Acoustical Treatments Master Thread - Page 9

A right tricorner is a reflector as well (that's how bicycle reflectors are made)...it has the characteristic of reflecting something (light, HF sound) back at the same angle it originated from.

Chris,

Not in the Navy but work for the Navy as a civil servant. The place I purchased my ductboard from is actually in Kent and the business name is Paragon Pacific. They sell to individuals but are really set up to sell in large quantities to contractors etc. It is kind of a pain to get one or two sheets (mostly for the warehouse guys). They are friendly and helpfull though. One sheet of 2"X4'X10' rigid liner board with tough guard on the face was about $93 with tax. If I thought there would be much improvement by going all the way to the ceiling, I might go back and buy another sheet on Jan 6 because I will be taking that day off with leave I must take or lose.

That's a tall order, but I'll try.

A minimum phase filter is one which alters the phase delay of a signal through it down to its theoretical minimum. There are implications for stability of the filter (its response is guaranteed to be stable and not blow up over time) as well as causality (no sound through the filter can create an earlier effect in time, which is, believe it or not, not true for every possible filter).

Every linear filter (one for which 2X the input produces 2X the output) can be decomposed into two different filters: a minimum phase filter and what's termed an all-pass filter. The latter introduces a delay across all frequencies. One example of a process which cannot be modeled without an all-pass filter is slap-echo, or any other discrete echo.

But the cool thing about minimum phase filters is that any particular one always has an inverse filter, which is also minimum phase. This means that you can cascade a minimum phase filter with its inverse (in either order), and completely undo the filtering process.

So what has all this got to do with room modes? Simply that the resonance of a single room mode, the back and forth wall reflection and amplification at a particular frequency, is minimum phase. So that means it can be completely canceled by its inverse filter. You can readily make such an inverse filter using an off-the-shelf parametric equalizer. You just dial in the frequency, bandwidth, and gain which reverses the effect of the room mode, and you have zapped it. And it doesn't matter that the inverse filter of the EQ came BEFORE the room resonance. The minimum phase principle doesn't care about what comes before what.

Hope this helps,
Terry

Guys, I have a bit of a situation in my new room. Tonight, I finished hanging all the rigid fiberglass. I'm using CertainTeed UltraGold Duct Board. My room is 26'L x 14'W 9'H (in center). There are two soffits on each side of the room. They extend 2ft into the room and down 1ft. I have covered the entire front wall from floor to ceiling. I have also extended the floor to ceiling treatment out three feet on each side wall. The rest of the room is covered from floor up to 4ft. I clapped my hands and was horrified to hear that awful concrete echo. My room has three exterior/underground walls that are concrete (front, left and right). The right concrete wall is separated by an airspace (alley) that is 26" wide and then a 2x6 wall with drywall that forms the HT interior wall. The left wall has a much smaller airspace of about 2" and is framed with 2x4. The front wall is also similar to the left. The rear wall is an interior wall and has a room on the other side. It is framed with 2x6 studs. ALL walls have R19 batts and one layer of drywall (damn!). The ceiling also has insulation. Could this echo be caused by the soffit?

My fiberboard properties can be seen here:
http://www.bobgolds.com/AbsorptionCoefficients.htm

Yes, but a center front subwoofer position cannot excite this mode. Yet there was a significant null in the corresponding frequency response.

jasplat88:
There's no axial mode for that room near 56hz, but there is a tangental mode at 55.8hz (1,1,0).

First few modes (axial, tangental and oblique): 24, 47.3, 50.4, 55.8, 69.6, 71.3, 78.6, 82.2, 87.8, 92, 93.4, 95.9, 96.4, 100.9, 103.7, 105.0, 106.5, 108.3, 111.7

OK, I went back and read a little about right tri-corners. I think I definitely need to treat the bottom of my soffits. On the other hand, it's quite different if I clap or whistle while seated in my chair...there is virtually no echo. I guess that's why you treat up to ear level while seated. Maybe, I'll only treat the soffit area in front of the seating area.

jasplat88:
If you get bored and find yourself with nothing to do, what happens with the subwoofer/mic at these locations:
- subwoofer at (6', 2'6.6", 1') and the microphone at (13'3.1"", 7', 3'8.9")
- subwoofer at (5'11.9, 1'11.9", 1') and the microphone at (12'4.7"", 7', 3'8.9")
( RPG Room Optimizer, 20hz to 80hz)

Hey Bob,

I will do this. Probably tomorrow (Monday) night and report back. I assume these are locations off the BACK wall correct? Read....for example in your first measurement (6' from the back wall, 2'6.6" from either side wall, 1' off the floor)? Just want to make sure before I get started. Thanks!

