The No.1 Website for Pro Audio
RT60 big differences for different frequencies REW
Old 7th December 2019
  #1
Here for the gear
 

RT60 big differences for different frequencies REW

Dear All,
I am using REW 5.1 on Mac El Capitan with a Scarlett 2nd gen 2i4 sound card and a Dayton Mic, both of them with a cal file.
I am measuring RT60 on an empty room that is about 58K2 cubic feet. ( 1650 m3 ). I used two speakers and one sub to generate the sweep. I tried sweep ranges from 256 to 1M on REW. I used different mic positions and different speakers positions and pointing. All the frequency responses and all the RT60 graphs on REW are very similar among them, and all of them have a big peak of about 2s. on 100Hz and 2KHz. but only near 1s. from 200Hz to 1KHz. My manual calculations are near 1s. for the space. Materials on the walls are glass and concrete. The ceiling has a perforated acoustic cover. Floor is concrete also. Has chairs and tables. All of those materials have no big absorption variations on frequency, so I guess RT60 should be very similar over all frequencies. Question is, is it RT60 valid at 100Hz and 2KHz? is this real or is some kind of mistake or issue? May I trust the whole curve from 60Hz to 8KHz with this huge peaks or should I just believe on the 200Hz-1KHz range?
Thanks in advance for your feedback.
My best regards,
Gerardo Ruiz.
Attached Thumbnails
RT60 big differences for different frequencies REW-rew-rt60-strange-graph.jpg  
Old 10th December 2019
  #2
Here for the gear
 

No one? really? help me pleeeease!

I just want to hear your thoughts even if you are not 100% sure about them...
Old 10th December 2019
  #3
Here for the gear
 
PJAcoustics's Avatar
 

Quote:
Originally Posted by gerardochile View Post
Dear All,
All of those materials have no big absorption variations on frequency, so I guess RT60 should be very similar over all frequencies.
Well your low absorption materials like concrete and glass don't have much variance over frequency, but the more absorbing materials do. For example, the perforated acoustic ceiling, could have an absorption coefficient in the region of 0.7 - 0.9 between 500 and 4kHz, whereas at 100 Hz, it may only be around 0.1 to 0.2.

I'm not too suprised by the 'spike' at 100 Hz, this is typical in most rooms. Low frequency is very difficult to absorb. Most surface finishes, including 'acoustic' products, have a poor low frequency performance when they are flush mounted onto a wall or soffit. What low frequency needs, is something thick, with an air gap, because it has a long wavelength. This is why many recording studios use bass traps (you don't see these in most other types of room, because 100 Hz just isn't as important for non-performance/music spaces). The bass trap normally has a depth which is equal to 1/4 the length of the wavelength you want to control. For 100 Hz, the wavelength is 3.43m, therefore the bass trap needs to be around 0.85m long. . You won't be getting this from any of your finishes (the closest you'll get is from the ceiling as this has an air gap above it, but probably not as much as 0.85m).

Also be aware that low frequency measurements of RT (<200 Hz) varies significantly as you move around the room, due to the room modes.

At a higher frequency, i.e. 2 kHz, it shouldn't vary too much. But, it is interesting why you have a spike at 2 kHz, this is quite unusual. A typical RT spectrum has a higher emphasis at lower frequency (which you have), then is pretty flat above 1 kHz. Hence I suspect this spike is something to do with how it's been tested, or, perhaps there's something in the room like an air con unit which is making a bit of a tone around 2 kHz?

The RT is a measurement of how long sound takes to decay by 60 dB, though you usually measure a 20 or 30 dB decay and then interpolate to 60 dB, because measuring the whole 60 dB range relies on there being basically no background noise. For example, your impulse is say 100 dB at 2 kHz, hence your RT measurement would be the time taken to decay from 100 dB to 40 dB. But there's something in the room like an air con unit producing a consistent 50 dB at 2 kHz, hence how do you measure? This reason is why you would normally measure a 20 dB or 30 dB decay (T20 or T30 parameter) and interpolate, it helps to get around problems with background noise.

It could be a couple of other things, but check that you're measuring T20 or T30 in the first instance. It could well be the room though such a spike at 2 kHz is not something I've experienced in 10 years of being an acoustic consultant.... so I suspect not.

So in short, 100 Hz - nothing unusual here, 2 kHz - unusual but more likely to be down to the test setup than the room itself.

Any more problems feel free to shimmy down to ParkerJones Acoustics and use the 'live chat' feature on the website, or shoot me an email to [email protected]. I don't mind answering some quick questions if I'm not too busy.

