John Sayers' Recording Studio Design Forum Archive

Hi everyone,

I am a desperate man who is eager for any and all assistance - I am not beyond paying for consultation, but as you will see below, finances are tight.

- The Sob Story:

I work for a smallish government high school in the Eastern Suburbs of Melbourne, Australia, and we have had a new Performing Arts facility built. Being at the mercy of the Department of Education, we had a scenario where we had to pack too much into too small a space, and instead of building from scratch we had to 'refurbish' and rebuild from a crumbly old gym. Fast forward a little over a year and the budget went way over, corners had to be cut, important considerations (how the rooms actually sound!!) were overlooked... blah blah blah. It's not all bad, but the Music Dept has been left with some rooms that desperately need treatment. And we have very little money to deal with it. We had a consultant who actually made the situation more difficult by suggesting terrible terrible ideas, but we are now at a point where the school understands that SOMETHING has to be done, and it's now my job to figure out what to do. (I'm a Music Teacher)

- The Room(s):

I will go one step at a time. There are a few rooms which need some serious, SERIOUS assistance, but the most urgent is our classroom. We run a contemporary performance-based program. Basically, the students make rock and pop bands (drums, dirty guitars, yelling etc) and perform every lesson. So it gets loud.

The dimensions of the classroom are:
16.5m (north to south) x
7.3m (east to west) x
around 8.2m (up and down).

Sorry, I still can't seem to find the exact vertical dimensions, despite my digging, but I believe it is close enough to work with. The long west wall is plaster, as is the ceiling. The other walls are brick and the floor is concrete with carpet tiles.

How bad does it sound??? It's an echo chamber!! Below are the results from a test performed by the aforementioned consultant:

Octave Band Centre Frequency (Hz) 125 250 500 1000 2000 4000 8000
Reverberation Time (Seconds) 3.0 3.3 3.4 3.4 2.3 2.0 1.2

I am yet to do my own tests because I haven't managed to get my hands on a suitable microphone yet. I did take an amp and a signal generator into the room. The lowest the amp could give me was around 60 Hz, and after I stopped the sig gen the sound just kept going... and going... and going.


- The Plan

I have endeavoured to sketch out a plan for how to treat this space, but I need some confirmation that I'm on the right track. I have laboured for more hours than I'd like to mention trying to make headway with Sketch Up, but I still don't have anything resembling our room. I'll keep trying, but in the meantime I have done some rough jpegs that will hopefully give you an idea. Even those took me hours to put together.

I've looked around at what is available in Aus, as well as what is affordable and preferably safe to handle and have landed on Tontine Acoustisorb 2 in both 50mm and 100mm depths. To keep costs down and as a teaching opportunity I intend to get students to help build broadband absorption panels with the 50mm, and bass traps with the 100mm. The plan is to follow the tried and true timber frame stuffed with the tontine covered with cloth method. The panels will be 1200mm x 2400mm.

The pictures below don't show the bass traps, which I intend to put 6 around the wall/ceiling bi-hedral corners and then another 4 as a super-chunk style up the wall/wall bi-hedral corners. Also, someone at school had the idea of using an old theater set backdrop on one of our walls. I thought it could be cool as long as it's breathable and could hide our panels underneath.
- The Help?

I feel a little like I'm stabbing in the dark. I'm pretty confident that going through with my plan will yield a positive result - but is it the best result for the money, and will it be enough? Here are a few questions:

1. Is there anyone in Melbourne who can offer assistance? After having such a hard time with our initial 'expert', I'm hesitant to just dig through Google and talk to anyone. I recognise that we will likely need to pay someone for their time, but hopefully there is someone out there who is happy to provide advice and an action plan who can then leave it to us to carry out - thereby saving us some $$. If you are such a person, please feel free to PM or post here.

2. If no luck with question 1, how is this looking? Are there better materials we should be using? Is just trying to cover a good percentage of the surface with absorption and bass-trapping corners going to be enough to make the room usable? It's hard to be specific with our ideal scenario - but we want to play music intelligibly in there, and right now it's too live for a maths exam!


Apologies for the essay-length post. Feel free to tell me to bark up another tree, but I'm not really sure where to turn.


Kind regards,
Pete

Hi Pete, and Welcome to the Forum!

Boy, do you know how to pick 'em, or what?

