Article: Solving the Klipschorn Throat Riddle (Edgar)

[WMcD]

Attached to the frame below are two, related articles by Dr. Edgar. There is also a correction of a graph by means of a letter to the editor. I've gone a bit overboard in typing in notice of these corrections. OTOH, I did want to make sure no one was misled.

Gil

Edited September 5, 2015 by William F. Gil McDermott

[WMcD]

This is the only source of information I know of regarding the effect of the throat restriction. Please also know I have a lot of respect for Dr. Edgar and all he had done to bring information on horn design and building to the amateur.

Gil

Solving the K-Horn Throat Riddle and Correction.pdf

[3dzapper]

Hmmmmmm....

It seems in the end, that Dr Edgar agreed with PWK, the slot sounds better but I don't know why. The good doctor hypothisizes reflections with no emperical evidence. Of course PWK did all the developement work without the benefit of Dr. Small's 1979 paper.

This all illustrates the "MAGIC" of the Klipschorn!

Thanks once again Gil.

Rick

[djk]

Dr. Edgar's test horn results are invalid because there is no splitter.

The splitter forms an exponential coupler (short horn) with the constricted throat, causing increased radiation resistance (more output) over a narrow range.

See Olson for an explanation of exponential couplers and horns with multiple taper rates.

[michael hurd]

Yes, Edgar was testing a straight horn, not a bifurcated path corner horn.

[D-MAN]

Gil, thank you.

Excellent reading.

This is just my opinion (and it may change as I learn more so feel free to disagree):

As far as the Khorn splitter, it is a simple Heugen reflector, and does not create a constriction at the throat cross-section as the available channel area on each side of the wedge is over 1-1/2" in width, the same as 1/2 of the throat cavity opening (3"). It merely turns (splits) the waveform into the bifurcated horn channels which are 3" in depth vs. the 1-1/2" of the "slot".

If it is not used, then the driver would be subject to 180 deg. reflections from the front board directly back toward the cone. However, there are many different bass horn designs that allow this to occur without much damage, but when they do they do not tend to be bifurcated-at-the-throat designs. I'm thinking specifically of the University "Classic" as an example.

The restriction/constriction is actually the throat cavity opening (slot) itself, being smaller than the optimal throat size, it increases the impedance to the movement of air through it, and reduces the upper band pass frequecies, the lower frequencies being less effected, but overall efficiency lowers at the same time, of course.

DM

[D-MAN]

In trying to explain my opinion further, here's what I mean in a hacked-up diagram...

Seems to me that some people just overcomplicate the situation - looks like a pretty solid explanation of the splitting wedge, agreed?

The "slot" size is reportedly calculated for overall efficiency and bandpass in the general rule of thumb "the wider the slot, the higher the crossover point can be." from "How to Build Speaker Enclosures", Badmaeff and Davis, W.L Sams Publishing, Pheonix, AZ, 13th reprint 1978. This was before the publishing of Thiele/Small driver parameters, so it tends to be rules of thumb.

DM

[D-MAN]

I think that the starting assumption of the article is incorrect. Here's why... {edit} Wait! I assumed this wrongly; the diagram is correct ; it is NOT representing the Khorn throat, but is an example horm only. However, it is NOT a "constricted" throat as used in the article, its an acoustic filter. The throat cross-section DID NOT CHANGE!

DM

[dodger]

As far as I figure the point D-Man, you are correct.

[D-MAN]

Here is an example of a bass horn using a compression chamber (more or less) in the horn throat. Usually this is used for MIDS and TWEETERS but this is the first example that I've seen in a bass horn in this example, much like an Altec A7 (828). THe reported effect of this compression is that the midrange dispersion characteristics are pronounced horizontally in a wider pattern. The compression chamber is clearly NOT used in a Khorn, nor is it the case for the acoustic filter slot.

DM

[D-MAN]

Here's an acoustic filter slot being employed in a 1954 patent. This is right along the lines of the acoustic filter slot used in the modern Khorn, and works from the same principles, that is, a physical high-pass filter whose width effects the amount of high frequencies that are passed through it. Note that at the time this was patented, the Khorn was "still" using the 6x13" opening, so this slot (in essence) was a precursor to the later narrower Khorn slot made in 1961. I believe that the filter slot was explored by Olsen in a 1950's publication. This reminds me of the old adage "those who ignore the past are doomed to repeat it".