All the best
Chris from ParkerJones Acoustics
Old 10th December 2019
  #4
Lives for gear
the topt measurement is not equivalent at the RT60 measurement. You can see the decay range in the Show data panel of the filtered impulse response window.

https://www.roomeqwizard.com/help/he...redir.html#top
Old 10th December 2019
  #5
Gear Addict
 

If you post pictures and dimensions of the room and the mdat-files you might get some sensible answers from people with knowledge.
Now all they can do is guess (& waist time).
Old 10th December 2019
  #6
Here for the gear
 

Quote:
Originally Posted by PJAcoustics View Post
Well your low absorption materials like concrete and glass don't have much variance over frequency, but the more absorbing materials do. For example, the perforated acoustic ceiling, could have an absorption coefficient in the region of 0.7 - 0.9 between 500 and 4kHz, whereas at 100 Hz, it may only be around 0.1 to 0.2.

I'm not too suprised by the 'spike' at 100 Hz, this is typical in most rooms. Low frequency is very difficult to absorb. Most surface finishes, including 'acoustic' products, have a poor low frequency performance when they are flush mounted onto a wall or soffit. What low frequency needs, is something thick, with an air gap, because it has a long wavelength. This is why many recording studios use bass traps (you don't see these in most other types of room, because 100 Hz just isn't as important for non-performance/music spaces). The bass trap normally has a depth which is equal to 1/4 the length of the wavelength you want to control. For 100 Hz, the wavelength is 3.43m, therefore the bass trap needs to be around 0.85m long. . You won't be getting this from any of your finishes (the closest you'll get is from the ceiling as this has an air gap above it, but probably not as much as 0.85m).

Also be aware that low frequency measurements of RT (<200 Hz) varies significantly as you move around the room, due to the room modes.

At a higher frequency, i.e. 2 kHz, it shouldn't vary too much. But, it is interesting why you have a spike at 2 kHz, this is quite unusual. A typical RT spectrum has a higher emphasis at lower frequency (which you have), then is pretty flat above 1 kHz. Hence I suspect this spike is something to do with how it's been tested, or, perhaps there's something in the room like an air con unit which is making a bit of a tone around 2 kHz?

The RT is a measurement of how long sound takes to decay by 60 dB, though you usually measure a 20 or 30 dB decay and then interpolate to 60 dB, because measuring the whole 60 dB range relies on there being basically no background noise. For example, your impulse is say 100 dB at 2 kHz, hence your RT measurement would be the time taken to decay from 100 dB to 40 dB. But there's something in the room like an air con unit producing a consistent 50 dB at 2 kHz, hence how do you measure? This reason is why you would normally measure a 20 dB or 30 dB decay (T20 or T30 parameter) and interpolate, it helps to get around problems with background noise.

It could be a couple of other things, but check that you're measuring T20 or T30 in the first instance. It could well be the room though such a spike at 2 kHz is not something I've experienced in 10 years of being an acoustic consultant.... so I suspect not.

So in short, 100 Hz - nothing unusual here, 2 kHz - unusual but more likely to be down to the test setup than the room itself.

Any more problems feel free to shimmy down to ParkerJones Acoustics and use the 'live chat' feature on the website, or shoot me an email to [email protected]. I don't mind answering some quick questions if I'm not too busy.

All the best
Chris from ParkerJones Acoustics
Thank you Chris for your feedback. I used REW to made the measures, I understand the software uses both RT20 and RT30. At the time of the sweeps there was nothing but noise floor ( near 50dBA ), no HVAC systems running, no people, nothing. Headroom on REW was near 10dB. I tried different positions for the mic but the curves are almost identical. I will go into the chat at your website, thanks!
Old 10th December 2019
  #7
Here for the gear
 

Quote:
Originally Posted by dinococcus View Post
the topt measurement is not equivalent at the RT60 measurement. You can see the decay range in the Show data panel of the filtered impulse response window.

https://www.roomeqwizard.com/help/he...redir.html#top
Hi Dinococcus. Thanks for your input. I understand from REW that Topt is the best guess from the software for the RT60? Beyond that, RT20 and RT30 are quite similar to the Topt curve.
Old 10th December 2019
  #8
Here for the gear
 

Quote:
Originally Posted by bert stoltenborg View Post
If you post pictures and dimensions of the room and the mdat-files you might get some sensible answers from people with knowledge.
Now all they can do is guess (& waist time).
Hi Bert, you are right. Attached you will find the mdat file for one of the measures and a couple of pics from the room to get an idea. As I said before, its about 58370 cubic feet. Concrete, glass and rigiton acoustic ceiling are the materials. I am not surprised on the 100Hz peak since the rigiton acoustic ceiling have a lower absorption coeficient on that frequency ( 0.3 ), but I am concerned about the 2KHz peak, since that coefficient is kind of flat from 1KHz and above ( 0.5 ).
Attached Thumbnails
RT60 big differences for different frequencies REW-img_20191129_165519957.jpg   RT60 big differences for different frequencies REW-img_20191205_165530767.jpg  
Attached Files
File Type: mdat Procer Punto 8.mdat (4.34 MB, 9 views)
Old 11th December 2019
  #9
Gear Addict
 

Did you point the microphone at the speakers?