Sadly, you are not alone: If I had a dollar for each person who came to the forum with stories about cut budgets, advice from acoustic "experts", and heart-breaking results... well, I wouldn't be a rich man, but I could sure get a few really nice dinners in some great restaurants!

So you might get some consolation from knowing that you aren't the first and won't be the last to arrive on the shores of the forum, panting and gasping, desperately seeking salvation (and an Aspirin).

The good news is that we seem to have helped the vast majority of those other folks get usable acoustics for their rooms, so I'm pretty sure you'll find the help you need here, just like they did. The other good news is that it probably can be done at low cost, and perhaps less expensive than you are planning on right now.

OK, so since you had a not-so-happy experience with the most recent acoustic expert, it would probably be wise to get a second check on those decay times you mentioned. From what you describe about the room construction, I'd expect them to actually be a bit longer than that. I'm wondering if there was some furniture, instruments, people or other stuff in the room at the time of the tests, that could have altered the results? The baseline tests should be done in a totally empty room: just the walls floor and ceiling.

So here's the way to do it: Go to the Home Theater Shack web site, and download the application called REW (for "Room EQ Wizard)

http://www.hometheatershack.com/roomeq/

It's free! And it's about the best darn acoustic analysis tool you could hope for, paid or not. Years ago, I paid a lot of money for other acoustic software that turned out to be not nearly as good. I have no idea why HTS gives away for free such great software, but they do, and I'm very grateful.

So install that on the PC of your choice that has a good audio interface on it, select your best mic that has an omnidirectional polar pattern, select your best speaker that has reasonably flat response and covers most of the spectrum, and go test your room! REW includes a PDF manual that walks you through how to do the calibration procedure, but if you find that too complex then let me know and I'll e-mail you my own set of instructions on how to do that.

So set up your speaker down low in one corner of the room (floor-wall-wall cornier), and set up your mic up high in the diagonally opposite corner of the room (ceiling-wall-wall). In other words, you want them as far apart as they can possibly get in that room, on the diagonal. (If you don't have an omni mic, then buy one! A good acoustic measurement mic costs around US$ 100 or so. And if you can't buy one fast enough, then do the initial tests with the best mic you have that ahs the flattest response and the pattern that most closely resembles omni. That's not great nor accurate, but can at east help get a basic idea of the room.

Now run a test with REW. Set up the start-delay in REW to give you enough time to get out of the room and shut the door, plus another 5 seconds for all the echoes of your noise to die away and things to get calm inside. REW will then do a sine sweep through the entire spectrum, and produce a series of wavy lines that I need to see. Save the data file from REW (it will have the extension "MDAT"), and upload it to some file-sharing service such as Dropbox, then post the link here, or PM it to me. I'll download the file and analyze it, to figure out what is up with your room.

OK, but don't send the file yet! I need another test in there too. REW allows you to do several measurements and keep all of them all inside one file.

You did the first test with the mic and speaker at diametrically opposed points. You did that for a very good reason: All room modes terminate in the corners, so by using those two "tri-corners" you are totally sure that you got them all. The speaker should have triggered all of the modes that want to be triggered, and the mic will have picked up the response for all of them. That's the worst-case scenario, and it tells me all about the low frequency response of the room, so we can figure out the best plan for your bass traps to damp that a bit.

But I'd also like a more realistic measurement of how the room sounds to normal people who don't play their guitars while kneeling down in one corner, nor listen with their ears glued to the ceiling in the other corner! So set up your speaker in a typical place where you'd normally have musicians making a lot of noise, not to close to any of the walls if you can help it, and set up the mic a fair distance away, at ear height, in a typical location where someone might be listening to the musicians making a noise, also not too close to the walls, if possible. Hopefully the mic and speaker will be several meters apart, and not on the room center-lines. Run another REW test like that, then save and send me the MDAT file with both tests (or rather send the link to the place where I can download the file).

The second test will show very different wavy lines on the graphs, but the combination of both sets of lines will tell me a lot about the room, and we can then figure out what to do to treat it.

OK, now about the treatment: You mentioned a specific acoustic product, which would probably do the job, but there are likely less expensive options down at your local building supply store. Go take a look at what they have in their insulation section, in terms of typical fluffy fiberglass insulation, and typical not-so-fluffy mineral wool insulation. Let us know what brands and product numbers they have in stock, and what thicknesses and densities, so we can see if any of that is useful for you. Compare the cost per cubic meter of each of those, against the cost per cubic meter of what you were planning to use, and you'll probably notice a worthwhile difference! If they happen to stock Owens Corning OC-703, OC-701, and OC-705 then your worries are over! You have found exactly what you need. But if not, then let's look at what they do have, and figure out what it can do for you.

