Would this idea work?

[startgroove]

Hmm, I'm not sure what you mean. The cone I tested with was closed at the small end, no hole at all. I intended to test it as a reflector, nothing more. Perhaps you could explain your meaning in a drawing?

[Marco Gilardetti]

Sure, will do a drawing on millimetric paper. It will be evident on the drawing. Will probably post it on monday as I have guests in my house in these days!

[Marco Gilardetti]

...and here's the drawing.

As in a horn the important factor is the surface offered to the expanding traveling waves, the target is to draw a horn (lower graph) with cross-section surfaces equivalent to the cross-section surfaces of the "modified" witch hat horn (upper graph).

Moving from the throat to the mouth, the "modified" and the equivalent horns will be absolutely identical from the throat up to the point "A". Here, the surface offered to the waves abruptly changes from a circle to an anulus (A-A'), that is a surface limited by an inner (radius = r) and an outer (radius = R) circle. The formula for the surface of the anulus is known, and in the equivalent horn it is drawn a section which surface is equal to the corresponding surface of the anulus. The equivalent radius of the section is given by the square root of the difference of the squares of the outer and inner radiuses of the anulus. It is easy to understand that from this point onward, the expansion rate of the equivalent horn will no longer be linear, so the calculations are repeated at points "B" and "C" and an approximate curve is drawn with segments. At point "D", the inner inverted cone is reduced to a point with zero surface, so once again the "modified" and the equivalent horn have the exact same mouth surface.

Looking at the equivalent horn section (lower graph) it is easily seen that all goes more or less well up to point "A" (we could argue that a conical horn is per se one of the worst when it comes to impedance adaptment, but we would digress, the relevant point is that up to "A" the horn is progressively expanding). At point "A" the horn retracts all of a sudden, and the waves are forced to pass through a "hole". This "hole" can be addressed in many ways: perhaps the easiest is to see it as a surface that limits the aperture of the mouth. In any case, it will do harm to the lower frequencies in first place, and will also prejudice the efficiency of the horn in general, as the horn no longer expands from the throat to the mouth, but it also contracts. After point "A", the horn expands again, but with a sort of "inverted quadratic flare", which will also prejudice a good impedance adaptment.

My conclusion is that this device spoils the sound in many ways, and ironically produces effects that are contrary to those suggested by the author.

[startgroove]

I would recommend that you repeat my experiment, as outlined above, using a "reflector" as described in the original post. It would be interesting to see if your results coincide. Who knows, a second set of ears might detect things differently.

[Marco Gilardetti]

Done as first thing!

My best guess is that the original author of the article suggested this modification (I mean: aside of a severe ignorance of how the horns work, but possibly little was known at his times should he have had the idea before Lord Raileigh works were re-discovered) because he got the impression to hear less high frequency noises due to the fact that the throat of the horn was concealed by the inverted cone and was no longer in "direct sight" in respect to the ears.

My comments about this are: first, this will completely kill the most emotional part of listening a good record on a good gramophone, that is (as I have read here and elswhere thousands of times; and I subscribe) to stand just in front of the horn and have the ears overwhelmed by the impacting sound delivered by a horn; second, the front peak lobe of the emitted sound is just moved elsewhere (sideways) by the new horn geometry.

The same effect, if desired, is better achieved by listening an unmodified horn sideways, or even from the rear: the listener would be out of the front peak lobe of the horn, but without compromising the bass sounds / efficiency / loading and impedance adaptment by changing the horn's geometry.

[justin ball]

My brain hurts.

[Daithi]

Why does the HMV 5A and 5B soundbox have a tiny cone in the center of its exit aperture ?

[AudioFeline]

Well to my suprise it actually worked i did try it with a very cheap thin steel horn so I think all it really did was remove the vibrations from within the steel by sending them into the balsa wood so I would advise this to anyone with a very thin metal horn as it does give a less harsh sound than original however it hardly compares with my HMV re-entrant!

Thanks, Jake

So from this description, it sounds like the benefits came from the mounting of the reflector acting as a brace, taming the horn's internal vibration.

It might help the discussion if you can scan an upload the relevant page(s) from the book, so some of the detail isn't lost in translation. My initial reaction to the concept is that I can't see how it would improve the sound. But I'm happy to be proven wrong!

[Inigo]

Probably all you are right. There's an effect of muffling the direct, straight treble tones, and also a direct effect of damping the horn resonances, which another member of this forum, Daithi, has achieved by means of mass-loading the horn by wrapping chains or something alike, around the external part of the horn. And there's also an effect of enlarging the sound path by the air column to surround the obstacle, which also has a resemblance to the folded horns. The sound path is forced in this way :

1594316365973.png

All together it tends to improve sound, killing that penetrating tone of these machines

[Daithi]

and also a direct effect of damping the horn resonances, which another member of this forum, Daithi, has achieved by means of mass-loading the horn by wrapping chains or something alike, around the external part of the horn.

I think its someone else was doing the chain thing Inigo. That would be too brute force for me. My instinct for making a horn inert would be to coat it in liquid rubber or if its made of steel attach that magnetic sheet rubber stuff they use at the back of fridge magnets. I am currently trying to make a horn out of thick section PVC pressure pipe which I think is fairly sonically passive to begin with. Back in the good old days they got good clean results by using Bakelite.