-Jason

jasplat88:
The program assumes a symetrical/rectangular room. So they are measurments off either the back or the front wall, as you prefer. They are all (23'6" dimension, 11'2" dimension, 7'2" dimension). I would have tried them from (front, left, floor), because then the microphone positions seem about right for a chair. I wouldn't swap start points between sub and microphone (i.e. sub measured from back and microphone measured from front -- would be bad).

Leon:

I paid about 120 bucks for two entire batts of 703-type fiberboard, so that sounds like a very high price...?

Jeff,

Did you do the batting on the 'untreated' area or is it still plain drywall?

Once again, asking whether it matters or not to room acoustics to use a perforated screen or not -- is the decision solely based on whether there's a speaker behind the screen?

I've treated screen wall with 4" 703 mounted 4" off the wall.

Speakers are out in the room, NOT behind the screen.

Would a perforated screen (mounted approx 4" in front of the 703) improve overall sound quality? Seems like it should, as it eliminates a largely reflective surface over a large fraction of the front wall. On the other hand, seems like most frequencies can find themselves around the screen anyway.

ps -- if answer is "it depends": room is 18' by 15' by 12'; first reflections off the side walls are being handled by 2" RPG binary diffusor panels or 7" skylines (will decide by ear); rear wall ceiling and screen wall corners all have soffits with bass trapping.

Brucemck2:

If you have a perforated screen, then by definition a lot of sound goes through it virtually without change (well they attenuate high frequencies a bit). If it's not perferated I've always assumed that it reflects highs but lows go right through it, but I've never seen measurements so I can't define 'highs' and 'lows'.
With the exception of first reflections, it makes less (none?) difference where absorbtion goes, as opposed to the absorbtive surface area and thickness (Sabine formula).

If your "Speakers are out in the room, NOT behind the screen", then I wouldn't get a perferated screen. The problem with perferated screens is moire, which is an visual interference pattern between the hole spacing of the perferations and the projector grid pattern. The second problem with perferated screens is you loose 10% of your brightness.

In my case I want a big screen in a short room, so I can't put my center above or below my screen, so it has to be perferated, so all the speakers are going behind it, along with, probably, a bunch of absorbtion.

Terry,

> Bill is right on both points <<br />
He usually is!

> inversion of the low frequency room modal response can be done with extremely simple EQ <<br />
Yes, but that's not what I was talking about!

I said, "Just as important is the way they reduce ringing, and in a way that EQ cannot." I didn't say anything about LF response, just ringing. I've heard from you and Dennis and others I respect that EQ can indeed help to tame the lowest modal boosts, and I accept that. Personally I wouldn't add an EQ to my system, but I understand why others do. Especially those who for whatever reason don't want to install a lot of bass traps.

--Ethan

Jeff,

Your question seemed to have gotten lost among all the chit-chat over phase response and axial modes.

> Could this echo be caused by the soffit? <<br />
Echoes are caused by large areas of reflective surface, wherever they may be located. If two reflective surfaces are opposite each other - ie: parallel walls, or a reflective floor and ceiling - then you get a special case known as flutter echo. Flutter echoes have a specific pitch whose frequency is related to the spacing between the surfaces. So it sounds more like a "boing" than like normal "hello .. hello" echoes.

--Ethan

bpape and Ethan,
First, I didn't want to add batting to the upper section because from all I've read, it isn't necessary and is only used if one wishes to add fabric to the upper half of the wall. We wanted to do some special paint treatments to the upper half.

As for the echo, it is only bad in a couple of spots. I'm pretty sure it is being caused by the soffit. I'm going to add some additional fiberglass to that area.

If I were to use a parametric Eq to correct low frequencies (in my case 90Hz for example) on a full-range speaker, does it introduce phase errors across the entire spectrum or only around the center frequencies I am correcting?

Andy K.

