In my application the levels will not be extreme (2-ch listening at home and the cabinets will be quite efficient to begin with). So my questions surround the issue of whether I need a rubber throat at all.

1) Are there any known disadvantages to using a rubber throat?

2)Has anyone actually a built front loaded cabinets with and without a rubber throat and measured any differences (esp distortion and output)?

Thanks in advance

• Hornresp

Conceptually I am wondering if the rubber throat section is acting like a chamber (after the throat) that happens to have a tapered flare). I believe that such chambers were sometimes used to effectively low pass filter the output. However I am going off-track in my speculations...

Anyhow I will give it a shot.
-Thanks

First, all horns are simply acoustic impedance transformers by nature, even in short sections. They can be designed to increase or decrease the impedance by flare rate and channel volume/cross-section between sections of the horn. The use of a differing flare rate (from the rest of the horn) section of horn AT THE THROAT is referred to as PWK called it a "rubber" throat, implying that it changes its impedance based on frequency, another characteristic of any horn. In this case, it is the horn throat section. Having such a throat horn section allows for greater control over the overall efficiency, distortion characteristics, and operational bandwidth of the horn, along with allowing for the desired selection of the throat cavity opening size (the main determiner of throat reactance as far as the driver is concerned). This allows for the "targeting" of a particular crossover point, a desired crossover slope, a particular driver, etc.

The "rubber" throat is a method used to transfer or alter acoustic impedance. It is typically used to LESSEN the throat reactance experienced by the driver by employing a more rapid flare rate than the subsequent horn sections which usually use a less rapid flare (lower Fc) rate. I think the so-called "rubber" effect is a misnomer.

The throat of rapid flare compared to the rest of the horn does not act as a chamber, it is a full-fledged horn (section) in its own right and operates exactly the way a "normal" horn does, and it will NOT effect the overall low Fc of the horn within reason, of course(that is determined by the terminal section of the horn).

The goal (usually) is to provide a reduced (i.e., less acoustically restrictive) throat cavity opening/throat cross-section as compared to a single-flare rate horn of a lower Fc, yet retain the overall low Fc of the horn.

Therefore, a wider operational bandpass with lower distortion is made possible by the "rubber" throat when the acoustic impedance is smoothly transfered between the differing sections. Take note that PWK typically used folds for his impedance matching, often associated with non-expanding "pipe" sections to achieve an adequate impedance match between the different flare sections. The Khorn, LaScala, and Jubilee all use different flare rates throughout. The LS is an interesting standout in that it uses a 60 Hz flare throat section and a 125 Hz terminal (mouth) section for an overall Fc of approx. 70 Hz.
Not the usual configuration, and was likely done to allow the use of the same driver, throat opening, and crossover/slope as the Khorn, an economic decision, I presume.

DM

DM: I understand the technique as a way for reducing throat overload distortion. However for home listening I am assuming the drivers are not being driven very hard to begin with. That is why I was wondering about the necessity of the initial rapid flare (rubber throat). You did mention a non-expanding pipe section. That is interesting, since such a section is found immediately after the rubber throat in the the Delgado & Klipsch article (JAES, 2000).

I had always assumed that was weak point in the design (the Jubilee) since 4 parallel walls (for about 5 inches after the radius reflector) would create standing waves. I had further assumed this "convenience" was considered too short to have an appreciable decrement.

Just something to think about. The flare changes at the folds make sense though, easiest to accomplish. The parallel wall cross-section lengths of the non-expanding channels are far too short to be much of a sub-multiple of the wavelengths being produced, so standing waves are highly unlikely, rather it is just an approx. 1:1 impedance transfer conduit over a short distance for strictly practical reasons, IMO.

DM

Look at 1990 back-issues

• Speaker Builder Back Issues