Help with first reflection "mirror trick"

[cruzmisl]

Hi All,
I am setting up my acoustic panels tomorrow and was wondering about the mirror test. My room is an odd shape and curious if the mirror should be held flat against the wall or 90 degrees to the wall. It may not matter but I had to ask. If I lay it flat on the wall along one wall I won't be able to see the mirror.

Thanks,
Joe

[Tonto]

The 1st order reflections are the path the sound waves travel from the speaker to the side wall, which then bounce off & travel to your seating position. If you can't see thespeaker in the mirror, those reflections aren't reaching you. I believe you should position the mirror flat on the wall such that you can see the speaker in the mirror. That should be where the panel needs to be--not an expert, just uneducated reasoning.

[Wayne A. Pflughaupt]

At the seating position, the mirror should be placed to where you can see the speaker in the reflection. So, it should be flat against the wall.

That said, I don’t take a lot of stock in the mirror trick. The reason is that sound doesn’t radiate from the speaker like a laser or flashlight beam. It radiates in a wide, dispersed pattern, progressively more so the lower the frequencies go. So your panels can be several feet in front of or behind the “optimal” mirror location and I doubt you will be able to discern any audible difference.

But as always, experimentation is never a bad thing.

Regards,
Wayne

[cruzmisl]

Thanks Wayne. Bryan P and I are working to get the reflection points sorted out via email. Wow, is he patient!

I've read somewhere that absorption with a cloth over it is not so good absorbent off axis. IOW, this is maybe not so good idea, if you want to eliminate first reflections. What about putting a perpendicular baffle on side walls? -- Mario Petrinovich

[JCD]

Wayne A. Pflughaupt wrote:

That said, I don’t take a lot of stock in the mirror trick. The reason is that sound doesn’t radiate from the speaker like a laser or flashlight beam. It radiates in a wide, dispersed pattern, progressively more so the lower the frequencies go. So your panels can be several feet in front of or behind the “optimal” mirror location and I doubt you will be able to discern any audible difference.

I'm not sure I completely get what you're saying here..

I understand the sound comes out of the speaker in an expanding "bubble" -- but the sound is going to hit you in one of two ways. The first is where your ears "pierce" the bubble. The second is where the sound wave reflects off of some object, the side walls being the most prevalent. That first side reflection point would be where you see the speaker with the mirror. Any sound that hits before or after that point will be reflected too far to the right or left of your head. This of course only affects the first reflection point, sound is going to be bouncing around the room like a pinball, but this trick does catch the earliest reflection which is the most important. The way I understand is that if the same sound wave his you within a certain amount of time, the brain combines the two waves and therefore muddies the sound. However, if there is enough of a time gap, the brain can distinguish between the two sound waves which won't come through as muddy. I'm not very eloquent with this explanation, but I think there's enough there to get the idea. Hopefully one of the guru's will chime in and correct something if I've got it wrong.

JCD

[Ethan Winer]

JCD wrote:

Hopefully one of the guru's will chime in and correct something if I've got it wrong.

No, you got it exactly right. That speakers have a "radiation pattern" is irrelevant here because whatever sound does strike the wall will reflect off the wall more or less like a cue ball on a pool table.

--Ethan

[Bob_99]

Ethan,

In regard to the first reflection point, I have read that any reflected sound that arrives within 15 to 30ms (depending on whose writing) is not resolved by the brain (normal brain). I'm by no means suggesting not treating first reflections but what are your thoughts on this. Also, how much lower in dB level does a reflection have to be before it not important. Again, I've read from 15 to 30 dB.

I now understand more fully why you hesitate to recommend books on acoustics.

Bob

[91BlckGT]

Not to thread-jack, but if one were to consider the "cue-ball effect" of the sound bouncing off the sides, if you have only a single 2' by 4' panel to place on the sides, wouldn't it make more sense to place the panel lengthwise, so that a multitude of seating positions would have first order reflection stifled?

The downside to this is that you have a rather narrow 2' height to work with. Considering that most theaters are set up with the mid/high axis at ear level, this doesn't sound too bad to me, if you center the panel at the speaker / ear mid-point.

If the radial pattern of speakers matter, then one would need to be install a 4' by 6' or greater panel for first order reflections, right? I generally don't see installations doing that.

Reason I'm asking is that I'm going to be building 4 2'x4' traps, and if I could get by with only one on each side, that saves me the other two for other uses.

[Wayne A. Pflughaupt]

JCD wrote:

I'm not sure I completely get what you're saying here.

I understand the sound comes out of the speaker in an expanding "bubble" -- but the sound is going to hit you in one of two ways. The first is where your ears "pierce" the bubble. The second is where the sound wave reflects off of some object, the side walls being the most prevalent. That first side reflection point would be where you see the speaker with the mirror. Any sound that hits before or after that point will be reflected too far to the right or left of your head.

