Hello,
In general, is it better to apply acoustic (say 703) panels on the top half or bottom half of a wall in a HT room?

I'm sure that the full wall is preferred but I would like to know if there is an optimal location. My feeling is the top 1/2 but I think I've read that the panels should be on the lower 1/2.

Anybody?
Thanks,
Doug

Hello,
In general, is it better to apply acoustic (say 703) panels on the top half or bottom half of a wall in a HT room?

I'm sure that the full wall is preferred but I would like to know if there is an optimal location. My feeling is the top 1/2 but I think I've read that the panels should be on the lower 1/2.

Anybody?
Thanks,
Doug
Actually, the middle half. :) This will assure that side wall early reflections are covered.

- Terry

They should be applied from ear level down, which is the bottom half. I've heard some people say that you should also do the top half from the front of the room to the first reflection point, which is what I did.

They should be applied from ear level down, which is the bottom half.
If any of your front speaker tweeters are above ear level, this oft quoted "rule of thumb" does not work.

- Terry

If any of your front speaker tweeters are above ear level, this oft quoted "rule of thumb" does not work.

- Terry


So, it appears that two factors are involved: where the sound is coming from height wise from the speaker, and your ear height relative to the sound bouncing off walls. If a front speaker was mounted 1 foot from the ceiling then having 'high' mounted panels would be important, but that would only partially solve the issue because the ear would/might be getting reflections off a reflective wall. Does this, ahem, 'sound' right?

If that is the case , is the optimal solution ear height speakers and panels that are basically at the midpoint of one's ear (1/2 above and 1/2 below)?

If any of your front speaker tweeters are above ear level, this oft quoted "rule of thumb" does not work.

- Terry

True, this was assuming the speakers were placed at the proper height.

So, it appears that two factors are involved: where the sound is coming from height wise from the speaker, and your ear height relative to the sound bouncing off walls. If a front speaker was mounted 1 foot from the ceiling then having 'high' mounted panels would be important, but that would only partially solve the issue because the ear would/might be getting reflections off a reflective wall. Does this, ahem, 'sound' right?

If that is the case , is the optimal solution ear height speakers and panels that are basically at the midpoint of one's ear (1/2 above and 1/2 below)?
There are other factors that make for an optimal solution. How much overall absorption do you need? How much diffusion? What is the directivity of the speakers? Where and how large is the listening area? How important is it to make the sound quality uniformly good over the entire listening area vs. just the prime seats? If the room is for music listening too, how much and what type of music?

This is what custom home theater acoustics is all about! I only have the "optimal" solution after I have built a 3D model of the room, run several test cases with my ray tracing software, plus applied a lot of experience to the problem.

It is why all "rules of thumb" have limited utility. They work in many situations, but not all.

- Terry

I just installed OC703 panels at the first reflection points (bottom half). The treatments were minimal (five 2x4 panels). I'm amazed at the audible difference.

Thanks.

Sounds like you can't go wrong by buying some 703 panels and sitting in the viewing area and just testing it out.

In a general set up where are the usual 'first reflection' points?

-Doug

Doug,

> In general, is it better to apply acoustic (say 703) panels on the top half or bottom half of a wall in a HT room? <

I agree with Terry that the middle matters most. But if you'll allow me to go into editorial mode for a moment...

I'm not sure where the idea of treating only the bottom of the walls first started, but it's not a good way to treat a room. Especially when the fiberglass is only an inch thick which unfortunately is also common. My guess is this started when someone saw carpet on the walls of a movie theater, put there mainly to keep kids from dinging up the walls, and figured they'd copy the same idea for home use.

What matters most in a domestic size room is having enough bass trapping in the corners to flatten the low frequency response and minimize ringing, and absorbing all early reflections which damage imaging and harm clarity generally.

A good general goal with acoustic treatment is to spread it around the room evenly, versus putting it all on one surface, or all on only the top or bottom of the walls.

--Ethan

That seems to make the most sense. Thanks.

Ethan, I *believe* (though am not 100% sure) that this methodology was adopted because this is the way Dennis Erskine designs his HT rooms.

The important part that is missing is that Dennis follows this treatment up with tuned resonant panel traps to target the bass modes. A lay person does not usually have the tools to design these traps and so end up implementing a partial solution.

I agree with you - lining the room with thin, flimsy 1" Linacoustic does absolutely nothing for the bass frequencies, which are at least as important as first reflections if not moreso.

Andy K.

I'm pretty sure Dennis has also said that 1st reflections are not ALWAYS a bad thing too. Depending on the off axis response of the speakers and the timing and volume of the reflections, they can widen the front sound stage and add spaciousness without adversely affecting imaging. I'd personally like to hear this talked about a little more....

Dan

Reflections in general that have a >20ms delay before reaching your ear are subjectively pleasing because they fill out the sound and add ambience.

First reflections (ie. reflections that bounce off only one surface before reaching your ear), unless your room is very large, will typicially be much less than 20ms of delay and so should be cancelled by treatments.

If you have a very directional speaker, its possible the first reflections will be less of an issue, but I dont think you could ever say they would be desirable.

Andy K.

I generally use "early reflections" instead of "1st reflections." This is the more commonly used term in acoustics.

