• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Open Discussion on Large/Larger Room Acoutics

Trdat

Addicted to Fun and Learning
Forum Donor
Joined
Sep 6, 2019
Messages
967
Likes
396
Location
Yerevan "Sydney Born"
I am hoping to instigate a discusion and possibly help others and of course myself out in the process. I am working on wedding hall that needs reduced reverberation times. Just so the thread doesn't attract basic answers I am not an amatuer(apologies for blowing my own trumpet) and know enough about small room acoustics at the very least I think I do and that this thread would rather serve as a guidance, discussion, and a solid platform for larger room venue type acoustics which might have been missed in small room acoustics conversations. Although this all also depends if there is a major discrepancy between the fundamentals of small room acoustics compared to that of larger venues but I am sure there is an overlapp.

The hall is 3000m3 23m from left to right and 17m front to back. The ceiling is a vaulted ceiling which tappers up to about 9m and starts from the stage at around 5m.

1. What is the most effective absorber for the lowest frequencies? 30cm thick fibreglass still doesn't seem to cut it on a pourous calculator. Perhaps a diaphragmatic absorber? I know its easy enough to check on a calculator but does anyone have sold experience, cause I have a back wall with a small portion 2 metre wide and 9 metre high 30cm thick and wondering how to utilise it. I know everyone will say soft fluffy stuff but how about some type of absorber like diaphraghmatic mentioned in Acoustic fields? https://www.acousticfields.com/how-to-build-a-diaphragmatic-absorber/

2. I have heard of the software Ease but are there any other free programs that can help with modelling and simulation for venue acoustics, prefeably something easy if it exists. Or do I just have to stick with a percentage rule if I don't have software. Of course I will measure but what do I use as the starting point to understand how much absorbtion according to what RT60...?

4. Another technique is tuned bass traps but is there a simple way to work out how much bass trapping you need and how much tuned bass trapping you need compered to a 4inch thick panel? Again is there a way to work this out or do I have to add and test as I go along?

3. Room modes occur less in larger rooms but would 4 sub woofers in corners still help to flatten out bass or provide any other advantage? There still will be SBIR from the floor and back wall so that also needs to be tackled right?


5. My understanding for larger rooms the detrimental early reflections are at about 30ms, and larger rooms have this issue much less due to the later arrival times. Griesenger mentions Lateral reflected energy in the 10ms to 50ms range reduces the “closeness” of the sound image • While this perception is pleasant, it is not musically essential. • too much energy in this time range can cause image broadening, timbre coloration, etc. • The ideal is to have the total energy in this time range two to four times less than the direct sound.

Then he goes on to say reflections between 50ms and 150ms are maximullly disturbing and saying reflections after 150ms add to envelopment. Now, I know he mainly works on opera and concert hall, and my application is a wedding hall so I would prefer lower reverberation times with a closer EDT to the RT60(while opera halls would need a longer gap between EDT and RT60 to increase envelopment). I just can't get my head around how you would reduce reverberation while reducing energy in that mid range ms. Any opinions?

6. "As a rule of thumb, sound is delayed 1 mS for every 34 cm of travel." I read this from a website.
Can I use this statement to give me rough idea of at least how long it will take for the direct sound to get to the specific wall, which will essentially let me know after what ms it will arrive to the listerner(of coure all different depending on where they are sitting or dancing)
 
Last edited:
OP
Trdat

Trdat

Addicted to Fun and Learning
Forum Donor
Joined
Sep 6, 2019
Messages
967
Likes
396
Location
Yerevan "Sydney Born"
I think this is going to be a first for someone answering there own questions.

Regarding a platform to start calculations for determining how much absorbtion is needed is the amacoustics reverberation calculator. The only thing I don't undertsand is the result which usually says.

"At 40Hz your current reverberation time is 2.96sec. That means that the equivalent absorption area is 1509.35 m2.
To reach your target of a reverberation time between 1.57 and 1.92 seconds at 40Hz, you need to add between 810.4 and 1326.5 m2 of equivalent absorption area.