Now for the theory: For a room with your dimensions, to get the decay time down to about half a second (what you need for a critical listening room) you'd need about 3200 sabins of absorption. That works out to about 290 square meters of perfect absorption. There's two problems with that: firstly, there is no such thing as a "perfect absorber", and secondly that would leave your room too dry for teaching music. I'd aim for something a bit more live than that, say around 800ms to maybe 1,200 ms, somewhere in that region. That would give you the best of both worlds: For speech intelligibility (so your students can actually understand what you are saying when they AREN'T killing you with noise), you need about 900 ms for that room. But for music to sound at its best in there, you need about 1,200 ms. So let's split the difference and call it one second. That would be a good target to shoot for.

To get 1 second decay time in that room, there are several equations for figuring it out, but the two best-regarded ones say 160 m2 and 140 m2 of absorption, respectively. So let's call it 150 to make them both happy.

So, figure that you will need about 150 m2 of absorption in your room, spread roughly evenly around the room AND AROUND THE SPECTRUM! (Perfect absorption would absorb all frequencies identically, but there is no such thing, so we'll have to use a combination of imperfect things over a larger area, to accomplish the task).

Your total surface area of the room is about 630 m2, and the floor is about 120 m2 of that, so there's 510 m2 left. But you have those doors and windows, and likely whiteboards and closets and other things that get in the way too, so you probably have something like 400 to 450 mt2 of actual open room surface that you can treat. You'll need to cover about one third of that with broad-band absorption, to get the results you want. But you can't just slap up the exact same thing everywhere!

HOWEVER! You already have some really lousy treatment in there, that isn't helping you at all. You said "the floor is concrete with carpet tiles". The concrete is great, but those carpet tiles are nasty things: they suck out a lot of high frequencies, randomly, but not much in the mids, and zilch from the low end. That's not good. Do you have to keep those, or could they go? Would it be possible, at some stage, to take out the carpet and lay laminate flooring in there instead? Or ceramic tile flooring, or linoleum flooring? They would all be better, acoustically, and also probably nicer visually! If you can't do that now, then you can live with it (grudgingly), but it would be nice to think about it for the future.

Assuming you can't do anything about the carpet, we have to be careful to not kill the high end too much, since it is already being killed by way to much randomly selective absorption (carpet). So you'll need light-weight insulation (lower density) than normal to ensure you absorb more bass and less treble, and perhaps also some reflective or diffusive panels of some type in front of some of the absorption, for the same reason. Poly's might be an option, or even "slot wall" type devices, or something similar (but not tuned). Maybe curved plywood panels hung from the ceiling, with absorption on top. Etc. Maybe even panel traps, perf panel, ... Lots of options to keep things smooth.

I did mention above that it isn't just an even spread around the room that you need, but also an even spread across the spectrum. You want to ensure that they decay times for all of the frequency bands are similar, and follow the normal pattern. But it isn't that easy: the things that treat in frequency, also treat others that you didn't want to treat.

Bass traps, for example are usually very large in size, and are designed to deal with low frequencies very well. But they also deal even better with high frequencies, so they need to have something in front of them to let only the lows in, and reflect the highs back into the room. We'll deal with that when the time comes to build the devices, but it is something that you need to keep in mind. If you put up only absorption, then you'd over-treat the high end, and you'd be left with a room that sounds muddy, dull, honky, tubby, dry, and all those yucky things. The treatment has to be balanced to take care of the right amount of each frequency band, and that's why I need the REW data, so I can see how the room is behaving at present, identify which parts of the spectrum need what type of treatment, then come up with a plan to do it.

At first glance, your plan looks like it probably isn't too far off the mark. You are proposing a total of 37 panels, each measuring 1.2 x 2.4m, which totals about 107 m2. That's two thirds of the theoretical requirement, but it is also just one type and two thicknesses of panel, which means you'd be targeting the same frequencies with every single panel.

But to add to the mix: during normal use, you'll have a bunch of people in the room, and people absorb sound too. You'll also have furniture, which could absorb, or diffuse, or both. And musical instruments, which could do straaange things. And you have that huge amount of carpet, which is already skewing things in unknown ways. In other words, your plan would definitely help, but wouldn't go as far as you need to, or might go way too far in some places, and not nearly enough in others, since it isn't taking into account all of the variables. It's good, but not enough.