In fact to measure RT you need an omni soundsource (and microphone, of course) to generate a stochastic soundfield.
I don't know REW very well, but MLSSA or ARTA can average measurements so you could point the speakers in f.e. 12 directions, measure these situations and average the results to get an impulse response containing the relevant information.

What PJ Acoustics says, f.e. that you would need 85 cm deep absorption to absorb 100 Hz is of course rubbish.
Old 11th December 2019
  #10
Lives for gear
 

Quote:
that you would need 85 cm deep absorption to absorb 100 Hz is of course rubbish.
+1!

Quote:
The bass trap normally has a depth which is equal to 1/4 the length of the wavelength you want to control. For 100 Hz, the wavelength is 3.43m, therefore the bass trap needs to be around 0.85m long. .
This is not correct. As Bert already pointed out, it is FAR from correct. In reality, there is substantial absorption from most types of porous absorber at 1/16th of the wavelength (take a look at the spec sheet for pretty much any typical fibrous absorption product). Also, the peak is at about 135 Hz, not 100 Hz. Thus, the wavelength is around 250cm, so absorption around 15cm to 20cm thick, correctly placed, would do a decent job. Thicker would be better, of course, but 15cm would still have a useful effect. (There are other types of acoustic treatment that can deal with 135 Hz in even less space that that.)

Quote:
What low frequency needs, is something thick, with an air gap,
Contrary to popular belief, an air gap does not lower the absorption frequency, nor does it increase the absorption coefficient. It does not necessarily even improve performance at all! Rather, it simply allows you to use less material to hit the same frequency range, but at the cost of possible uneven absorption at higher frequencies.

Quote:
This is why many recording studios use Bass Traps (you don't see these in most other types of room, because 100 Hz just isn't as important for non-performance/music spaces).
Actually, the reason why you don't see bass traps in larger rooms is because they are not needed, since large rooms do not suffer from the modal spread issues that affect smaller rooms. Diffusion and some broadband absorption is what large rooms commonly need. 100 Hz would be treated in different ways in a large room, such as this, and would likely still need some treatment in most large rooms. It just would not be in the form of bass traps, since 100 Hz is not going to be a modal issue in a large room, with 60,000 ft3 volume.

Quote:
Also be aware that low frequency measurements of RT (<200 Hz) varies significantly as you move around the room, due to the room modes.
Again, modal issues are not going to be a major problem in a room with a volume of nearly 60,000 cubic feet. Modal spread is only an issue in small rooms (less than about one tenth that volume). Large room acoustics is very different from small room acoustics, because there actually can be a statistical reverberant field in large rooms, so RT60 measurements really re valid (unlike small rooms, where there is no reverberant field).

Quote:
At a higher frequency, i.e. 2 kHz, it shouldn't vary too much.
Why not? Decay times can vary significantly across the spectrum in any size room, large or small, if it is not designed/treated correctly.

Quote:
A typical RT spectrum has a higher emphasis at lower frequency (which you have),
That is not normally true for large rooms, such as the one we are discussing here. Large rooms have far fewer problems in the low end that small rooms do.

Quote:
For example, your impulse is say 100 dB at 2 kHz,
That's not the way impulses work. Sorry. By definition, an impulse is extremely loud (way more than 100 dB), and has no frequency, because it is a simple pulse of extremely short duration: it is not a repeating cycle, thus it has no period, and no frequency. That's why it is called an "impulse", not a "tone". So there never was any "100 dB impulse at 2 kHz here". You seem to be confusing the impulse response with the sine wave sweep that REW uses to generate the impulse response.

Quote:
Hence I suspect this spike is something to do with how it's been tested, or, perhaps there's something in the room like an air con unit which is making a bit of a tone around 2 kHz?
No. The problem is acoustic, not mechanical, nor testing, nor HVAC related. How could an HVAC issue that "makes a tone around 2 kHz" also produce the spectral decay see in the graphs and in the actual MDAT? That would require someone to turn of the HVAC system at the exact instant when the REW sweep hit 2.5 KHz.... Not very likely. If it was an HVAC tone, it would not have any decay: it would be a constant peak visible in all the time-domain graphs, but there is no such peak. Hence, it is an acosutic issue, related to the room, not to the HVAC system, nor to the way the measurement was done. There might have been other issues with the REW testing procedure (see here to find out how to do it right: How to calibrate and use REW to test and tune your room acoustics ), but such issues would not show up in the way these results are indicating.

