Calculate how much absorption after measuring RT

[chickenfingers]

Hi

A friend and I want to treat his music rehersal room. Is far to reverberant.
I managed to get my hands on a good Sound Level Meter and large loudspeaker to measure the RT. The loudspeaker is to play pink noise, however we may use an air gun instead. Which method you recommend? The room´s about 6m,6m,3m. I am aware of the RT methods locating the loudspeakers towards the corners to try to recreate a diffuse field...

After, how can we predict the amount of absorption needed to bring down the RT? Which equation? (Sabine, Eyring). We haven´t got any software to do a simulation. Seems a bit contradictory to me to measure RT and then use an equation to predict the desired RT. Would instead do it all with a formula from the beginning.

Any help is much appreciated.

Thanks!

oh and we haven´t got any money to hire an acoustic consultant. I´m fairly confident I can deal with it.

Sergio

 

[Savant]

Sergio,

Which source to use - loudspeaker or gunshot - depends on what the SLM can do. How does it (or, indeed, can it) determine RT?

For the size of room you're looking at, any RT equation and no RT equation are the answers. The room is a little too small (108 m³) to be considered statistically reverberant, except at high frequencies. Therefore, none of the equations are going to give numbers that jive with the measurements. However, for a general idea of how much absorption to add, any of the various equations should work fine. Sabine will be the easiest with which to work.

The calculated RTs will have to be adjusted to match the measurements. E.g., if the measurement says 2.4 s and the Sabine calc says 1.8 s, negative absorption should be "added" to the calcs to get the Sabine number up to 2.4 s. To get this to be as "accurate" as possible, make RT measurements at as many different points in the room as is reasonable. I would suggest between 4 and 6 different measurement points, none of which are too close to a room surface (1-1.5 m away, minimum).

Having said all that, calculations (and measurements) can be skipped completely, IMO. Assuming mostly hard surfaces to start with, 20-25% coverage of the walls and ceiling (roughly 22-27 m² for the room in question) with absorption - spread around and staggered - is always a good starting point for a rehearsal space. If that's not enough to get the desired room sound, keep treating with absorption and/or diffusion until it sounds good. It would be best to select broadband absorption (8 to 10 cm thick, minimum) and some "traps" for the corners.

HTH.

 

[jostenmeat]

After, how can we predict the amount of absorption needed to bring down the RT? Which equation? (Sabine, Eyring). We haven´t got any software to do a simulation. Seems a bit contradictory to me to measure RT and then use an equation to predict the desired RT. Would instead do it all with a formula from the beginning.

Hi Sergio, firstly, I am talking over my head. But, AFAIK, the measurements of sabines of any particular product is not exactly accurate, since every room and positioning is different. Your room is not the same as the laboratory room. You can find comparative results between products if they use the same testing methods, in the same room, on the same day. I've read there is not a US testing facility that is even certified to report absorption data below 100hz.

Anyhoo, the folks that seem to be the most serious about treating a room seem to add treatments, measure, add treatments, measure, add treatments, measure . . . and they keep going till they have like 40!

Again, I'm a noob, and I'd just as soon listen to Savant instead.

 

[chickenfingers]

Savant

Thanks for your quick reply

If I were to design a room where I needed a precise RT60, say a high school classroom (according to goverment REGULATIONS they must have 0.9-1sec RT60). Which method would you aim for? Taking into account only two measurements of RT60 can be made. One, before the design stage. And the second after the design has been installed!!!

Cheers

Sergio

 

[Savant]

Sergio,

In general for regulatory situations, calculation methods should generally "match" the measurement methods being used. Since most standardized measurements of RT, as well as published absorption coefficient data derived from standardized tests, are based on or otherwise conform to the Sabine equation, the Sabine equation would be the correct method. There are probably exceptions, but I cannot think of any off the top of my head.

Let me know what country you are in and I'd be happy to elaborate, particularly with regards to measurement.

