the axis, go to lower frequency flare. At least one illustration showed that he had yet a third section with a lower frequency flare. But that starts looking like a conical horn to me.

In any event, the very first KHorns had a rapid flare at first and used a 12 inch driver. PWK called this a rubber throat. There was no rubber in there. It was just that at high frequencies it looked like a small mouth, and then at lower frequencies looked like big mouth. This is what you see in PWK's early publications.

Per PWK in an interview, he said when he switched to a 15 inch driver, he didn't need the rubber throat. So the throat size just at the flare after the wedge went to 78 inches square. Generally, there are three flares and they are probably all close to an Fc of 48 Hz.

The next factory mod was to put a restriction at the motor board, 3 x 13 inches. It was upon that that Dr. Edgar wrote. He eventually said, Whoops, I'm wrong about my theory. Jim Hunter from Hope sent him some sound curves showing the difference in response. They were different, but which is better is hard to say. The goal was to increase output at 300 Hz were the KHorn is drooping.

The issue was somewhat confused in SpeakerBuilder in that the original graphs published had the effect of the restriction reversed with the normal. There was a correction.

Bruce is a brilliant guy, but I think his approach was incorrect. I've not seen anything from Klipsch about the theory behind the restriction. Some years ago I published here my theory of reverse engineering. It may have disappeared with revisions to the bbs.

One theory you see in books on horns and drivers is that a front chamber can help high frequency output. However, the 78 square inch throat is almost as large as the working area of the driver diaphragm. You can't form a chamber with one end open to the throat. Therefore, the restriction could have been envisioned to form such a resonant chamber even if it messes up the situation at the throat.

Now you'd think this makes no sense because there is a contradition between the restrictor plate and the throat size. There was a comment somewhere, I think in Olsen, that the pressure situation on either side of a small restriction (e.g. 3/4 inch plywood) it tough to judge. The pressure is close to the same on either side; on the other hand, a resonant chamber implies there is a localized pressure increase in front of the diaphragm.

The solution may be that, like the rubber throat, it (the restriction) is only present at high frequencies. Note, this assumes your stuck with effective thoat size. In HF drivers, the throat size is usually much smaller than the diaphragm.

My conclusion is that the best minds think it is kludge, it does something, and is difficult to predict.

It seems the rubber thoat still lives on in the Jubilee. If you read the Klipsch-Delgato paper, there is a quicker flare at the first expansion. My guess is that they wanted to hold the same footprint as the original K-Horn. When you do that, the width of a 15 inch driver starts taking up too much real estate, so they went to 12 inch driver. Then had to use a rapid flare.

Like others have said, I don't suggest you change anything. I'm just describing the evolution of the present situation.


Link to comment
Share on other sites

It seems the LaScala does the same thing. The expansion left and right are very quick. Then there is NO expansion as the air moves toward the front for a few inches. And then the final flare rate. I did some two dimensional drawings of this, using the area at the different points along the horn. There has to be some reason, because it wouldn't have been that hard to do differently. PWK didn't guess at all this like some folks seem to imagine.


Link to comment
Share on other sites

Gil and others:

I hope I was successful in attaching a .jpg of a cross-section diagram of the throat area of my 1962 Klipschorns. Basically, five-sided longitudinal wedges about 1/2" thick and 4" long are solidly attached (nails and glue, I think) immediately above and below the 13" X 3" opening to the woofer horn, on the other side of the woofer. Some have thought this was an experiment by Klipsch, apparently deemed inconsequential enough to ship the speakers out as regular production items.

This has intrigued me ever since discovering it. I became the third owner of the horns around 1980; however, like the previous owner, I was not fully satisfied with how they sounded. "Updating" with new drivers and crossover in the early 1980s made it much worse. The bass section did not seem to integrate sonically with the midrange/tweeter.

I sent the top section to Klipsch around 1984; they said they found nothing wrong, but modified the tops and flush-mounted the tweeter. They were definitely better but still not how I thought they should sound. After extensive equipment changes over the years including an AK-3 update, and more recently going to a tube amp (Joule-Electra) and installing the AK-4 tweeter/MR/crossovers a few months ago, I feel it finally sounds pretty good but still wonder what the effect is of those woofer throat "enhancements". Those I've talked with think it would be impossible to remove them without a lot of damage.

Any thoughts? I hope the jpg attached; if it didn't, I'll try to find out how to do it.


Link to comment
Share on other sites


Your post started me thinking...

