Baffle Slots, Horn Mounting, and Various Horn Considerations...

[D-MAN]

The question has been brought up about the issues of flush surface (front) mounting of horns (tweeter) vs. the rear-mounting of the same horn, i.e., from behind the baffle board... cabinet diffraction remaining the same, I would tend to vote for recessed mounting as having a beneficial effect - for VERTICAL orientation only for wider horizontal dispersion.

There is some argument that the rear-mounted ones would have a wider dispersion IF the "slot" was oriented vertically. For instance, the EV T35 is a diffraction horn that should technically be mounted vertically for the widest horizontal dispersion.

Consider the attached patent. I've only included the drawings for those of you who cannot get the uspto TIFs to display but you will have to look it up and read the text. Also consider that the Klispch Heritage horns use a slot-filter in the horn mouth - same considerations also apply. Anyway, it's interesting.

Here is the link: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=/netahtml/srchnum.htm&r=1&f=G&l=50&s1=2646851.WKU.&OS=PN/2646851&RS=PN/2646851

DM

[D-MAN]

Drawing 2 of 3

[D-MAN]

3 of 3

[D-MAN]

Some further considerations in light of the above concerning a baffle layout for high frequency horns (in particular using a diffraction horns such as the EV T35 and its ilk). This would change if one was using a non-difraction horn such as a JBL 2404, where perhaps a central axis - surface mount (from the front) approach would then be better.

As it has been stated previously, the central axis mounting relieves the requirement of producing pairs in opposite mirror versions (one for left and one for the right). So economics plays an important part in the decision, so altogether, it is more complicated than it seems at first.

The attached drawing assumes an EV T35 -type tweeter, of course.

The issue of surface or rear mounting of a diffraction horn (which already incorporates some of the qualities of the slot for dispersion characteristics in that plane) would benefit or be subject to excessive beaming is a matter that I can only guess at. I estimate that a recessed vertical mount would increase the horizontal dispersion as the thickness of the baffle board (being 3/4" thick at most) probably is not enough to cause any harm and could conceivably aid in the horizontal dispersion characteristics by essentially "exagerating" the slot effect slightly.

DM

[D-MAN]

Some further considerations in light of the above concerning a baffle layout for high frequency horns (in particular using a diffraction horns such as the EV T35 and its ilk). This would change if one was using a non-difraction horn such as a JBL 2404, where perhaps a central axis - surface mount (from the front) approach would then be better.

As it has been stated previously, the central axis mounting relieves the requirement of producing pairs in opposite mirror versions (one for left and one for the right). So economics plays an important part in the decision, so altogether, it is more complicated than it seems at first.

The attached drawing assumes an EV T35 -type tweeter, of course.

The issue of surface or rear mounting of a diffraction horn (which already incorporates some of the qualities of the slot for dispersion characteristics in that plane) would benefit or be subject to excessive beaming is a matter that I can only guess at. I estimate that a recessed vertical mount would increase the horizontal dispersion as the thickness of the baffle board (being 3/4" thick at most) probably is not enough to cause any harm and could conceivably aid in the horizontal dispersion characteristics by essentially "exagerating" the slot effect slightly.

DM

Dang! I misposted - meant to edit.

It is well known that a column (i.e., non-expanding sides) does not detrimentally effect the passage of a plane wave through it. Therefore, the short passage through the columnar baffle cutout from a rear-mounted horn though the width of the baffle should not have any noticable effects as long as it is uniform. Same thing for the bass horn driver with an additional mounting board between itself and the baffle (I have read that some think this is an issue - well, it's not).

To illustrate my point, this drawing shows 3 different baffle thicknesses, resulting in a longer column - the diffraction, dispersion characteristics, and overall efficiency is virtually the same in all 3.

Accordingly, it can be assumed that the more reinforced the baffle becomes, the less prone to vibration it is. Therefore, the conclusion can be drawn that (contrary to some's opinions) - separate woofer mounting boards provide a structural benefit that actually aids the propogation of sound energy by reducing baffle vibration!

DM

[D-MAN]

Extrapolating the vertical "slot" concept, it leads the issue of vertically-oriented horn throat constrictions in common use, especially in midrange horns.

However, it is also used in bass horns for exactly the same reason (increasing horizontal dispersion), here's an example, the Tattersall patent.

DM

[D-MAN]

The same thing is commonly implemented in alot of cast midrange horns, again for exactly the same reasons: Keele's EV midrange horn patent, however, in this case, the "slot" is horizontal which promotes vertical dispersion characteristics. It is even more extreme in the Smith horn (JBL).

DM

[D-MAN]

The famous Smith horn.

DM

[D-MAN]

Part 2 of Smith article.

[D-MAN]

Part 3 Smith article.