 

[highfigh]

Hi

A friend and I want to treat his music rehersal room. Is far to reverberant.
I managed to get my hands on a good Sound Level Meter and large loudspeaker to measure the RT. The loudspeaker is to play pink noise, however we may use an air gun instead. Which method you recommend? The room´s about 6m,6m,3m. I am aware of the RT methods locating the loudspeakers towards the corners to try to recreate a diffuse field...

After, how can we predict the amount of absorption needed to bring down the RT? Which equation? (Sabine, Eyring). We haven´t got any software to do a simulation. Seems a bit contradictory to me to measure RT and then use an equation to predict the desired RT. Would instead do it all with a formula from the beginning.

Any help is much appreciated.

Thanks!

oh and we haven´t got any money to hire an acoustic consultant. I´m fairly confident I can deal with it.

Sergio

I would recommend using software based programs for this. You'll be able to see the graph and some have RT60 tests.

 

[chickenfingers]

Sergio,

In general for regulatory situations, calculation methods should generally "match" the measurement methods being used. Since most standardized measurements of RT, as well as published absorption coefficient data derived from standardized tests, are based on or otherwise conform to the Sabine equation, the Sabine equation would be the correct method. There are probably exceptions, but I cannot think of any off the top of my head.

Let me know what country you are in and I'd be happy to elaborate, particularly with regards to measurement.

Savant

With regards to your last comment Im in Colombia. Specific Regulations such as RT60 in classrooms are not enforced at all, the focus is background noise levels.
I think for the time being Im gonna go with the following measuring/calculation method:

Two representative loudspeaker positions (at the front of the classroom, where the teacher usually is. Loudspeaker pointing forward about 1.5m high), 6 measurements positions (3 measurements per position, mic on a tripod 1.5m high)-similar to ISO 3382-, 3sec long pink noise (3secs for the noise to build up in the room before turning it off to measure the RT60).
Measurement Mic: Behringer SM8000 with a preamp (don´t know which one yet). Will record the samples in Wavelab 6.0 and then generate individual files for each octave band using a parametric Eq plug-in. Finally I will get the RT60s using a code I wrote in Matlab.

Two questions:
1)What level shall I set the PreAmp to? It doesn´t matter right? as long as the measurement doesn´t clip.
2)Was thinking of using a 1/3 oct band eq to get the flattest possible noise before measuring. But does it matter? Im not measuring freq. response only RT60, so I guess not.

Later I will use Sabine as in your first reply to predict the amount of absoprtion required.

I know I could avoid all these with a good SoundLevelMeter (which would do all for me i.e.B&K 2260, but all I could get was a RadioShack SLM) or a proper software. I have RoomEqWizard but it uses sweeping freq for the measurements (rather do it with pink noise, besides haven´t fully gotten my head around it, going straight to its forum after this), Smaart seems to be more oriented towards freq. response and loudspeaker setups. Just getting SpectraPlus now. Can you recommend one?

Sorry about this messy reply but if you don´t mind, I could do wit a second opinion.
Thanks a lot!!!

Sergio

 

[Savant]

Sergio,

I'm not familiar with the exact regulations in Columbia, but I was going to mention ISO 3382. Thus, it appears you are already on the right track!

1) Set the preamp gain as high as possible while avoiding clipping.
2) I guess using Wavelab EQ filters is OK. I don't know how likely the filters in Wavelab's EQ module conform to ANSI and IEC standards. But they're probably going to be "close enough."

For other programs, I would suggest ETF from AcoustiSoft or Ymec's RAD, though I will admit the latter may have a steeper learning curve. Many people have had success with ETF and the price is great.

The other thing you could do is check out the Sencore SP395, or its predecessors, the Audio Toolbox series. While subtly different, each Audio Toolbox model has an RT60 measurement module (or two or three).

HTH.

 

[chickenfingers]

Thanks SAVANT!

Will look into the software you recommended.
Hopefully everything will go smoothly from now on.

Sergio