Makes sense I guess, but the problem I have with that idea is that it assumes the reflected sound arriving in front of or behind your head is not going to be heard. Of course that’s just not possible.

And if we’re only interested in treating the sound that’s arriving precisely at the ear, from the mirrored cue-ball reflection point, then the acoustical panel would only need to be a few inches square. After all, if it’s any bigger, it’s soaking up sound that would be hitting in front of and behind us, which we don’t care about... ???

Obviously, more treatment is typically better, not less. I assume we use treatment panels that are large enough to absorb reflected sound arriving in front of and behind the ears because that sound matters, too.

But – experimentation is always best. If it sounds best placing the panels where the mirror “says,” vs. a foot or two one way or the other, then by all means go for it.

Regards,
Wayne

[JCD]

91BlckGT wrote:

Not to thread-jack, but if one were to consider the "cue-ball effect" of the sound bouncing off the sides, if you have only a single 2' by 4' panel to place on the sides, wouldn't it make more sense to place the panel lengthwise, so that a multitude of seating positions would have first order reflection stifled?

The downside to this is that you have a rather narrow 2' height to work with. Considering that most theaters are set up with the mid/high axis at ear level, this doesn't sound too bad to me, if you center the panel at the speaker / ear mid-point.

Are you saying to put position these as "landscape" vs. "portrait"? I think the reason for the "portrait" is because you're blocking the output of more than just one driver. You make a good point though -- and I think they'd tell you to use two panels in this situation.

91BlckGT wrote:

If the radial pattern of speakers matter, then one would need to be install a 4' by 6' or greater panel for first order reflections, right? I generally don't see installations doing that.

I'm not sure how this one relates.

91BlckGT wrote:

Reason I'm asking is that I'm going to be building 4 2'x4' traps, and if I could get by with only one on each side, that saves me the other two for other uses.

I think there is something about bass traps that makes them different. I get the feeling it's, for lack of a better word, more like "trapping" (i.e., bass trap) the bass rather than absorbing it like the panels for first reflections are supposed to do.

JCD

[JCD]

Wayne A. Pflughaupt wrote:

Makes sense I guess, but the problem I have with that idea is that it assumes the reflected sound arriving in front of or behind your head is not going to be heard. Of course that’s just not possible.

It will be heard, but only after it's reflected off some other walls.. By then, the assumption is the "damage" that it could have done to the imaging has been negated or minimized. I'd also say a truly properly treated room is going to have more than 2 panels for accoustical treatments, it's just that the first reflections off the side walls are going to be the most "damaging".

Wayne A. Pflughaupt wrote:

And if we’re only interested in treating the sound that’s arriving precisely at the ear, from the mirrored cue-ball reflection point, then the acoustical panel would only need to be a few inches square. After all, if it’s any bigger, it’s soaking up sound that would be hitting in front of and behind us, which we don’t care about... ???


Obviously, more treatment is typically better, not less. I assume we use treatment panels that are large enough to absorb reflected sound arriving in front of and behind the ears because that sound matters, too.

This statement I think starts to wander in to the graduate level area -- which I am SO not equiped to answer properly. Of course, I'll try anyway.

Well, there is more than one driver in the speaker, so you want a panel that can take care of more than just the tweeter. The other issue is that we would want the decay within the room to be relatively short. Not that we'd want to have a room that completely absorbs all sound (I've been in an over treated room -- it sounds awful), just more than the average room.

I think the current preferred method is a dead (read absorption) front of the room and a live (read diffusion) back of the room.

Wayne A. Pflughaupt wrote:

But – experimentation is always best. If it sounds best placing the panels where the mirror “says,” vs. a foot or two one way or the other, then by all means go for it.

Couldn't agree with you more!

JCD

[Ethan Winer]

Bob_99 wrote:

I have read that any reflected sound that arrives within 15 to 30ms (depending on whose writing) is not resolved by the brain (normal brain). I'm by no means suggesting not treating first reflections but what are your thoughts on this.

Echoes that arrive quickly are not heard as echoes, but that doesn't mean they have no damaging and very audible effects! In this case, instead of hearing an echo you notice a hollow sound caused by the numerous comb filtering peaks and nulls. There's a bunch more on the RealTraps site about this, including a video. The image below is from my RFZ article.

Also, how much lower in dB level does a reflection have to be before it not important. Again, I've read from 15 to 30 dB.

It depends on how you define "important." The normal rule of thumb is 10 dB minimum, but I'd aim for at least 15 dB.

--Ethan

[Bob_99]

Ethan,

As always, many thanks for the explanation. Your answer explains quite a bit and I am sincerely grateful for your time.

Bob