And it mentions that for each frequency range as you move your mouse across the frequency so essentially that is the range of absorbtion we need overall? We don't have to add it for each frequency right? So the total surface area that needs to be covered is in that range?

And then I answer another part of my question, I can see what frequency needs more absorbtion and tune the traps to that frequency. And use the cooefficents of each absorber to to work out how much of what frequency I have absorbed.

The only point of confusion is that if the range of absorbtion is say, 800 to 1400m2 surface to be covered than how do I divide this to find how much absorbtion is needed for each specific frequency?
 

Hipper

Addicted to Fun and Learning
Joined
Jun 16, 2019
Messages
753
Likes
625
Location
Herts., England
There may be people on here that can help - I'm not one of them! - but small room acoustics, which is all I've seen discussed on here, is about getting good sound at one or at most a few positions, whereas you are presumably wanting a good sonic experience throughout the hall.

You may well be familiar with the book 'Master Handbook of Acoustics' by Everest and Pohlmann. There is a chapter (no. 22) on the subject of 'Acoustics in Large Halls'. It points out that you have to decide on whether you want intelligibility for speech or music. I would also think that psychoacoustics will be completely different because of the longer distances sound has to travel.

I realise this is a 'basic' answer. Oh well, someone had to do it!
 

raindance

Major Contributor
Joined
Sep 25, 2019
Messages
1,037
Likes
968
Will you be adding a sound system to the room or are you simply trying to make it quieter when people are in it? The location and type of speakers will play a huge part in the treatment options needed.
 
OP
Trdat

Trdat

Addicted to Fun and Learning
Forum Donor
Joined
Sep 6, 2019
Messages
967
Likes
396
Location
Yerevan "Sydney Born"
Will you be adding a sound system to the room or are you simply trying to make it quieter when people are in it? The location and type of speakers will play a huge part in the treatment options needed.

It will have full sound system which I am also working on but of course with the audio engineer. So full acoustics to improve the sound system, bring down reverberation time and also to improve speech intelligibility which we might to with side speakers from a seperate matrix.

I have pretty much answered most of the questions I was dwelling on back then, except for number 5. Although number 5 is self explanatory, my difficulty is understanding the spectral content of the reflection in order to understand what I need to do with it and how i find a range of reflections? Anyway, I will be adding a new thread to ask these and learn about the interpretation of REW graphs.

Overall, I got the basic down pat but its the measurement interpretation which will catapult the knowledge to the next level.
 

kyle_neuron

Active Member
Joined
Jun 18, 2021
Messages
149
Likes
254
I’m afraid you are all outta luck on ‘free’ and ‘easy’ room acoustics software. The closest you’ll find is Akabak, which has a demo version that only lacks the ability to save the calculated boundary element matrix once solved. There are a couple of ‘room’ demo files in the included examples, and you can import specific acoustic wall impedance data to map onto surfaces. It has a pretty steep learning curve though, so if you’re not familiar with general CAD modeling concepts like nodes, vertices, planes and meshing then budget a lot of time to learn.

That goes for EASE and the like too. Odeon is the most user-friendly UI-wise, but you’ll need over £10k in your pocket to get started. There are also a few plugins or standalone apps from research projects, like RAVEN, but these aren’t really developed to the standard where I’d want to have a client spend money based on the results.

Unfortunately, the simple calculated models often aren’t accurate enough in this type of environment either, as you need to know the makeup of the surface materials.

Geometric acoustics models have their own limitations, of course. They’re all linear acoustic models. You won’t get any data below 100 Hz or above 10 kHz out of it, and the calculation times can be obscene; EASE is a single core app by default, unless you spring the extra cash for the AURA module. A big part of this work is knowing what you can leave out or optimise in the model, how to analyse the results and tweak things, and the like.