Chris,

I am not sure when you bought yours but the price has increased quite a bit since January of this year. How thick was your 703 and how big is a bat? I could have purchased the 1"X10'X4' boards for less than half the price of the 2" ductboard per piece.

it was 1-inch thick boards, 16 per batt. I paid like 130 for two batts, more than enough for the whole room.

So if I understand the info on this page. I could actually just put R-11 or R19 unfaces batts on the front wall, cover it will GOM or something similar and it would have the same effect as using something like 703, JM or other rigid fiberglass?

If you look at the absorption coefficients, you'll see that R11 has signifigantly more absorption at some frequencies than 703:

125 250 500 1000 2000 4000 NRC
703: 0.11 0.28 0.68 0.90 0.93 0.96 0.70
R11: 0.34 0.85 1.09 0.97 0.97 1.12 0.95

Now here's the part I'm not sure about, because I was thinking of using R11 as well. The vast majority of my front wall is taken up by the screen. The screen is reflective to certain frequencies (I'm guessing > 500 hz or so...). Since most of the insulation will be behind the screen, does the screen serve to "counter" the extra absorption of the R11. So in practice, perhaps I can use R11 behind the screen, and then save the $$ 703 panels for elsewhere in the room.

But I won't know w/o measuring, and I unfortunately don't have any measuring equipment yet (I'm an all Macintosh/Linux shop at home, still need to build my Windows HTPC and even then, I'd need to invest the mic and whatever else is needed to measure).

j.

Having the R11 behind the screen will give you the absorbtion from the lower mids down while the screen will reflect the highs.

You CAN use R11 everywhere IF that is what your room needs and if you have a way to hide it. It doesn't tend to be as easy to make 'visually friendly' as the 703 @ 1".

Andy,

> If I were to use a parametric Eq to correct low frequencies (in my case 90Hz for example) on a full-range speaker, does it introduce phase errors across the entire spectrum or only around the center frequencies I am correcting? <<br />
A lot of people don't realize this but phase shift is totally benign. The only time phase shift has an impact on audio quality is when two versions of the same signal are combined and one of those has been shifted. So if you add an EQ to your system and measure some frequency response, what you measure is what you have and any accompanying phase change at any frequency is irrelevant. (Assuming the same phase shift is applied to all channels.)

--Ethan

Ethan:

I thought there were several types of phase shift (time delay > 1ms):
a) equal delay on all speakers, but may be out of sync with the video.
b) different delay between two (or more) speakers -- as you mentioned
c) different delay for different freqeuencies - caused by bad EQ electronics, or bad crossover electronics, or bad bass management electronics, or a bad filter.
d) different delay because some speakers are further away from the listener than others (always happens, think 'two' listeners)
e) multi-driver (single speaker) time delay because if you're head is higher than the speaker then likely the tweeter is closer to you than the woofer
f) subwoofer phase shift control knob (really useful for modally placed subs that are tied to the front left and right channels using high-low filter hardware in the subwoofer), or for a center placed subwoofer.

Okay back to tri corners for a moment please. I will have a trayed ceiling in my room so I will have soffit on all walls. Do I treat the corners in the facial area of the soffit as well, they are 11" high? Also is the treatment at 45 degrees on all 3 axies for any tri corner treatment?

When you deal with a tri-corner, you are dealing with the intersection of length width and height. Treat the soffit tri-corner as the real one. The other one (depending on how wide the soffit is) is generally pretty far away from the length dimension.

Ok Bob,

I finally got to it....and wow....pretty impressive. Now what? The sub is off my front stage at both those locations, and the mic height is about a foot higher than where a head will be (although the seating location is almost spot on----from my front row prime viewing seat).

-Jason

A phase change is introduced on both sides of the frequency you are correcting.

It may be argued that this is not a phase error, but a phase correction. If the result of parametric equalization is a flatter spectrum, then the "distorted" spectrum of the room before EQ can be regarded as having incorrect phase, which the equalizer corrects. The resulting signal phase after parametric EQ will be the same as if the offending room modes weren't present to begin with.

Thanks Terry and Ethan. So, can you tell me why EQ is considered a last resort then? I agree that it doesnt solve the 'ringing' and long decay of problem frequencies, but I've always always read that EQ causes phase distortions which lead to a 'smearing' of the sound.

Andy K.