The MDAT does, in fact, show some low-level continuous tones at around 1 kHz, 1.3 kHz, and 1.4 kHz (plus a few others), perhaps from a power supply of some type, but there is nothing in the 2.5 kHz region, and the intensity is too low to be of concern.

Quote:
It could well be the room though such a spike at 2 kHz is not something I've experienced
It's not that uncommon to have uneven decay times, reverberation, or resonances of one type or another in large rooms such as this. It happens quite often, in fact. Below is the graph for a room I tested and treated a couple of years back (a church hall), about half the size of this one in this thread (in terms of volume: around 30,000 cubic feet), but with a lower ceiling. The decay peak at around 500 Hz is clearly visible, well into the mid range. That's not uncommon in larger rooms. I see it all the time.

Quote:
... ceiling have a lower absorption coeficient on that frequency ( 0.3 ), but I am concerned about the 2KHz peak, since that coefficient is kind of flat from 1KHz and above ( 0.5 ).
The ceiling is not the problem: the walls are the problem. Also, the peak decay is not at 2 kHz, but more like 2.6 kHz. That's easier to deal with than the 135 Hz issue.

It should not be too hard to get this room under control with suitable treatment,bBut as Bert said, we can't help you much with practical treatment suggestions until you provide the DIMENSIONS of the room. Not just the total room volume, but the actual dimensions. It would also be useful to know the purpose of the room, since different uses need different acoustic performance. If the room is to be used for lectures, then it is way, WAY too reverberant. But if it is to be used for performances by Gregorian chant choirs, it's probably really good just like it is!

- Stuart -
Attached Thumbnails
RT60 big differences for different frequencies REW-c3-man-rt60-example.png  
Old 11th December 2019
  #11
Gear Addict
 

Welcome back 20000 words a post Stuart!
Old 11th December 2019
  #12
Lives for gear
 

Quote:
Originally Posted by bert stoltenborg View Post
Welcome back 20000 words a post Stuart!
Thanks Bert! I'll try to ration the words a bit, so I don't hit the quota too soon each day...

- Stuart -
Old 11th December 2019
  #13
Gear Addict
 

If he measured with the directional tweeter hitting the mic in the face that could explain the 2 kHz RT 'anomalie' as the room isn't exited as it should.
If you display the measurement on a decent scale it's al not so strange and extreme, IMHO.

I like your remark on modes; a lot of people (and seemingly even 10-years-of-practice-acousticians) don't get the concept of large and small room acoustics.
Old 11th December 2019
  #14
Lives for gear
 
avare's Avatar
 

Great post Stuart. Adding to lf treatment is ©Invisible Alpha. Great article that has disappeared from New Yorker about Harris' design for the Avery Fisher Hall. Amongst other things the mass of fronts of the dress boxes varied making wider bandwidth membrane absorbers. Totally invisible absorption even to the construction crew.
Old 12th December 2019
  #15
Here for the gear
 

Quote:
Originally Posted by bert stoltenborg View Post
Did you point the microphone at the speakers?

In fact to measure RT you need an omni soundsource (and microphone, of course) to generate a stochastic soundfield.
I don't know REW very well, but MLSSA or ARTA can average measurements so you could point the speakers in f.e. 12 directions, measure these situations and average the results to get an impulse response containing the relevant information.

What PJ Acoustics says, f.e. that you would need 85 cm deep absorption to absorb 100 Hz is of course rubbish.
No, I did not. I put the mic on 14 different places all over the room, Just two of them "kinda" pointing the speakers, but all the graphs show practically the same RT on REW.
Old 12th December 2019
  #16
Here for the gear
 

Quote:
Originally Posted by Soundman2020 View Post
+1!

This is not correct. As Bert already pointed out, it is FAR from correct. In reality, there is substantial absorption from most types of porous absorber at 1/16th of the wavelength (take a look at the spec sheet for pretty much any typical fibrous absorption product). Also, the peak is at about 135 Hz, not 100 Hz. Thus, the wavelength is around 250cm, so absorption around 15cm to 20cm thick, correctly placed, would do a decent job. Thicker would be better, of course, but 15cm would still have a useful effect. (There are other types of acoustic treatment that can deal with 135 Hz in even less space that that.)