As with any model or simulation, the results need to be validated. My usual process is to make a defeatured, simple CAD model of the space and assign a generic list of materials based on photos or visual assessment on site. Once that’s built, measurements are done on site with dodec sources and the like to the ISO standard, and the results are used to tweak the model parameters to match the generated binaural room impulse responses and room acoustic metrics. Once the model is verified using the spherical or theoretical sources, actual loudspeaker data is swapped in using GLL or CLF files, and then tweaks are made to the surface materials, geometries, and the like with intensive ray or cone tracing done at specific ranges of ‘seats’ to determine the angle and location of the reflections reaching listeners, via a ‘3D hedgehog’ map or an animated ray or particle bounce.

The process takes a fairly long time, and the clients who need or want it are aware of the costs involved…

If you’re curious about EASE in particular - it is the cheapest of these tools by far, too - then the ‘manual’ on their site is a good guide for room acoustics modeling and the metrics of worth in general.

Edit: Just seen your new post. Measurements need to be done to the ISO standard, and interpreting the results is the real skill. If you want to derive the directions of reflections, you’ll need multiple calibrated microphone capsules taking coincident measurements of the sound intensity field - that allows the derivation of the acoustic velocity and angle between the capsules via complex transfer functions and correction weightings.

You can cheat your way there a bit by using a high-quality Ambisonic mic with spherical harmonics decomposition, and a license for GratisVolver Pro. The cheapest way to do that at sufficient quality within the ISO bands is going to cost around £1000. You want a Core Sound Tetramic, a high quality linear four-channel portable recorder with stable phantom power and 96 kHz capability, and the relevant software.

The problem then is knowing what to do with the data. Not having a model to feed a test, verify, sim, tweak, repeat method is costly in time and money as you try things that mean reasonably large modification to the building.
 
Last edited:
OP
Trdat

Trdat

Addicted to Fun and Learning
Forum Donor
Joined
Sep 6, 2019
Messages
967
Likes
396
Location
Yerevan "Sydney Born"
I’m afraid you are all outta luck on ‘free’ and ‘easy’ room acoustics software. The closest you’ll find is Akabak, which has a demo version that only lacks the ability to save the calculated boundary element matrix once solved. There are a couple of ‘room’ demo files in the included examples, and you can import specific acoustic wall impedance data to map onto surfaces. It has a pretty steep learning curve though, so if you’re not familiar with general CAD modeling concepts like nodes, vertices, planes and meshing then budget a lot of time to learn.

That goes for EASE and the like too. Odeon is the most user-friendly UI-wise, but you’ll need over £10k in your pocket to get started. There are also a few plugins or standalone apps from research projects, like RAVEN, but these aren’t really developed to the standard where I’d want to have a client spend money based on the results.

Unfortunately, the simple calculated models often aren’t accurate enough in this type of environment either, as you need to know the makeup of the surface materials.

Geometric acoustics models have their own limitations, of course. They’re all linear acoustic models. You won’t get any data below 100 Hz or above 10 kHz out of it, and the calculation times can be obscene; EASE is a single core app by default, unless you spring the extra cash for the AURA module. A big part of this work is knowing what you can leave out or optimise in the model, how to analyse the results and tweak things, and the like.

As with any model or simulation, the results need to be validated. My usual process is to make a defeatured, simple CAD model of the space and assign a generic list of materials based on photos or visual assessment on site. Once that’s built, measurements are done on site with dodec sources and the like to the ISO standard, and the results are used to tweak the model parameters to match the generated binaural room impulse responses and room acoustic metrics. Once the model is verified using the spherical or theoretical sources, actual loudspeaker data is swapped in using GLL or CLF files, and then tweaks are made to the surface materials, geometries, and the like with intensive ray or cone tracing done at specific ranges of ‘seats’ to determine the angle and location of the reflections reaching listeners, via a ‘3D hedgehog’ map or an animated ray or particle bounce.

The process takes a fairly long time, and the clients who need or want it are aware of the costs involved…

If you’re curious about EASE in particular - it is the cheapest of these tools by far, too - then the ‘manual’ on their site is a good guide for room acoustics modeling and the metrics of worth in general.

Edit: Just seen your new post. Measurements need to be done to the ISO standard, and interpreting the results is the real skill. If you want to derive the directions of reflections, you’ll need multiple calibrated microphone capsules taking coincident measurements of the sound intensity field - that allows the derivation of the acoustic velocity and angle between the capsules via complex transfer functions and correction weightings.

