Horns with good timbre and tonality?

@ei001h wrote:

"I appreciate speed and dynamics with good timbre and tonality. I know horns are good with speed and dynamics, but not sure if they can do timbre and tonality like SF can."

I’ve been designing and building horn hybrd horn loudspeakers (horn + direct radiator woofer) with timbre and tonality as a high priority, as I’m also a dealer for SoundLab fullrange electrostats, which are arguably superb in those respects.

Imo the secret to natural timbre is, getting the reflection field right. When there is a spectral discrepancy between the first-arrival sound and the subsequent reflected sound, even if the first-arrival sound is very smooth, the net result is unnatural timbre and in some cases listening fatigue. So we want the reflections to have essentially the same spectral balance as the direct sound, modified only by the room’s acoustics.

Note that with live music there is relatively little discrepancy between the direct sound and the reflections. You can step outside the room where a live piano or cello is playing and listen from beside the open doorway with no line-of-sight to the instrument and instantly tell whether it’s live or not, even though you are not getting any direct sound.

Getting the spectral balance of the reflection field correct calls for paying particular attention to the loudspeaker’s radiation pattern, and in this respect not all horns are created equal. Ime what works well is a constant-directivity horn which does not rely on diffraction. Most horns do not fit this description.

Timbral richness also benefits from having a fairly well-energized reflection field, which implies that you don’t want the horn speaker’s radiation pattern to be too narrow.

You mentioned full scale symphony. The ability to convey the hall ambience on a good recording again goes back to getting the reflection field correct, as the in-room reflections are in effect the "carriers" for the reverberation tails on the recording. I can go into more detail about this if you’d like.

Best of luck in your quest.

Duke

From a previous post of mine:

“You mentioned full scale symphony. The ability to convey the hall ambience on a good recording again goes back to getting the reflection field correct, as the in-room reflections are in effect the "carriers" for the reverberation tails on the recording. I can go into more detail about this if you’d like.”

To which @phusis replied:

“I’m certainly all ears for an elaboration here.”

In the playback room there is in effect a “competition” between these the “venue spatial cues” on the recording (whether they be real or engineered or both) and the “small room signature” cues inherent to the playback room. The goal is to make the “venue spatial signature” perceptually dominant.

At the risk of over-simplifying, the ear/brain system gets room size information from the earliest reflections; from the reverberation tails; and from the “temporal center of gravity” of the reflections. The latter refers to the interval between the first-arrival sound and the “average” of the reflection energy; the further back in time the “temporal center of gravity” is relative to the first-arrival sound, the larger the indicated room size.

We can disrupt the “small room signature” by taking advantage of the directionality of horn loudspeakers. We can toe them in aggressively such that the speaker axes criss-cross in front of the listening area. This way the radiation pattern of each speaker does not illuminate the same-side wall; rather, the first significant lateral reflection for each speaker is the long, across-the-room bounce off the opposite side wall. This also pushes the “temporal center of gravity” of the reflections back in time somewhat, further disrupting the “small room signature” of the playback room.

At the same time, we WANT plenty of spectrally-correct later-arriving reflections arriving from many different directions. We want to preserve the spectral balance of our in-room reflections by using as little absorptive treatment as we can get away with. If the spectral balance of the late reflections is preserved, the ear/brain system will detect the recording venue reverberation tails therein by recognizing the overtone patterns. The ear/brain system is actually able to detect and follow the reverberation tails down into the noise floor IF the overtone patterns are intact. But if the higher overtones (the higher frequencies) have been absorbed, the recording’s reverbration tails will not be effectively presented to the ears by the in-room reflections; they will have ceased to be "signal" and will have become "noise". Also by preserving the later reflections instead of absorbing them, we are again pushing that “temporal center of gravity” of the reflections back in time a bit.

In my own designs I often cheat and inject a bit more late-onset reflection energy via rear-firing drivers.

One of the reasons I prefer constant-directivity horns is that their off-axis energy tracks their on-axis energy well, such that the later-arriving reflections have more intact overtone patterns than if the off-axis energy had started out with the highs already rolled off.

If all goes well, the playback room’s inherent “small room signature” package of cues will have been disrupted by the early reflection arrival times presenting an inconsistent (and therefore weakened) picture, and also by the “temporal center of gravity” of the reflections indicating a larger room size than the actual room. If at the same time we have reverberation tails coming from all around which are indicating the much larger acoustic space of the recording venue, this can tip our perception in favor of the recording venue package of spatial cues. When this happens, and it is to some extent recording-dependent, the result is a “you are there” presentation wherein, with eyes closed, one has the perception of being within the acoustic space portrayed by the recording. For instance instead of the soundstage extending several feet behind speakers, it extends as deep as the recording cues indicate.

These are not the only factors in play, and the relevant principles can be applied to non-horn systems, but the typically narrower and more consistent radiation patterns of horn speakers can contribute to that elusive “you are there” presentation.

@phusis wrote:

"What about, and maybe when is the diffraction part most troublesome here; what’s inherent to - as a distinct sonic imprinting/coloration - the narrow slot section leading up to the horn widening at every volume level (i.e.: at lower levels as well), or more predominantly at higher, and perhaps only very high SPL’s? My understanding is it’s more the latter than the former, which urges the more pragmatic question of relevance in a given, domestic setting when, or rather if the issue seems to arise only at, say, +120dB levels."

Agreed, diffraction introduces a type of distortion to which the ear has a non-linear response; that is, the sonically detrimental effects of diffraction become more audible and objectionable as the SPL increases. If you’ve ever heard a PA system that sounded increasingly harsh as the volume level went up your instinct may have been that the system was distorting, but it was probably diffraction. Especially if the PA system could do this repeatedly without being destroyed, as power levels that drive speakers or amps into audible distortion can and does cause permanent damage to the speakers.

Fortunately in a home audio setting the SPL demands are not as high, and large diffraction horns that have found favor in home audio tend to be relatively "gentle" diffraction horns. Some of the bigger ones are imo quite benign even at loud home audio listening levels.

@phusis also wrote:

"I find horn size to be a factor as well, certainly with a lower crossover point and trying to maintain a fairly uniform dispersion pattern at the crossover with directivity control all the way down to the crossover. My finding here is that, generally, the larger the horn the less it sounds like a horn, and by that I mean a more relaxed, properly (i.e.: realistically) sized, dense and visceral sphere of sound. What are your thoughts on sheer horn size here?"

Yes!! There are definitely advantages to pushing the crossover point down fairly low, and there are advantages to the fairly narrow and exceptionally uniform radiation pattern of good big horns. And big horns tend to use large-format compression drivers, which ime impart a sense of ease to the music and sense of palpability (for lack of a better term) to the sound images that make it hard to go back to lesser compression drivers, let alone cones ’n’ domes.