There's also the issue of how you build the panels. There's quite a difference between a slab of 50mm mineral wool all by itself out in the middle of the room, and the very same slab stuck in a wooden frame, covered with fabric, then hung on the wall. They both have rather different absorption curves. Drastically different. And you'd have yet another curve if you hung that exact same panel 15cm away from the wall, instead of right up against it. And yet another other OTHER other curve, if you cut large holes or slots in the wooden frame that holds the slab, exposing the edges.

So lots of variables, and they can all make a differences.

Yes, your plan would to accomplish something, but not necessarily everything you hope for. But you found us here, so you are in the right place to get your plan modified so it does do what you want.

So I'd suggest that you do those REW tests first, to see exactly how the room is behaving right now. If you are able, then do one set of tests with the room totally empty, and another set with the typical compliment of furniture, people, instruments, etc., so we can get a better handle on what the room is doing under different situations. When you do the tests with the people in it, instruct them to block their ears with their hands while you run the tests: there could be some unpleasant stuff going on, or even damaging stuff, potentially. So play it safe. Do the tests at a measure calibration level of 85 dB (using a hand-held sound level meter that you beg, borrow, or buy), but be aware that modal resonance could cause the level to be much, much louder than that at some specific frequencies, and that's not good for your ears.

While it would be nice if there were someone in Melbourne who could go do this for you, it really isn't necessary: I have tuned rooms remotely like this before from thousands of miles away, and it can be done, thanks to tools like REW and the internet.

So that's what I would suggest.

And one other thing: Please post a few good quality photos of the room, from different locations and angles, so we can get a better idea of what you are dealing with!

I'm looking forward to this: It's going to be fun!


- Stuart -

Wow! Thanks so much for your extremely helpful reply and your willingness to help, it is greatly appreciated!

My first stab at REW was slow, but I believe I managed to sort out the calibration of my soundcard, not sure I really got beyond that, but I've got some more reading and testing to do next week. I have an Omni (Rode NT2), but I wouldn't say it's all that linear - I've placed an order for the Behringer ultra-linear model, but it might take a week or two, so I could redo tests once I have that as well.

As far as insulation materials are concerned, the brands available are Tontine, Bradford, Autex, Knauf and Pink Batts (no OC in Australia as far as I can find). There is a good stockist in my area that can get most things - anything from 25mm to 100mm in either polyester, glasswool or mineral wool, anywhere up to about 60 kg/m3 I believe. The criteria I was looking at was their coefficients at low frequencies and whether they are low-irritant.

Anyway, I'll leave all that for once I have some tests. Your other suggestions of using diffusion and curved panels etc sound very interesting - I'm quite keen to learn more and more and don't mind doing reading, I've already done plenty and feel like I've only touched the surface. So far, there is no furniture in the space, but there will be some once we have properly moved in - not going to do that until we can make real music. Here are some photos below.
Thanks again for your help - I'll post again once I get the REW results.

Cheers,
Pete

Oh wow. That's a nightmare! It must sound pretty awful in there right now!

But it can be made to sound good, though, so don't despair.

With those large opposing windows at both ends of the room, I'm thinking that maybe some hanging baffles in the middle of the room might not be a bad idea, in addition to the wall-mounted treatment.

I have an Omni (Rode NT2), but I wouldn't say it's all that linear

That's fine. It will work to get a basic idea of how bad the room is. The room is going to be so bad that the issues will be many, many, MANY times larger than the small non-linear deviations in the mic response. They will be absolutely visible in the data, no doubt at all. We are talking wild ocean waves smashing the shore in a force 5 hurricane, compared to ripples in a bathtub....

By the way, you are going to need a reaaaaaaallllly tall mic stand to get your mic up into the ceiling corner, though! Maybe a ladder with a clip of some type on it? Scaffolding? The mic needs to be pretty much right in the corner with just a couple of cm from gap on each side (to both walls and to the ceiling).

I've placed an order for the Behringer ultra-linear model

The ECM-8000? That's their acoustic measurement mic. It's not too bad, but I prefer something a bit better quality. I have a Presonus PRM1, and that's pretty good. The DBX RTA-M is also a good one. Earthworks has some nice ones too. There are plenty of others too, but some of them can get a bit pricey! You just need a basic one, like the Presonus or DBX.

anything from 25mm to 100mm in either polyester, glasswool or mineral wool, anywhere up to about 60 kg/m3 I believe.