You can cheat your way there a bit by using a high-quality Ambisonic mic with spherical harmonics decomposition, and a license for GratisVolver Pro. The cheapest way to do that at sufficient quality within the ISO bands is going to cost around £1000. You want a Core Sound Tetramic, a high quality linear four-channel portable recorder with stable phantom power and 96 kHz capability, and the relevant software.

The problem then is knowing what to do with the data. Not having a model to feed a test, verify, sim, tweak, repeat method is costly in time and money as you try things that mean reasonably large modification to the building.
I really apprecaite your reply and I came a across a few simulating softwares which were quite expensive. Maybe in future I will invest in Ease depending on if I get more jobs in future techncially its only a hobby. I dont think simulation software is an absolute need but I presume it makes things easier and with it there is the potential to design great sounding rooms from the CAD model and from the start of the operations. Currently, we are limited on what we can do to the hall. I will take in consideration the software mentioned and look into all of them, appreaciate reply.

There is a 60cm gap between the drywall and the above concrete ceiling which is acting as some sort of mebrane, Helmonts absorber as the RT60 between 60hz and 100hz is 2 and 2.5s while the midband goes up to 4.5. Back wall is glass allowing bass to run through and its a vaulted ceiling rising up to 9 metres. I used REW and the measurement seems decent in its accuracy no anomolies, but like you said what to do with the data.

My weaknesses will probably be made up for with simulation, I struggle to understand the spectral content of the reflection and where its coming from. Which is what I will start a new thread for, see if I can get some input on interpreting REW measurements. Otherwise my understanding of how to deal with them and what the literature supports with the various theories is something I have a basic grasp of and might do. I figured out the online calculators and they might not be that accurate but putting together all the literatrue you get an idea of what treats what and how to tackle each frequency range.

This wedding hall is on a tight budget, and its doesn't need to sound like a concert hall, bring down reverberation time and some decent diffusion and keep it simple but of course it will be great to know how to maintain the reverberation and reflection after 150ms and reduce the ones between 50 and 150ms it all sounds easy on paper. And getting rid of detrimental reflection to the seated guests but keep the envelopment(after 150ms reverb tail) but that's not gonna happen as I can't touch side walls but I woudn't want to absorb them anyway so how to diffuse the side walls would also be good to know.

So looks like I will have to do it old skool which would be totally against what a profesional like you will preach. I can get a consultant, but considering the budget and the limited testing equipment how much more it will give me I am not sure and even if it will would it be worth it?

The sound system is also a challenge, room modes and the whole bass suck out in large room acoustics is another ball game with PA compared to home 2 channel listening. So my work is cut out learning the basics of PA so I can work with the audio engineer to choose a decent sound system. But of course, not my responsibility, I am taking care of acoustics.
 

kyle_neuron

Active Member
Joined
Jun 18, 2021
Messages
149
Likes
254
Oooh well, here we reach a limit. None of the models include transference or sound transmission. You need AURA for scattering coefficients if they're in the library for your material...

You need BEM for bass array and design. That's my forte, and a lot of things are in the classic acoustic texts but need tweaking for modern implementation. Mostly you're limited by the real world- where you can feasibly place things.

I recently had some very elucidating conversations with Phil Newell, and the rear wall ‘problem’ is more easily solved with some physical reduction in room volume.

If you really get stuck, feel free to message me here. I have a personal EASE license (no AURA) and am happy to advise without credit on the results. At the very least, a half-hour chat is always free with me - I would always rather people get better sound than not. It's why I'm a poor business man ;)
 
OP
Trdat

Trdat

Addicted to Fun and Learning
Forum Donor
Joined
Sep 6, 2019
Messages
967
Likes
396
Location
Yerevan "Sydney Born"
Hey Kyle, I wrote you a private message by the way.

Yes, I noticed many of your comments are around bass array and PA design.

Just to get this correctly, when you say non of the models include transference or scattering coeficients you mean that they do not simulate the reflected spectral balance?
 
Top Bottom