Here's the EBU Technical document 3276 definition:

2.2. Early reflections
Early reflections are defined as reflections from boundary surfaces or other surfaces in the room which reach the listening area within the first 15 ms after the arrival of the direct sound. The levels of these reflections should be at least 10 dB below the level of the direct sound for all frequencies in the range 1 kHz to 8 kHz.

The "10 dB down" is a good rule of thumb, though that's a simplification of a complex psychoacoustic phenomenon. Alpha Certification® uses a more precise metric.

Weak early reflections (or later reflections), the kind provided by diffusive surfaces or sufficiently attenuated off-axis speaker response, can definitely enhance the spaciousness of the sound stage. You don't necessarily want to kill all such reflections. But it is hard to completely kill off anything (darn dB log scale! :) ), so this is often not an issue.

- Terry

High, low, middle of wall?? Well pull out a mirror, sit in your listening spot and have someone run the mirror down each side wall. When you can see the speaker then that is where the panels would go.

Glenn

Andy,

> I *believe* (though am not 100% sure) that this methodology was adopted because this is the way Dennis Erskine designs his HT rooms. <

I always assumed it started longer ago than that. People go to a real movie theater and see that style of treatment done with carpet, so they assume it must be good acoustically. But the carpet is there just to keep little kids from scuffing up the walls!

> The important part that is missing is that Dennis follows this treatment up with tuned resonant panel traps to target the bass modes. <

In this case I'm considering early/first reflections only. I tell my customers to center reflection panels at the mirror point, so the coverage area extends beyind that pinpoint in all directions. Most sound waves don't travel like a laser pointer beam. :D

--Ethan

Dan,

> I'm pretty sure Dennis has also said that 1st reflections are not ALWAYS a bad thing too. <

I've heard that theory but I've never seen a case where early reflections do anything but harm the sound. I have two completely different systems in my home, and both use very high quality professional studio monitors with a very good off-axis response. My HT room has Mackie 624s and my home studio has a pair of huge JBL 4403 loudspeakers with an even better off-axis response than the Mackies. In both cases the improvement in imaging and clarity with RFZ absorption is obvious and undeniable. I even use that as a demo of the importance of RFZ treatment. I'll have a visitor step back a few feet (or behind the couch in my living room) and notice that the sound is clearly coming from the loudspeakers. Then I have them step forward into the Reflection-Free Zone, and all of a sudden the sound is coming from a very wide and clear phantom image.

Now, all that said, I don't know what speakers Dennis uses and recommends, and even if I did I've probably never heard them in person.

--Ethan

Terry,

> I generally use "early reflections" instead of "1st reflections." <

Me too, though sometimes I forget. :D

> Here's the EBU Technical document 3276 definition: <

Yes, early reflections are those that arrive within some short time window. I've seen some people say as long as 25 milliseconds. I think it also depends on frequency.

As for the distinction between early and first, any reflection that arrives within that window is early, but for me it's a "first" reflection only if it's on the side walls, floor, or ceiling. A lot of people sit too close to the wall behind them, so those reflections are definitely early and cause terrible comb filtering. But that doesn't seem like a first reflection point to me. Or maybe it is because it's still the first room boundary that's hit.

Likewise, when we talk about treating the front wall, that's mostly a secondary reflection point, after the sound passed your head, hit the wall behind you, then made it back to the front wall, then finally gets back to you ears. So the front wall is a second (or third?) reflection point, though by then the sound is not early unless it's a 4 by 6 foot bathroom.

> The "10 dB down" is a good rule of thumb, though that's a simplification of a complex psychoacoustic phenomenon. Alpha Certification® uses a more precise metric. <

Not to contradict you, but this is not complex at all. The 10 dB rule derives from ancient recording engineer's practice, and is often called the "three-to-one" rule*. The basis for this rule gets us to the real issue with early reflections - comb filtering. It's not phase shift or "time smearing" etc, but simple comb filtering. When the reflected sound is 10 dB or more below the direct sound, the comb filtering peaks and nulls are not as objectionable.

--Ethan

* The three-to-one rule says that if you have, for example, a guitar player sitting next to a banjo player (no banjo jokes please), the microphone meant for the banjo should be at least three times farther from the guitar. And vice versa. Assuming the instruments have a similar acoustic volume level, being three times farther makes the guitar sound in the banjo's microphone 10 dB below the sound from the guitar's microphone. When both microphones are then mixed together, the comb filtering from the delayed version is less severe.

Ethan,

You are certainly correct about the fundamental connection between perception of comb filtering from early reflections and the problem of sound localization in the presence of early reflections. Sound localization at low frequencies is controlled by interaural time difference (ITD), while at high frequencies it relies on interaural intensity difference (IID). IID for high frequencies must be significantly cued by comb filtering. Sound at 2 kHz and above is critical to the sensitivity of the image shift phenomenon of early reflections. If you band limit the early reflections to lower than this frequency, the image shifting threshold gets worse.

As usual, the devil is in the details. :) Experiments have shown the detection threshold for sound coloration from early reflections to be consistently lower than that for spatial shifting effects. And they both change with the type of sound and the number of milliseconds of delay. However the 10 dB down rule for early reflections and the 3-to-1 rule (9 dB down) to avoid comb filtering derive from the same principle, and are both quite serviceable under most conditions.

Once again, we agree much more than we disagree. :D

- Terry

Excellent Terry.

test