Contrary to popular belief, an air gap does not lower the absorption frequency, nor does it increase the absorption coefficient. It does not necessarily even improve performance at all! Rather, it simply allows you to use less material to hit the same frequency range, but at the cost of possible uneven absorption at higher frequencies.

Actually, the reason why you don't see bass traps in larger rooms is because they are not needed, since large rooms do not suffer from the modal spread issues that affect smaller rooms. Diffusion and some broadband absorption is what large rooms commonly need. 100 Hz would be treated in different ways in a large room, such as this, and would likely still need some treatment in most large rooms. It just would not be in the form of bass traps, since 100 Hz is not going to be a modal issue in a large room, with 60,000 ft3 volume.

Again, modal issues are not going to be a major problem in a room with a volume of nearly 60,000 cubic feet. Modal spread is only an issue in small rooms (less than about one tenth that volume). Large room acoustics is very different from small room acoustics, because there actually can be a statistical reverberant field in large rooms, so RT60 measurements really re valid (unlike small rooms, where there is no reverberant field).

Why not? Decay times can vary significantly across the spectrum in any size room, large or small, if it is not designed/treated correctly.

That is not normally true for large rooms, such as the one we are discussing here. Large rooms have far fewer problems in the low end that small rooms do.

That's not the way impulses work. Sorry. By definition, an impulse is extremely loud (way more than 100 dB), and has no frequency, because it is a simple pulse of extremely short duration: it is not a repeating cycle, thus it has no period, and no frequency. That's why it is called an "impulse", not a "tone". So there never was any "100 dB impulse at 2 kHz here". You seem to be confusing the impulse response with the sine wave sweep that REW uses to generate the impulse response.

No. The problem is acoustic, not mechanical, nor testing, nor HVAC related. How could an HVAC issue that "makes a tone around 2 kHz" also produce the spectral decay see in the graphs and in the actual MDAT? That would require someone to turn of the HVAC system at the exact instant when the REW sweep hit 2.5 KHz.... Not very likely. If it was an HVAC tone, it would not have any decay: it would be a constant peak visible in all the time-domain graphs, but there is no such peak. Hence, it is an acosutic issue, related to the room, not to the HVAC system, nor to the way the measurement was done. There might have been other issues with the REW testing procedure (see here to find out how to do it right: How to calibrate and use REW to test and tune your room acoustics ), but such issues would not show up in the way these results are indicating.

The MDAT does, in fact, show some low-level continuous tones at around 1 kHz, 1.3 kHz, and 1.4 kHz (plus a few others), perhaps from a power supply of some type, but there is nothing in the 2.5 kHz region, and the intensity is too low to be of concern.

It's not that uncommon to have uneven decay times, reverberation, or resonances of one type or another in large rooms such as this. It happens quite often, in fact. Below is the graph for a room I tested and treated a couple of years back (a church hall), about half the size of this one in this thread (in terms of volume: around 30,000 cubic feet), but with a lower ceiling. The decay peak at around 500 Hz is clearly visible, well into the mid range. That's not uncommon in larger rooms. I see it all the time.

The ceiling is not the problem: the walls are the problem. Also, the peak decay is not at 2 kHz, but more like 2.6 kHz. That's easier to deal with than the 135 Hz issue.

It should not be too hard to get this room under control with suitable treatment,bBut as Bert said, we can't help you much with practical treatment suggestions until you provide the DIMENSIONS of the room. Not just the total room volume, but the actual dimensions. It would also be useful to know the purpose of the room, since different uses need different acoustic performance. If the room is to be used for lectures, then it is way, WAY too reverberant. But if it is to be used for performances by Gregorian chant choirs, it's probably really good just like it is!

- Stuart -
Wow, what a reply! Thanks for the class. This is the biggest room that I have to deal with so far, so maybe is that's why I am not so sure about the numbers. The room will be used for a restaurant / venue. I was also discussing the graph with the author of REW on another forum and he said that looks fine and should be useful. I re calculated theoretical values for RT60 at different frequencies and I have a delta of about 25% on the 2KHz point against REW, so I guess I just have to accept it and treat the room accordingly to that RT60 values.
Old 12th December 2019
  #17
Here for the gear
 

Quote:
Originally Posted by Soundman2020 View Post
+1!
Contrary to popular belief, an air gap does not lower the absorption frequency, nor does it increase the absorption coefficient. It does not necessarily even improve performance at all! Rather, it simply allows you to use less material to hit the same frequency range, but at the cost of possible uneven absorption at higher frequencies.