Where exactly is the throat of a Klipschorn?

PWK refers to the "throat opening" and the "throat" interchangably in his 1941 paper "A Low Frequency Horn of Small Dimensions" or so I take it.

So for clarity, let me rephrase it as "I changed the size of the THROAT OPENING"; of course, I was not referring to changing the dimensions of the exponential horn itself. The slot-shaped opening at the throat can also be referred to as an "acoustic low-pass filter" which is much harder to type and contains the spelling hazards of big words so I want to avoid it.

I would disagree with you that the throat is 6 x 13 sq in. in that the volume displaced at the throat by the splitting wedge must be taken into account as it further restricts the available area of the horn throat in particular with the 3" slot in place. It appears that Klipsch reduced the size of the low-pass slot without changing the horn which retains the dimensions suitable for the wider (original) 6 in. wide throat opening. Clearly, the intent was to restrict the broadcast of higher frequencies available with the K33E.

Link to comment
Share on other sites


From your first post pertaining to this thread:

"...I have experimented with other drivers, and found the 6x13 in throat to be the best for other 8 ohm drivers; they tend to be far more punchy and efficient than the K33..."

With which other drivers did you experiment? Did you try them with the 3" x 13" restrictor slot or only the "6 x 13"?

Link to comment
Share on other sites

I have tried 4 different sets of drivers in my corner horns so far. Here's my opinion of them, for what its worth.

I tried both versions of the Eminence Kappa-C's.

Stay away from the 4 ohm Kappa C, you won't like it... deep bass but very soft, almost no mid or upper bass at all.

However the 8 Ohm KAPPA C-LF is promising. It will require the 6x13" throat opening and a change to the xover low-pass coil to about a 3.2 mH or so (although it is listenable without changing it- the upper cutoff won't be quite correct, though.

It is quite punchy (great mid-bass) and is a little more efficient than the K33E but doesn't go quite as low (its Fc is 39 Hz. and its rated at 98 db, if I remember correctly). They are well built and are cheaper than the K33. In a corner horn, they will change your heartbeat at 30 watts per channel and a Tower of Power cd in even the most questionable of listening spaces (which is what I've got). But bear in mind that they will not blend in seemlessly with the midrange unless you change the low pass coil to match the 8 Ohms. If you don't the midrange will definately become harsher.

I have also tried a set of Seleniums now out of production(I forget the model#, but WPU202 seems close- the ones with the user-replaceable cones) that were rated at 99 db at 1 meter. These kicked major butt, but then soon went out of alignment and scraped the voice coils off. Now we all know why they went out of production... That's what I get for believing what I read on the internet!

I then bought a pair of K33E's from Klipsch and like them alot. I knew I liked them best from the first note. They don't have the punch of the LF but they are a better match to the midrange/tweeter/xover setup that I currently have.

I have been eye-balling the new Selenium (non-replacable)WPU?? (I want to say 505 but I could be wrong) with the published sensitivity of 96 db and an xmax of 9mm and an Fc of 36 Hz. These may be worth a try; the specs look good.

Has anyone tried the Dayton 500w 15" driver?

They look reasonable for a horn application...

Link to comment
Share on other sites

If I remember right, Klipsch used different expansion rates for each section of the K-horn. He refered to it as a rubber throat. NOt sure which paper he thus discribed it, but it was in one of his writings.

Upon further checking, he discribes this effect in "A Low Frequency Horn of Small Dimensions", 1941 and in "Improved Low Frequency Horn", 1942.

Jim N

Link to comment
Share on other sites


Serial Nos. 1A836 and 1A837. Yes, very interesting, but as you can tell from my comments, not necessarily positive. After I discovered them, I got in touch with an engineer, Jim Hunter, in Hope; he found notes in their files that, in July 1961, a modification of "four (?) multi-tapered blocks" was installed in the bass horn mouth. I believe he talked with Paul K about it, but apparently got no further info. My notes indicate that Hunter said the mod was in production for two to three years, from 1961-1963, so I guess mine weren't experimental one-offs after all.

As purely my own surmise, I wondered if Klipsch may have been trying to deal with the sonic afteraffects of having to change bass drivers after E-V discontinued the Alnico 15WKs. I recall that a local dealer (and I myself) didn't think the new woofers sounded as good as the old ones. I'd certainly be interested in authoritative views on this.

Link to comment
Share on other sites

  • 2 weeks later...

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Create New...