[D-MAN]

With the established fact that "squeezing" the horn throat increases midrange dispersion characteristics in bass horns due to the wavelengths involved, what do you think the results would be for this well-known bass horn, the EV "Eliminator" 2701?

I would say that it probably has some increased upper-bass as a guess.

However, it's main purpose is to simply elongate the horn channel to the full height of the enclosure (however the cross-section from the constant expansion rate along the pathway is maintained) - it just happens (as a natural by-product) to increase the midrange dispersion as it does so IMO.

DM

[D-MAN]

The University Classic and Dean also employ what appears to be a "squeeze" again to match the throat channel to the height of the enclosure, (exactly as I did in my bass horn design for the same purpose although after reading the Tattersall patent, I had expected some increase in the mid-and-upper bass range - to my suprize, there was more than I expected).

I find that this "adjustment" to the channel proportions also has an effect on the horns capability to pass upper bass frequencies as a matter of physics, and therefore, the University Classic and Dean, if given a good driver with an appropriate uper frequency corner, could expectedly have a higher crossover point limited by the horn folding which is pretty minimal in these particular designs. Technically, these are both excellent designs.

DM

[djk]

"Therefore, the short passage through the columnar baffle cutout from a rear-mounted horn though the width of the baffle should not have any noticable effects as long as it is uniform."

At low frequencies.

At high frequencies the slot is a problem with the K77 in long-axis horizontal mounting. The problem is still there with long-axis vertical mounting, but confined to the vertical plane(much less noticable than with the long-axis horizontal mounting).

[D-MAN]

The EV T35 is assumed. The propagation anomoly mentioned above is probably due to the long-axis expansion curve is subject to an abrupt discontinuity (interuption) caused by the straight walls of the cutout when rear mounted.

I would have hoped that the tendancy for vertical diffraction around the straight edge would sort of "adjust" for that to a degree, but it could be that the discontinuity at the mouth is just too disruptive.

DM

[boom3]

Looking at the Keele patent, it seems that this an application of the convergant-divergant principle of the DeLaval nozzle, which is widely used in rockets:

http://en.wikipedia.org/wiki/De_Laval_nozzle

A mathematical treatment of the design is here:

http://www.innovatia.com/Design_Center/rktprop3.htm

The Beveridge electrostatic uses a similar arrangement matching a rectangular radiator with a wide-dispersion lens

http://www.beveridge-audio.com/Technology.htm

[D-MAN]

"Therefore, the short passage through the columnar baffle cutout from a rear-mounted horn though the width of the baffle should not have any noticable effects as long as it is uniform."

At low frequencies.

At high frequencies the slot is a problem with the K77 in long-axis horizontal mounting. The problem is still there with long-axis vertical mounting, but confined to the vertical plane(much less noticable than with the long-axis horizontal mounting).

This example is different than a mounting, as the driver and horn are separated by a column. Some area for speculation exists as to what the effect on hounting a horn BEHIND the column would do, however details of a flanged duct have been made, and the outcome is well understood.

Assuming that the wavelength being passed in the column "fits" in its cross-section area, I would guess...

Here is the Adair Time-Alignment patent for long bass horns horns with horn tweeters which is a coulmn BETWEEN the driver and the horn....

DM

[D-MAN]

We are talking about a horn transmitting into a duct, somewhat like this:

The rear-mount seems likely to have some diffraction issues that the front mount (flush with baffle) would probably not.

The reactance at the mouth of the horn would seem greater for the rear-mounted version, also.

However, a baffle of 1/2" thick or less would be much more acceptable and certainly less noticable, I should think. Some more examples of "duct"/baffle waveform propagation follows. You can easily see some of the issues of propagation that will be in play.

DM

[D-MAN]

The duct (column) with and without flange (baffle)... From http://www.kettering.edu/~drussell/GMI-Acoustics/Filters.html

For a circular duct containing air at room temperature, the highest frequency at which only plane waves will propagate is given by f = 100/a where a is the radius of the duct cross-section. Once plane waves are generated inside the duct, they will propagate down the duct, even if the duct has bends or turns in it. A propagating plane wave may encounter a change in the acoustic impedance of the duct when the duct (i) opens into free space, (ii) is connected to another section of duct with a different cross-section, (iii) branches off into two ducts, or (iv) is terminated in some other way. This impedance change causes partial reflection and partial transmission of the incident plane waves.

DM

[D-MAN]

Propagation through a circular opening (however, it is not too tough to extrapolate to a rectangular opening.<?xml:namespace prefix = v ns = "urn:schemas-microsoft-com:vml" /><?xml:namespace prefix = w ns = "urn:schemas-microsoft-com:office:word" /> The issues of diffraction remain the same.

DM

[jlossint]

The horn in your picture is not an EV design. It was published in Sweden 1975 in a paper called Radio & Television. The design was originally done by me. The picture shows a hungarian replica.