- Stuart -
That is new for me, as Alton Everest says in his Master handbook of acoustics; "the absorption of porous material is much greater with an airspace between the material and the wall"
Old 12th December 2019
  #18
Here for the gear
 

Quote:
Originally Posted by Soundman2020 View Post
+1!

It should not be too hard to get this room under control with suitable treatment,bBut as Bert said, we can't help you much with practical treatment suggestions until you provide the DIMENSIONS of the room. Not just the total room volume, but the actual dimensions. It would also be useful to know the purpose of the room, since different uses need different acoustic performance. If the room is to be used for lectures, then it is way, WAY too reverberant. But if it is to be used for performances by Gregorian chant choirs, it's probably really good just like it is!

- Stuart -
The room has an irregular shape, I am attaching it.
Attached Thumbnails
RT60 big differences for different frequencies REW-plano-espacio.jpg  
Old 12th December 2019
  #19
Lives for gear
Quote:
Originally Posted by gerardochile View Post
That is new for me, as Alton Everest says in his Master handbook of acoustics; "the absorption of porous material is much greater with an airspace between the material and the wall"
when you use acousticmodeling, you can see the effect describe by Stuart with the ondulation showing by the curves.

http://www.acousticmodelling.com/porous.php
Old 13th December 2019
  #20
Lives for gear
 

Quote:
Originally Posted by avare View Post
Great post Stuart. Adding to lf treatment is ©Invisible Alpha. Great article that has disappeared from New Yorker about Harris' design for the Avery Fisher Hall. Amongst other things the mass of fronts of the dress boxes varied making wider bandwidth membrane absorbers. Totally invisible absorption even to the construction crew.
Thanks Andre. You are right: I completely neglected to mention one of your favorite subjects! "Invisible Alpha". Something that a lot of people aren't even aware of, but should be.

It's a pity about that article on the Avery Fisher Hall: I would have liked to have read that: it sounds fascinating. To totally mis-quote Shakespear: "There are more types of absorption in heaven and Earth, Horatio, than are dreamt of in Newell's philosphy!"...




- Stuart - (Dramatically)
Old 13th December 2019
  #21
Lives for gear
 

Quote:
Originally Posted by dinococcus View Post
when you use acousticmodeling, you can see the effect describe by Stuart with the ondulation showing by the curves.
Right! I'm adding another similar one here, which shows how that can change for different combinations. The green curve is for very high density absorption, 40,000 rayls, 600mm deep. Blue is very light absorption, 3,000 rayls, also 600mm deep. Red is for a layer of 3,000 rayls 200mm thick over a 200mm air cavity, with another 200 mm of the 40,000 rayls stuff at the back. And orange is for 400 mm of 3,000 rayls at the front, a 100mm air cavity, and 100mm of 40,000 at the back. I don't recall why I did that graph originally, but I think it was to show something about how you can layer insulation in a bass trap to achieve different things, and how useless 40,000 rayls is for bass traps... But the graphs is useful here as well, to add more weight to what you I said above.

- Stuart -
Attached Thumbnails
RT60 big differences for different frequencies REW-layered-absorption-graph-blue-600x3k-green-600x40k-red-200x3k-200xair-200x40k-orange-400x3k-1.jpg  
Old 13th December 2019
  #22
Lives for gear
 

Quote:
Originally Posted by gerardochile View Post
That is new for me, as Alton Everest says in his Master handbook of acoustics; "the absorption of porous material is much greater with an airspace between the material and the wall"
Correct! But what Everest is saying there is the same as what I'm saying. He is talking about the case where you have an absorber, for example 4" of OC-703, against the wall: if you then move it away from the wall, for example by 6", so the front is now 10" away from the wall, then it will now work down to lower frequencies. But is is not the air gap per sé that makes it work down to lower frequencies! Rather, it is simply the fact that the front face of the insulation is now further away from the wall. The frequency range that any absorber can cover is partly defined by how far away the front face is, with respect to the wall behind it, so the simple act of pulling the same 4" absorber away from the wall will now make it work down to the same frequency as if it were 10" thick. The difference is that the absorption spectrum is no longer smooth, due the the air cavity behind the absorber, as you can see in the graph in my previous post and as mentioned by dinococcus. If you were then to stuff that air cavity with even more insulation, that would not change the frequency (since the front face is still 10" from the wall), but it would smooth out the absorption curve again, as show in that graph. So, what sets the frequency is the depth from the wall, regardless of whether or not there is an air cavity there. The air cavity can be useful, if you know how to tune it, or it might be a problem for someone who doesn't know what they are doing, as you can see in the graph below. All of those are for the exact same depth of absorber: 2" of insulation over a 4" air cavity. The only thing that changes is the GFR (Gas Flow Resistivity) of the different types of insulation. Which brings up the second factor that controls the frequency range of a porous absorber trap: it's acosutic impedance. GFR is a measure or acoustic impedance. You can see that all of the devices in this graph have roughly the same peak frequency, around 600 Hz., but the amount of absorption for varies widely across the spectrum for different values of GFR.

