Psychoacoustics of Room Reflections

Duke LeJeune said:

[cue Hank Hill's voice]Yes sir, Ah sure can talk a good game now, Ah tell ya what.[/Hank Hill]



Depends on the specifics of course, but in general I recommend diffusion of the early reflection zones, and absorption only as a last resort. My analogy is this: What would you do if this was a room for a grand piano? Would you turn it into a padded cell, or would you want to get your money's worth in terms of that deliciously rich sound field?

For a point source... well, I think it depends on what the radiation pattern is like. If it varies a lot, for instance such that there's excess off-axis energy going out into the room at the bottom end of the tweeter's range (usually 2-4 kHz ballpark), then we may have a problem. It's hard to selectively absorb the lower treble region without over-absorbing the upper treble region. It can probably be done, but I don't know how to do it.



The grand piano analogy applies even more to the MBLs than to the SoundLabs. Bass trapping may indeed be beneficial in many if not most cases, but we don't want to alter the spectral balance of the reverberant field by absorbing the high frequencies... and of course the shorter the wavelength, the more vulnerable to absorption it is.

Note that the smaller the room, the more effective a given amount of absorption is at altering (and potentially ruining) the in-room spectral balance. This is for two reasons: First, a square yard of absorptive material is a larger percentage of the room's surface area in a small room than in a large room. Second, because the reflection path lengths are shorter in a small room, (within a given time interval) more reflections will hit (and be killed by) our square yard of absorption in the small room.

I'm harping on absorption because many people's idea of room treatment is thick slabs of egg-crate foam stuck on the walls. Very few people go overboard on diffusion or bass trapping. Since having the good fortune to work with a real professional in this area (Jeff Hedback), I have gained a great deal of respect for the disproportionate benefits of using the right material in the right amount in the right place. It's sort of like crossover design - you can't just add capacitors at random and expect to improve things. It takes the right amount of inductance, capacitance, and sometimes resistance in the right places to get really oustanding results. And when room treatment is done right, your room literally sounds twice as big as its physical dimensions. I heard that in a recording studio that Jeff Hedback designed. (I also heard that my crossover design still had a problem, which had completely eluded me up until then!)

Ah tell ya what.

Click to expand...

Duke LeJeune said:

...
Depends on the specifics of course, but in general I recommend diffusion of the early reflection zones, and absorption only as a last resort.

Click to expand...

Duke LeJeune said:

...
My analogy is this: What would you do if this was a room for a grand piano? Would you turn it into a padded cell, or would you want to get your money's worth in terms of that deliciously rich sound field?

Click to expand...

JonFo said:

Regarding diffusion / absorption for large panel dipoles, I've come to quite the opposite conclusion after more than a decade of research, measurements and a massive deployment of treatments in my system.

It all starts with identifying challenges that need attention, and with large panel speakers like Soundlabs or my MartinLogan Monoliths, it generally has to do with modal ringing effects in the mids and highs. As when one is energizing the room with more than 8 square feet of radiating material per speaker it is very easy to engage significant ringing artifacts that not only blur the sound, but actually are uncomfortable.

So while I designed my room around my speakers and therefore have no placement or room shape compromises, the results in the early days of this setup were limited by the room resonances, I could not turn it up as the high-frequency ringing was painful, and the bass modes a challenge as in any room.
Back then I leaned towards diffusion as the answer and even deployed 2D (behind the panels) and 3D diffusers in the room. While it changed the soundstage (wider but more diffuse) it still did nothing for the resonances other than shift their center points a bit.

After many tests and measurements of different products and playing with placement, I arrived at a setup that actually has a good bit of absorption, especially for the rear-wave of the dipoles in the front of the room. It also has a fair amount of diffusion, mostly in the rear of the room.
Full details of the acoustic treatments (45 commercial + 2 large scale custom treatments) are in this thread at the MartinLogan Owners forum.

The result has been an incredibly wide, deep and focused soundstage, such that one can hear instruments well behind/outside the plane of the speakers or anywhere between the listener and the speaker. Most planar setups have muddled imaging, it seems huge, but is not precise. This setup has inch-perfect sound placement with surround music and movies.

The other significant achievement is the taming of most of the ringing issues, as I can now play at up to 105dB levels without major modal buildup. It’s pretty amazing to hear a system that does not change ‘character’ as one turns it up. A consistent in-room power-envelope is a sign of good speakers in a well treated room.


It’s interesting that you’d choose a grand piano as the analogy, as the ability to resolve a piano accurately is what brought me into the MartinLogan line. And that’s because I grew up in a household with a grand piano played by a Julliard grad (dad) concert pianist who would practice endlessly on it.
For many years, that piano was in a large room with cement walls, tile floors and minimal furnishings, and my god was that painful at times. Would have killed for some judiciously applied absorption and diffusion, as the modal ringing in the mids and highs were extreme (he played loud stuff like Rachmaninoff & Liszt).
Finally, I can play those same pieces from high-rez SACDs and DVD-Audio discs on my system at natural levels and enjoy the music to its fullest without the room marring the sound.

So I believe the trick for large panel speakers is managing the rear-wave of dipoles and getting the spectral and decay balance in the room right with absorption, then use diffusion for imaging tuning.

Click to expand...

JonFo said:

Well, even with the speakers moved 5 feet into the room, all that changes is the rear wave propagation delays and likely places the prime listening position closer and receiving more of the direct (front wave) sound. So soundstage perception definitely changes, but still does zero to affect room modes and HF ringing. Only room treatments can solve those challenges.

Click to expand...

JonFo said:

Large panel speaker manufacturers make a design decision, but not one that I necessarily agree with anymore, they call dipole a 'feature', I call it a 'bug'. If it really were desirable, then most 'high-end' speakers would use that topology, but they don't.

Click to expand...