Pics new 4-way Azura horn system, plus BIG bass horn

[atilsley]

Lynn, I'm one of those lucky people with a good size, essentially symmetrical, room with two very usable corners. I currently use Lowther All Fun horns, the smaller brother of the BigFuns, driven by flea-power SET amps.

I like what Andrew has done with the Bass/Mid horn and, as I alluded to before, I've been wondering about possibilities around a corner firing derivative; it would seem possible that by exploiting the corners the bass cutoff point could be lowered a bit, than Andrew's 70Hz, without physically taking over the room. I have little knowledge or experience of speaker design and I'm looking for a steer on how one might develop the initial concept 'on paper'? Would something like HornResp be the place to start?

Just to add a little spice to the mix I'm fascinated by the possibility of a two way by utilising the ribbon mid/treble unit that DIYHiFiSupply use in their Crescendo speaker;

Crescendo Ribbon Horn Speaker

That would require a higher crossover frequency so the mid/bass would need to be extended upwards a little too. Is that asking too much?

I appreciate that there's a bit of work (and learning) involved and it could all go nowhere.

Thanks for any pointers.

Ray

Ray, I tested my horns with 50Hz xover point (30db slope) today and no subs. The bass was still very good. You could almost have this as a great two-way system. I'll be running some recordings soon.

Oh,and this was without the large horn mouth extenion pieces added. The mouth, therefore, is 41cm x 70cm.

I've also moved the Azura's back on top of the subs...this gives a wider sound stage...and a more sensible look.

Certainly presents a massive wall of sound...very good.

 

[pooge]

At wavelengths that are several times the size of the diaphragm (say, 100 Hz to 500 Hz), direct-radiators are very close to omnidirectional, and all room modes are driven. So long as the horn is above cutoff, any horn will be several times more directional than the equivalent direct-radiator, much less energy will be directed into room boundaries, and more energy directed right at the listener. That's where some of the efficiency gain comes from ... what's sometimes called "antenna gain".

I think most of the gain comes from the resistive loading of the horn. The horn actual puts out more energy before "focussing" that energy directionally.

Take a look at the unsmoothed response of a typical direct-radiator in a domestic listening room and you will see very large response deviations in the 100 Hz to 500 Hz range ... 10 to 20 dB peak-to-dip ratios are not usual, thanks to room modes, and complex summation of first reflections off the rear, side, floor, and ceiling surfaces. In this frequency range, damping materials have limited effect, carpets do very little, and the walls have an acoustic reflectance of more than 90% (more than the mirror in your bathroom).

The wavelengths are long enough that direct-arrival and room reflections cannot be separated by the listener, and the nulls are deep enough they cannot be equalized. (Equalization can smooth out the peaks.) In effect, the room and the loudspeaker merge, with the loudspeaker acquiring all the colorations of the room. This is why small changes in location change the modal patterns, as well as the complex summation of first reflections, which then changes the sound of the loudspeaker. It's not a good sign when a 1" change in location results in a noticeable change in sound; that means the room is controlling the sound of the loudspeaker.

This is the region where increased directivity will have a strong effect on room interaction. Dipoles have a moderate directivity gain (at the expense of increased excursion compared to direct-radiators), while horns are even more directive and have better excursion control than direct-radiators.

This could be why bass horns with problems in the response (folded horns, for example) can still sound better than direct-radiators in a typical room. The response deviations of a "problem" horn are very likely not as large as what the room itself does to a direct-radiator. (That said, direct-radiator arrays have increased directivity, mostly in the vertical plane, and may also complex lobing patterns at the top of their working range.)

The rooms modes and dips likely result in differing loading of the driver, like mutual acoustic coupling between bass drivers on the same baffle. I don't have any numbers on this, but the room modes might affect the actual output of the drivers in addition to additions and subtractions in the rooms due to mode-caused peaks and dips within the room volume. A horn loaded driver will have a much stronger motor than a direct radiator, so likely will not be decelerated as much as direct radiators by any increase in loading brought about by any rooms modes. Also, the resistive load seen by a horn loaded driver is much greater than a direct radiator, due to the horn, so any load changes on the driver due to room modes would cause less of a change in the overall loading of the driver, percentage-wise.

 

[POOH]

For a corner horn, you are absolutely right. There's no difference in directivity between a direct radiator and corner horn, for the simple reason that everything is constrained to 90-degree radiation in every plane, no matter what kind of loudspeaker it is. A corner horn may be more efficient, but the directivity is exactly the same as a direct-radiator (in the same location).

That's very true. What can be realized is higher efficiency in less space and lower cost due to less drivers. My room has perfect corners made od concrete block and buried under ground. Corner bass horns below around 70 Hz are really excellent but after trying many different types i could not live with them above that frequency because of the lack of pure directional sound of the straight front horn. A straight front horn sounds very different from the corner horn - precise, fast and pure versus powerful and diffuse. Sure we all wish to get away with a 2 or 3 way but with horns it's a big compromise.

 

[Lynn Olson]

The room's modes and dips likely result in differing loading of the driver, like mutual acoustic coupling between bass drivers on the same baffle. I don't have any numbers on this, but the room modes might affect the actual output of the drivers in addition to additions and subtractions in the rooms due to mode-caused peaks and dips within the room volume. A horn loaded driver will have a much stronger motor than a direct radiator, so likely will not be decelerated as much as direct radiators by any increase in loading brought about by any rooms modes. Also, the resistive load seen by a horn loaded driver is much greater than a direct radiator, due to the horn, so any load changes on the driver due to room modes would cause less of a change in the overall loading of the driver, percentage-wise.

The effects of a horn loading on a diaphragm is usually visible in the impedance curve, particularly where the horn goes through the cutoff region, or there are ripples in the acoustic loading. There are usually corresponding ripples in the impedance curve.

If the acoustic load from a room is significantly changing the diaphragm loading, there should be a (perhaps very) small change between the impedance measured outdoors and in a room. I haven't heard of this happening, but I haven't heard of anyone making this kind of comparative measurement, either. So it's possible.

A simpler way of making a comparative measurement, short of the hassle of dragging a big loudspeaker outdoors, might be to simply measure the impedance of a loudspeaker in the middle of the room and then again in the corners. The magnified delta between the two impedance curves would be the result of room loading, if present. Any volunteers?

 

[pooge]

Might be difficult to interpret the results, as moving the horn from the center of the room to a corner or wall would change the horn loading as well. For example, since room modes find corners attractive, measuring the horn there would have another variable due to corner loading, depending on the horn and measurement conditions.