So that's what Everest was saying, and what I was saying too: contrary to popular belief, having an air gap does not make it work down to lower frequency: getting the front of the panel further away form the wall is what makes it work down lower. If all you have is a 4" panel, then as Everest says, it makes more sense to move it away from the wall, because that extra distance to the front face will make it work lower. But what he doesn't mention is that it might also skew the rest of the absorption curve, if you don't fill the cavity that you created. Or at least choose a distance and a type of insulation that works for the situation in question.

- Stuart -
Attached Thumbnails
RT60 big differences for different frequencies REW-porous-absorption-graph-6-inch-various-rayls.jpg  
Old 13th December 2019
  #23
Lives for gear
 

Quote:
Originally Posted by gerardochile View Post
Wow, what a reply! Thanks for the class.


Quote:
This is the biggest room that I have to deal with so far, so maybe is that's why I am not so sure about the numbers.
As Bert mentioned: there's a fundamental difference between acoustics in small rooms, and acoustic in large rooms. So maybe that's what is messing with your expectations!

Small rooms do not have a statistically valid reverberant field: large rooms do. What that means is simply this: one of the defining characteristics of a "reverberant field" is that no matter where you stand in the room, the sound seems to come at you equally from all directions, at the same level. In a small room, that is not the case, since there simply is not enough distance for that to happen: most of the sound seems to come at you from a few specific directions, so there is no true reverberance. There might still be resonance, yes, but the total sound field is not really reverberant . Thus, you can't validly talk about "RT60" times in a small room, because the very definition of RT is "Reverberance Time": it's the time taken for the reverberant field to decay by 60 dB after the initial impulse. But there is no "reverberant field" in a small room! So it can't decay...

OK, so this is sort of splitting hairs, because the resonances can still decay, and that sort of looks like RT60 on a graph... but technically it isn't.

So sound just behaves different in a small room, than it does in a large one. Reverberance is just one way that is is different: there are other differences too. It's confusing. If you try t apply small-room acoustic solutions to large rooms, they won't work very well. That's why I said that bass traps are not necessary in large rooms, because they don't have the same bass problems that small rooms do. You might still need treatment in the low end, at the same frequencies, certainly, but you would not use the same type of treatment in a large room, because the fundamental problem is different.

Quote:
The room will be used for a restaurant / venue.
Great! When you say "venue", I'm assuming that some type of music will be performed there? A small live band, perhaps? A singer with a piano or guitar? Something like that? That narrows down the target decay times considerably, so the goal is more clear.

Quote:
I was also discussing the graph with the author of REW on another forum and he said that looks fine and should be useful.
Yup! I agree. Your results are about what I would expect from that room, untreated.

Quote:
I re calculated theoretical values for RT60 at different frequencies and I have a delta of about 25% on the 2KHz point against REW, so I guess I just have to accept it and treat the room accordingly to that RT60 values.
To a certain extent! But with large rooms, there's more to take into account than just the acosutic treatment itself. For example, you say that this will be a restaurant: thus you will have tables and chairs. You need to consider those in your calculations. They will have an effect on the acoustic response of the room. Having a hundred wooden or plastic chairs in there will be rather different from having a hundred upholstered chairs. Ditto for the tables: If they are just simple light-weight open metal "cafe" type tables, that's different from having large, bulky heavy wood tables. Maybe not a huge difference, but still something to consider. Even wall decorations could make a difference. For example, if you plan to have a lot of fabric drapery on the walls, vs. leaving them bare. Significant difference.

Quote:
The room has an irregular shape, I am attaching it.
Ahhh! OK, that makes more sense! That clarifies a lot.

And you have a more complicated situation than you think: one entire wall is a concave cylinder! That's a problem. Curved surfaces tend to focus sound, sometimes in ways that you would not have suspected. You might find this article interesting: http://digistar.cl/Forum/viewtopic.php?f=7&t=22 It's about a church I was hired to treat a few years ago, where they had a curved front wall behind the stage that made it impossible for the band to play. I got called in late in the project, after the architect had already designed it and they were well advanced with the construction. Too late to change the design: all I could do was treat it, and design the sound system to go with it. It's not identical to your case, but there are some similarities, so maybe worth reading. The treatment will be different in your case, obviously! However, there might be something in there that you find useful.

