Friday, 12 January 2018

The Western Electric KS12024 horn in more detail.

This post is to explore why the KS12024 horn stands out as one of the best sounding of all small horns. What is different from other horns? How do these differences influence its sound and can we understand them? It is important to remember that if we observe a difference and cannot explain it this lack of explanation does not mean there is no difference.  It only means we have not yet been able to catalogue and attribute it but still the difference remains. Some electrical engineers have disagreed with this and argued that if it cannot be measured and explained then it does not exist. This is a banal error (so lacking in originality as to be obvious and boring) and can cripple eventual understanding. Science is a tool, and a very useful one, but if you let it run the show it will still be a tool. When science can reproduce a Stradivarius, or better still, have thought of and created one, let us speak again.

Let us begin with the observation - the KS12024 horn sound. I shall use other observations than my own - a conversation from a Japanese publication:

"Atarashi: The first listening of the KS - 12024 horn is pretty good and very impressive. Colourful expression but very soft and right tone.
Shinoda: Good compatibility with the WE555 – very balanced.
Doi: KS12024 horn is the core of Westrex sound systems.
Atarashi: You understand when you hear ‘’Casal’’ with such a good quality. The fact to clearly hear both the sound of the ringing and the sound of the string body, both being organically overlapped, gives the firm presence impression of the cello. Emotion is well transmitted, as if to express the music with the help of like Casals Bach. It is an emotionally rich sound.
Shinoda: I am not focusing on scratches on the SP record; I can concentrate myself on music. I feel a sense of depth and it sounds quite well.
Doi: More noteworthy, it is no fun listening to music. No likes and dislikes of music genres.
Atarashi: Yes. "Casal" as well as "Sonny Rollins" or "Kay Starr", all are lively and enjoyable performances.
Shinoda: I think this system tends to give a HiFi tone, I’ve felt the energy impression and the thickness of the sound together. For example, "Sonny Rollins" in front of the stage, plays saxophone lively first, joined by strong sidemen from behind with Max Roach, you can hear the music even more impressive. That’s because there is a bodily sensation, you can feel the thickness part of the sound, performance is close to the human perception without being disordered.
Doi: There is an energy impression in sound, at the same time also the tone is beautiful. However, it is wide range, as if it was a full range system. This is a very valuable horn, which gives the best quality of the WE555 driver, according to me.
Atarashi: When you heard the sound of this combination, you think that extension of WE555 is not difficult to obtain (laughs!).
Shinoda: I agree. The WE555 is splendid, and this sound is difficult to obtain with many horns."

Quotations taken from this publication.

On initial examination the horn incorporates:

1) Simultaneous and varying exponential and linear expansions.

2) Multicell seperations which begin about 70mm from origin of throat - before the bend and change in opening angle.

They look like this at their beginning.

Profile x-ray view shows where these begin - before the introduction of the curve and increase in linear profile.

3) A 45° bend 90mm from throat origin where the linear opening angle increases from 30° to 50°.

This poses the question of a possible correlation of the bend and increase of linear opening angle. 

4) Throat size is WECO standard 0.7" or about 18mm

5) Cast aluminium has roughed surface perhaps performing boundary layer excitation (BLE) - the intentional creation of small modes which prevent larger ones from developing. Earlier in wooden horn building I was surprised that the horns sounded better before they were sanded and finished. The better they looked the worse they sounded. This is perhaps why. Of course we stopped finishing them...

A simular texture is used in reproduction for this reason.

6) Exponential expansion begins after 180mm from throat origin alongside the linear which continues and determines the output pattern. The linear propagates in the lateral expansion - the exponential the top and bottom. You might want to call this a radial horn however it is purely linear for the first 180mm, then becomes the hybrid which is termed radial - but with multicellular separations.

In conclusion the expansion is 30° opening angle linear for 70mm then this is divided into 4 asymmetric linear cells which approximate exponential expansions. These increase to 50° linear   in total at 90mm and go for an additional 90mm. At this point another exponential expansion begins on the top and bottom (going into radial) further increasing the expansion accordingly. It is a thus compound horn: linear 30°linear/multi cell 50°, then radial multi-cell.

Further interior examination shows there are two different countours of the four cells. 

These are grouped as the two internal cells and the two external cells. Each being slightly different to each other. These individual linear expansions appear to approximate exponential.

They are different even if we consider the spherical wave expansion as can be seen in another x-ray view.

This is probably enough to separate the 12024 horn from other other horns. One can only wonder how this horn was developed.

Horn material makes a difference in the sound - as sound waves inevitably travel through solids with increased speed the material also alters them - accentuating some frequencies and attenuating others. These waves then leave the material ahead of the sound traveling only in the air inside the horn. The original aluminium lends a ringing character (or glare) when compared to plastic, wood or bronze - as can be expected. Plastic improves on aluminium in reduction of glare and is noticeable when directly comparing the two materials side by side. AL was probably selected for production or/and ruggedness in a commercial environment rather than for epicurean reasons. Bronze takes on a quieter, clearer and more present aspect. It also lends a sweetness or naturalness something like nickel does in transformer cores. Wood has its own and lighter form of natural sound. Combining these works very well.

In summary, all of these elements combine in a horn which is singular in its natural presentation. It can go lower than other horns with similar mouth areas because there is less distortion near the mouth cut-off frequency. Most horns need a margin of an octave, or half and octave at best, above the mouth cut-off frequency to avoid noticeable distortions. The KS-12024 does not seem to require as much margin resulting in a smaller form factor for the frequencies it can propagate well. It is one of the best of all small horns in doing what it does.

In the another post I will talk about different drivers in the KS-12024 as well as compare the LM555, TAD 2001 and our proprietary driver with phenolic diaphragm.

Until then...

1 comment:

  1. Where are you crossing over? Sorry if I missed that somewhere.