Sounds like you have a good selection! When the time comes, we can take a closer look at the options.

The criteria I was looking at was their coefficients at low frequencies and whether they are low-irritant.

You'll be needing a few different types of treatment, not just for the lows (although that will be a large part of it), as you'll have several different issues going on there, at different frequencies and in different places.

Also, don't worry too much about the health factor: The scare-mongering is quote a bit overblown. It's worth taking a look at this:

http://www.gearslutz.com/board/bass-tra ... eview.html

That's not to say that there aren't issues: there are, but they can be dealt with as part of the design of your devices. For examples, anything that goes overhead will have plastic on the front, to ensure that the fine fibers that are shed over time don't filter down onto your instruments and students. All of them will have fabric on them, which is a second barrier, and if you can get semi-rigid insulation, then that doesn't shed much anyway. So don't sweat that aspect too much.

I'm quite keen to learn more and more and don't mind doing reading

Cool! There's enough reading right here on the forum to keep you going for many months! It's actually amazing how much you can learn by following the build threads. Some of those run to dozens of pages, and they are packed with useful stuff.

But if you want a more theoretical background, then I'd suggest "Master Handbook of Acoustics" by F. Alton Everest. That's sort of the Bible for acoustics.

I've already done plenty and feel like I've only touched the surface.

Don't worry: it gets worse as you learn more! I've been at it for many years now, and the more I learn, the more I realize that I don't know very much! It's like one of those terrible nightmares, where no matter how fast you run toward it, the exist always seems to be getting further and further away...

By the way, when you do the REW tests, make sure you carefully measure the exact physical locations of the speaker and the mic, so you can get them back to the identical points in the future, for subsequent tests as you treat the room. They must be in the same locations for all tests. If not, then you cannot validly compare the future tests against the baseline test you are doing now.

I'm looking forward to seeing the MDAT file!


- Stuart -

Awful is an understatement - it is alarmingly bad!

It is the ECM8000 I have on order, I'll try and see if I can change it with the DBX or Presonus if it isn't too late.


That's a bummer about the mic needing to be right in the corner, and in a repeatable position. No scaffold available, but I should be able to get a ladder long enough, but not sure how to position the mic yet. Trial and error, I'm sure I'll get it eventually. Stacking tables and chairs on my longest boom stand was still a meter or two shy of the target.

I'll see if I can hunt down that Everest book, sounds like the ticket for my bedtime reading! It's also nice to know that the fibreglass isn't as bad as I thought, because it's pretty easy to come by.


Thanks again for your help, I should hopefully have that MDAT file soon - I'm still getting through all the REW reading so hopefully I can understand the results too!

Cheers,
Pete

It is the ECM8000 I have on order, I'll try and see if I can change it with the DBX or Presonus if it isn't too late.

If you can't swap it, then don't worry too much: It's not a bad mic. I have one too, that I use for jobs where it might get damaged: I save the Presonus for the more important jobs. But the ECM works OK, so don't sweat it too much if you are stuck with it.

That's a bummer about the mic needing to be right in the corner, and in a repeatable position.

I wouldn't worry too much about the repeatable part for that mic: It's more for the one in the room. Since that mic up high is basically for checking the low end response, where waves are many feet long, high precision repeatability isn't too important! But it is for the free-field mic, where high frequencies have very short wavelengths, measured in fractions of an inch.

not sure how to position the mic yet. Trial and error, I'm sure I'll get it eventually.

As long as you can get it fairly close to the corner, but not touching any of the surfaces, then you'll be fine.

I'll see if I can hunt down that Everest book, sounds like the ticket for my bedtime reading!

Here ya go: I'll make it easy for ya!

http://www.amazon.com/Master-Handbook-A ... +acoustics

I'm still getting through all the REW reading so hopefully I can understand the results too!

If you get stuck, let me know. I wrote a brief set of instructions on how to calibrate it a while back: I could send those to you: might help.


- Stuart -

Amazon to the rescue, thanks again!

If you could send through your calibration cheat sheet, that would be a great help - I'm enjoying all the learning of going through the Help Index point by point, but I might not be done for quite some time.

Thanks again, I'm so grateful for your assistance Stuart.