DM

[D-MAN]

A few very rudimentary diagrams that partially form a basis for slot theory, hacked from a couple of websites. Note that these are referencing a round "hole" vs. a rectangular one.

DM

[D-MAN]

Here's a hack job on the Khorn patent to make a point. PWK originally used a compression chamber (capacitance) in front of the driver but it was a 12" WE field-coil woofer. It is NOT used for any of the larger production 15" woofers employed in the Khorn. Also the patent discloses a wider proportioned "slot" for the same reason. The chamber was located between the driver and the throat opening, not actually in the throat!

DM

[D-MAN]

About the effect of the narrow slot on the Khorn with the K33E, it can be seen that the "sag" at 400Hz was somewhat addressed by the narrower 3x13" slot than the original 6x13" size. This corresponds to PWK's statement(s).

However, lower efficiency and some other response "dips" were accepted at the same time. As to whether it actually flattened out the response is somewhat open to interpretation, as far as I can tell. Where this leads is that the Khorn itself was SLIGHTLY modified (only the slot and driver mounting) to "fit" the K33E SLIGHTLY better, and that, the most likely reason for the K33E being the determining factor was for economic reasons as Edgar stated. (I'm warming up to him).

I stated previously that a narrower slot would not raise the a "so-called dip" in the response, rather unequivecally, too. I'm not sure who's right here... it sort of works, but not to an "important" degree. I suppose that its a matter of interpretation. I call it a "slight" change, not enough to treat as a general rule. The important part is that it ONLY pertains to a K33E, doesn't it?

Remember: If it isn't a K33E, the slot should be wider!

DM

[Klewless]

While we are on this thing about the 3 X 13 slot in the Khorn,

What is the story between it and the LaScala????

[Deang]

In the graph, why does the 6x11 throat response look better to me? What am I not seeing here?

[djk]

"The compression chamber is clearly NOT used in a Khorn, nor is it the case for the acoustic filter slot."

The air trapped between the cone and the opening in the motor board constitute a compression chamber in the Klipschorn.

The acoustic filter slot does not as there is no horn to load it.

[D-MAN]

----------------

On 5/25/2005 8:57:28 AM Klewless wrote:

While we are on this thing about the 3 X 13 slot in the Khorn,

What is the story between it and the LaScala????

----------------

The using the same woofer (K33E) means "use same slot size" for exactly the same reason to alter the response of the driver.

However, the throat cross-section is the same 6x13" in the Khorn, LS, and Belle. The folds, expansion rates and overall horn lengths change between them though.

{edit} This is somewhat contrary to the findings of the "correction" article by Bruce Edgar as he concluded that the particular folds of the Khorn were the main reason for the mysterious throat opening behavour. Since PWK used the same throat opening sizes for the LS and Belle, it seems that PWK did not consider the foldings or other considerations (horn-length/expansion rates, etc.) to be of much concern in the matter of throat opening size selection. Rather it seems that the use of the same driver in the same throat cross-section area in all three horns being the determining factor, I'm guessing.

{edit2} did you know that the Khorn and the LS and Belle have the same MOUTH area? It's true, about 4 square feet.

The lengths are quite a bit different though, at approx 65" for the Khorn and about 19" for the LS/Belle. The THROAT area and cavity openings are also the same size!

DM

[ajsons]

As far as the Khorn splitter, it is a simple Heugen reflector, and does not create a constriction at the throat cross-section as the available channel area on each side of the wedge is over 1-1/2" in width, the same as 1/2 of the throat cavity opening (3"). It merely turns (splits) the waveform into the bifurcated horn channels which are 3" in depth vs. the 1-1/2" of the "slot".

If it is not used, then the driver would be subject to 180 deg. reflections from the front board directly back toward the cone.

DM

Here's one that I find rather interesting. The other "throat", at the TAILBOARD.

A copy of the 1945 Klipschorn construction plans was sent to me by forum member ap sass. (Remember, the patent was in the same year, April 1945). The 28 sheets of the construction plans were drawn at different times in 1945, from what I can see. Anyway, here's something we don't see in the patent drawing. A simple Huygen reflector at the tailboard.

[ajsons]

Here's the part to build it.

It seems logical that the horn would work better with it, based on earlier posts by Dana, but somehow Klipsch dropped the idea for whatever reason. Maybe a full Huygen reflector is needed, vice a radius reflector, something that's probably worth looking into.

Armando