The good news is that it is possible to treat your venue for what you want, without too much complexity. It will need a bit of careful design, and some more targeted testing with REW, but it can be done, and using typical acoustic treatment.

- Stuart -
Old 17th December 2019
  #24
Here for the gear
 

Quote:
Originally Posted by Soundman2020 View Post


And you have a more complicated situation than you think: one entire wall is a concave cylinder! That's a problem. Curved surfaces tend to focus sound, sometimes in ways that you would not have suspected. You might find this article interesting: http://digistar.cl/Forum/viewtopic.php?f=7&t=22 It's about a church I was hired to treat a few years ago, where they had a curved front wall behind the stage that made it impossible for the band to play. I got called in late in the project, after the architect had already designed it and they were well advanced with the construction. Too late to change the design: all I could do was treat it, and design the sound system to go with it. It's not identical to your case, but there are some similarities, so maybe worth reading. The treatment will be different in your case, obviously! However, there might be something in there that you find useful.

The good news is that it is possible to treat your venue for what you want, without too much complexity. It will need a bit of careful design, and some more targeted testing with REW, but it can be done, and using typical acoustic treatment.

- Stuart -
Thanks again for your time on this thread Stuart. And yep, that curved wall is not the best. I am planing to treat the room with a material called sonoglass that has a good coefficient on the 100-120Hz region and maybe to add some diaphragmatic resonators. By the way, I am writing from central south Chile. Do you live in Chile also? This place will have a lot of beer once done, I owe you some!
Old 4 weeks ago
  #25
Lives for gear
 

Quote:
Originally Posted by gerardochile View Post
Thanks again for your time on this thread Stuart. And yep, that curved wall is not the best. I am planing to treat the room with a material called sonoglass that has a good coefficient on the 100-120Hz region
Sonoglass is good stuff. I have used that to in treatment before, and it should work fine for what you want.

Quote:
and maybe to add some diaphragmatic resonators.
I don't think you'll need those for a room that size. The decay tails you are seeing are not modal, so it's unlikely that you'll need resonant traps. I suspect that absorption, located suitably, in combination with some type of large diffusion (I'm partial to poly-cylindrical diffusers myself, for large spaces), or perhaps even slotted polys (which I also use a lot). It is possible to get some aesthetically attractive treatment like that, in pleasing curved shapes. It's a little more expensive to do that, but it could look really nice... especially if you could do it in stained or varnished Raulí or Alerce, if you happen to like warm-looking wood (and if you can get it!). Coigue is nice too, but even plain old pine is fine as well... and cheaper!

Quote:
By the way, I am writing from central south Chile. Do you live in Chile also? This place will have a lot of beer once done, I owe you some!
Cool! Yep, I sure do live in Chile. In Santiago, actually. What city are you in? I'm planning a trip to the south sometime next year (I love Puerto Varas), so maybe I could drop in on the way, and take you up on that offer!

- Stuart -
Old 4 weeks ago
  #26
Here for the gear
 

Quote:
Originally Posted by Soundman2020 View Post
Sonoglass is good stuff. I have used that to in treatment before, and it should work fine for what you want.

I don't think you'll need those for a room that size. The decay tails you are seeing are not modal, so it's unlikely that you'll need resonant traps. I suspect that absorption, located suitably, in combination with some type of large diffusion (I'm partial to poly-cylindrical diffusers myself, for large spaces), or perhaps even slotted polys (which I also use a lot). It is possible to get some aesthetically attractive treatment like that, in pleasing curved shapes. It's a little more expensive to do that, but it could look really nice... especially if you could do it in stained or varnished Raulí or Alerce, if you happen to like warm-looking wood (and if you can get it!). Coigue is nice too, but even plain old pine is fine as well... and cheaper!

Cool! Yep, I sure do live in Chile. In Santiago, actually. What city are you in? I'm planning a trip to the south sometime next year (I love Puerto Varas), so maybe I could drop in on the way, and take you up on that offer!

- Stuart -
Actually I am doubting about resonators, but even that the room is big, the place will not only be a bar/restaurant, it will also be used as a live rock music venue as well. I wonder if I can treat that LF, since sonogalss is not helping me so much below 80Hz, and I am trying to flat the RT60 all over the spectrum.

Diffusors are on the list too, and probably polys, but that is not only on me, but on the interior designer that is also involved on the project. So far I guess they do not want a lot of wood. Will see.

The offer is real! lol. I live in Chillan, so before your trip, pm me in advance; +56 9 3198 3420. If you come past march, the place should be operating with a lot of hand crafted variety of beers. If not, we can go directly to the brewery.

Gerardo.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump