Trying to address a null in a relatively small studio. Help?

[CurtZHP]

First time poster... Hoping I don't screw up and forget anything right out of the gate.

I built this room in my basement several years ago. It's 176 inches long, 134 inches wide, and 85 inches high. Actually, the wall to the right of the mix position is slanted slightly toward the front of the room, so the width at the very front is 130 inches, while it's 137 inches at the back. (I figured 134 was a close average...) The entry door is to the left, 75 inches from the back; and there's another door leading to a closet/crawl space on the opposite wall (the slanted one) 30 inches from the back. Each door is 32 inches wide. At each end of the room there are soffits along the top of each wall, running the width of the room. These soffits are about 12 inches square. (These are for HVAC ducts.) The walls are standard drywall on studs, packed with fiberglass insulation, with the exception of the back wall, which is concrete. The concrete floor is covered with carpet. The ceiling is insulated with fiberglass and finished with fiberboard ceiling tiles.

Also, in the middle of the front wall, there's a small glass window, about 3 feet by 14 inches, looking into what was supposed to be a small tracking room.

About 3 feet from the front wall is the desk with the console and computer monitor. The speakers are located just behind the desk on stands.

As far as acoustic treatment presently in the room....
I built a bunch of DIY acoustic panels. They are all 2 feet by 4 feet by 4 inches. Each consists of a wood frame made of 4-inch by 3/4-inch boards. Inside the frame is mineral wool, and the whole thing is covered in cambric (the thin, tightly woven, black fabric you find on the bottoms of chairs). These panels are placed thus: one panel straddling each corner of the room, hanging from the bottom of the soffits. Two on either side of the mix position at the first reflection point, hung directly on the wall. Two panels above the console/desk, straddling the corner created by the soffit and ceiling.

Most of the work I've done in this room has been radio or small video projects, and a few of my own "demo" recordings. For most of the time I've been doing this, I hadn't noticed too many issues. But recently, a couple projects have brought some low end issues to my attention. So, I got myself a copy of REW, broke out the old Radio Shack SLM, and set to investigating. I did a few (probably horribly unscientific) tests of my own, and came to the following conclusion. I've got a pretty significant dip in response centered around 130Hz.

Visited all the usual sites on the interwebs, and decided that what I needed to do was build a few panel/membrane absorbers. Bought some lumber, bought some mineral wool. Built two (to start with) boxes, 70 inches by 24 inches by 6 inches. Sides, top, and bottom of 6-inch by 1-inch boards. Front and back of 1/4 plywood. In a few places, I had read that it was important to have a small air gap between the front panel and the mineral wool. The mineral wool I got was a little thicker than I thought it would be, so it essentially fills the box. I'd have to figure out a way to create the gap. But then, a few sites suggested that this wasn't necessarily necessary. ??? Ethan Winer, for one, suggests that, for small rooms, broadband absorption is better than trying to mess around with panel or tuned traps. Anyway, after putting the two new panels in room, I did see a slight improvement, especially by putting them in the front corners. Taking the plywood off the front made it even slightly better.

So, now, I'm sitting here wondering what to do next. Do I just proceed with my plan and hope for the best? Do I remove the front panel off of these and just create larger and thicker panels for the corners, foregoing the whole membrane absorber idea? Do I just get rid of all the gear, shut it down, and turn it into a gunsmithing shop??

So I found this place. Figured I should ask for help. I followed the instructions as posted here for setting up REW, as closely as I could without buying any additional gear. Put the room back the way it was and took a measurement. It confirmed by earlier measurements. Definitely have issues around 130Hz.

Apologies for the ridiculously long post right out of the gate. Thanks for sticking around up to this point. If there's anything further I can provide, please advise.

[Soundman2020]

Hi there "CurtZHP", and Welcome!!!

So I found this place. Figured I should ask for help.

You have, indeed, finally arrived at the right place.

Hoping I don't screw up and forget anything right out of the gate.

And you have, indeed, NOT screwed up anything! In fact, that was a pretty darn good first post, that succinctly described the room and the situation. The only thing that would make it even better, is some photos of the room, and an accurate diagram, or if you really wanted to make it "outstanding", then adding a SketchUp model would do the trick!

Seriously, that's a great first post. I wish all new members would take the time and effort that you did.

OK, first things first: Did you plug in your room dimensions to a room mode calculator? Use one of these to predict the modal distribution for the dimensions of your room:

http://www.bobgolds.com/Mode/RoomModes.htm

http://amroc.andymel.eu/

Both of those are very good. They give you tons of information that is really useful to help figure out what's going on with your room.

I already did that using your dimensions, and this is what I got:


39.0 hz (1,0,0 Axial)
50.6 hz (0,1,0 Axial)
63.9 hz (1,1,0 Tangential)
77.9 hz (2,0,0 Axial)
79.6 hz (0,0,1 Axial)
88.6 hz (1,0,1 Tangential)
92.9 hz (2,1,0 Tangential)
94.3 hz (0,1,1 Tangential)
101.3 hz (0,2,0 Axial)
102.0 hz (1,1,1 Oblique)
108.5 hz (1,2,0 Tangential)
111.4 hz (2,0,1 Tangential)
116.9 hz (3,0,0 Axial)
122.3 hz (2,1,1 Oblique)
127.4 hz (3,1,0 Tangential)
127.8 hz (2,2,0 Tangential)
128.8 hz (0,2,1 Tangential)
134.5 hz (1,2,1 Oblique)
141.4 hz (3,0,1 Tangential)
150.2 hz (3,1,1 Oblique)
150.5 hz (2,2,1 Oblique)
151.9 hz (0,3,0 Axial)
154.7 hz (3,2,0 Tangential)
155.9 hz (4,0,0 Axial)
156.8 hz (1,3,0 Tangential)
159.2 hz (0,0,2 Axial)

Those are your first 26 modes, up to about 160 Hz, and I don't see anything suspicious at around 130 Hz, so I'm inclined to suspect that it's either an SBIR issue, or the typical floor bounce or ceiling bounce that is almost inevitable in any small room.

So let's see your REW data, and compare how well theory matches reality. You say you already did some REW tests, and I'm hoping you followed these instructions here: http://www.johnlsayers.com/phpBB2/viewt ... =3&t=21122

If so, please upload your REW file to a file sharing service, such as Dropbox, and post the link here, so I can download them and take a look.

...and 85 inches high.

You probably already know this, but that's REALLY low, and not too good for acoustics. However, how is your ceiling buit? Do you have drywall up there? You mentioned "fiberboard ceiling tiles: Do you have a link to data about those? When you say '85" ', do you mean from the surface of the floor up to the bottom of the fiberboard, or the bottom of the joists, or the bottom of the subfloor above you?

About 3 feet from the front wall is the desk with the console and computer monitor. The speakers are located just behind the desk on stands.

Speakers on stands is great, but your room is very small, so your speakers are in the wrong place. The room is too small to be able to have them located away from the front wall, because you cannot get enough distance between the speakers and the wall to force the SBIR dip down low enough. Assuming that your speakers are actually about 2'6" from the front wall, that would give you an SBIR dip at around.... welll... wait for it.... did you guess? .... Roughly 130 Hz!!!! I'm not saying that this is your problem, but it's a good possibility.

In a small room, your best bet is to have the speakers tight up against the front wall, as close as you can get them, with just a thin 4" gap where you insert a panel of 4" OC-703, between the rear corner of the speaker and the front wall. That forces the SBIR dip up to a higher frequency hopefully around 400 Hz or even higher, which is great. Number one, it's not so noticeable up there, as it's firmly in the mid range, and number two, it is now high enough that you can treat it with suitable insulation, hence the 4" of OC-703!

So do a test like that: Do nothing else except slide your speakers up to the front wall with a 4" panel pf 703 in there, set them exactly 38" away from the side walls (location of acoustic axis, not edge of speaker cabinet), which means they will be 54" apart. Not great, but a decent starting point. Now aim them carefully so the acoustic axes of both speakers intersect about a foot behind the back of your head. That places your ears on-axis, and should give you a reasonably accurate stereo image and sound stage.

Do another set of REW tests like that, and also post those results.

Inside the frame is mineral wool,

What density? That's a key point. You need to know the density of the insulation that you used. Not the R value, which has no meaning for acoustics, but the weight per unit volume. Also, how thick is that insulation?

and the whole thing is covered in cambric (the thin, tightly woven, black fabric you find on the bottoms of chairs).

Cotton or nylon? I have seen some cheaper chairs with course weave nylon "fabric" underneath, but that's not much use acoustically. Photo? Link? Make and model of speakers?

one panel straddling each corner of the room, hanging from the bottom of the soffits.

At what height? You mentioned that the panels are rather small (only 4 feet high), so they are not tall enough to cover the full floor-to-ceiling distance. How high did you hang them? In reality, your corner traps really do need to cover the entire floor-to-ceiling distance. All room modes terminate in corners, so treating the corners is key, and treating the "tri-corners" es even more key than "key"! The "tri corners" are the eight places in a rectangular room where two walls meet the ceiling, or two walls meet the floor. Are you covering your tri-corners?

Two on either side of the mix position at the first reflection point, hung directly on the wall. Two panels above the console/desk, straddling the corner created by the soffit and ceiling.

What about the back wall? That's the most important wall of all in a small room: What treatment do you have on the rear wall?

So, I got myself a copy of REW, broke out the old Radio Shack SLM, and set to investigating.

Great! What mic did you use for the REW tests?

Visited all the usual sites on the interwebs, and decided that what I needed to do was build a few panel/membrane absorbers.

You probably visited the wrong sites, then, if that's what they told you to do in a small room with probable SBIR issues!

Bought some lumber, bought some mineral wool. Built two (to start with) boxes, 70 inches by 24 inches by 6 inches. Sides, top, and bottom of 6-inch by 1-inch boards. Front and back of 1/4 plywood.

Technically, that's more of a panel trap than a membrane trap, but that's splitting hairs.... Anyway... What surface density is your plywood? Did you measure that? To calculate the frequency accurately, you need to know exactly how much each square foot of your plywood weighs. Not how much is predicted by tables, or graphs, or equations, but rather the ACTUAL density of YOUR plywood.

In a few places, I had read that it was important to have a small air gap between the front panel and the mineral wool.

That depends on what you are trying to accomplish...

The mineral wool I got was a little thicker than I thought it would be, so it essentially fills the box.

And what density is that wool? You need to know that, in order to estimate the Gas Flow Resistivity numbers, so you can accurately predict the frequency, Q and performance.

I'd have to figure out a way to create the gap. But then, a few sites suggested that this wasn't necessarily necessary. ???

That depends on what Q you want for your device. If you want a broad bandwidth (low Q), then fill the box. If you want a sharper, higher Q, then don't fill the box, and leave a gap. The empty air in the "gap" behaves differently than the air trapped inside the mineral wool: sound behaves differently as it moves though those two types of "air". The actual physical properties of the air are different: energy transfer happens adiabatically in the empty air, but isothermally in the insulation. Different process, different results, different tuning.

Built two (to start with) boxes, 70 inches by 24 inches by 6 inches Sides, top, and bottom of 6-inch by 1-inch boards.

What was your predicted resonant frequency for that combination, along with your plywood and insulation? Did you actually calculate that, or did you just get it from a table or graph?

Front and back of 1/4 plywood.

That won't work! You have TWO tuned panel traps opposing each other on opposite ends of the box. A panel trap must have a very solid, massive, rigid back on it to work. One that cannot bend, flex or vibrate anywhere near the range you want to tune the box for. Probably 1" MDF would have been a better choice for the back panel.

Also, did you seal the box totally airtight, carefully caulking all seams, joints and holes? If not, it won't work. Panel traps work on the principle of "mass-spring-mass". The front panel is one "mass", the thick, rigid, immobile rear panel is the other "mass", and the air trapped between them is the spring. But it won't be a spring if the box is not sealed air-tight!

Ethan Winer, for one, suggests that, for small rooms, broadband absorption is better than trying to mess around with panel or tuned traps.

Absolutely! Yes, Ethan is correct. Firstly, trying to tune a resonant device of any kind is a major challenge, as there are so many variables to take into account, and materials are not perfect. That's even assuming that you managed to identify the offending issue correctly and accurately... Plus, in a small room you don't just need to treat one frequency: you need to treat many. And there's not enough physical space to put all of that treatment in! Plus, each device has to go in the correct location: there's no point in putting a device on the rear wall that is tuned to a problem that only occurs between the floor and the ceiling! Since the device will never "see" that problem, it won't do anything at all, even if it is perfectly tuned to the exact frequency.

Broadband porous absorption is the way to go in a small room.

Anyway, after putting the two new panels in room, I did see a slight improvement, especially by putting them in the front corners.

Please post the "before" and "after" REW data for those tests as well.

Taking the plywood off the front made it even slightly better.

... which implies that they weren't working very well as tuned devices to start with! Perhaps because they were not tuned close enough to the problem frequency, or the Q was too low, or several other things. If they work BETTER as broadband, then that's a clear indication that they were working pretty lousy as tuned devices, if at all. The results you saw might simply have been because they were occupying volume in the room, thus changing the acoustic response so the room just be being there, but not because they were tuned to the right frequency.

As near as I can tell, a 1/4" of plywood over a 5 1/2" cavity filled with insulation would be tuned to around 61 Hz, so it's hard to see what effect that would have on 130 Hz. Not to mention that if your problem really is SBIR or floor bounce, a panel trap would do zilch to treat it, no matter where you put it in the room, nor how well it was tuned: panel traps cannot fix SBIR or floor bounce, since those are phase cancellation problems, and a panel trap can't do a single thing about phase cancellation....

So, now, I'm sitting here wondering what to do next. Do I just proceed with my plan and hope for the best?

"Hoping for the best" is never a good approach for tuning a room! The chances for success are somewhere between "zero" and "nothing at all". Maybe even as high as "nope"! It's far, far better to first analyze and identify the real problem, then treat that

... foregoing the whole membrane absorber idea?

Yes. At this stage, it seems unlikely that you are dealing with a resonant issue at 130 Hz, so a resonant trap is not going to fix it. I may be wrong, but I'll only know for sure once I see your REW data, but from everything you say, I'm suspecting that this is not a resonant problem. Trying to fix SBIR with a panel trap is like trying to fix a leak in your roof by stapling flyscreen over the hole... a lot of work, for no effect at all!

Do I just get rid of all the gear, shut it down, and turn it into a gunsmithing shop??

Now THERE'S an idea! But you won't be able to make much music like that...

Apologies for the ridiculously long post right out of the gate

Not long at all! No apology necessary. If you want to see a REALLY long post, go searching for my first one... or actually, pretty much any of my posts! I don't think I'll win any prizes for minimizing word usage. After all, why use only three words when fifty will do the job so much better?

If there's anything further I can provide, please advise.

Photos. Diagram. SketchUp model. Make and model of speakers. But most of all, REW data.

OK, so here's the plan: Move your speakers and set them up accurately as I outlined above, then follow the procedure for using REW that I linked to above, then post ALL of your REW data (the stuff you already had, plus the new stuff), and when I get a chance during the week I'll analyze that and get back to you with what I see, perhaps asking you to do some more REW tests to confirm it. Then based on that, we'll come up with a plan for treating your room. In the meantime, please answer all the questions I posed above, and add the photos, diagrams, etc.

Hopefully we'll be able to nail this guy! If it is SBIR, it's hard to fix but not impossible. If it is floor-bounce, that's a bit tougher. If it is modal, that's simple (relatively). If it is flutter echo, that's dead easy. If it is something else... well, let's hope it isn't "something else"....

- Stuart -

[CurtZHP]

Thanks so much for the informative reply. A lot of information to go over. Like drinking from a fire hose!

The only thing that would make it even better, is some photos of the room, and an accurate diagram, or if you really wanted to make it "outstanding", then adding a SketchUp model would do the trick!

I'll whip up some drawings and dig up some pictures over the next couple days and get those posted.

Did you plug in your room dimensions to a room mode calculator?

I fooled around a little with the room simulator in REW, but since I clearly know just enough about this to be dangerous, I wasn't sure of what to do with the results. Thanks for going ahead and running the numbers for me already. What you had to say about it is already a giant step forward.

....and 85 inches high....You probably already know this, but that's REALLY low, and not too good for acoustics. However, how is your ceiling buit? Do you have drywall up there? You mentioned "fiberboard ceiling tiles: Do you have a link to data about those? When you say '85" ', do you mean from the surface of the floor up to the bottom of the fiberboard, or the bottom of the joists, or the bottom of the subfloor above you?

Yes, I know that's not ideal, but I couldn't raise the ceiling or lower the floor, so there it is. As for the actual construction, I put R-23 or R-30 fiberglass (can't recall which...) between the joists and covered the joists and insulation with the fiberboard ceiling tiles. These tiles are made of similar material to drop ceiling panels but are designed to be flush mounted to either bare joists or existing drywall ceilings. They're relatively smooth as far as surface texture goes, and far lighter than drywall. When I measured 85", that was from the carpet to the tiles.

The room is too small to be able to have them located away from the front wall, because you cannot get enough distance between the speakers and the wall to force the SBIR dip down low enough. Assuming that your speakers are actually about 2'6" from the front wall, that would give you an SBIR dip at around.... welll... wait for it.... did you guess? .... Roughly 130 Hz!!!! I'm not saying that this is your problem, but it's a good possibility.

I wouldn't be surprised if this is at least contributing to the problem. I wouldn't mind pushing them back at all, since that would free up much needed space in the room. I had them away from the wall because I thought that's what you're supposed to do with them. SBIR? I had to look that up! This is the only place so far that I've even come across that term! And now that I know what it is, I can take that into consideration.

So do a test like that: Do nothing else except slide your speakers up to the front wall with a 4" panel pf 703 in there, set them exactly 38" away from the side walls (location of acoustic axis, not edge of speaker cabinet), which means they will be 54" apart. Not great, but a decent starting point. Now aim them carefully so the acoustic axes of both speakers intersect about a foot behind the back of your head. That places your ears on-axis, and should give you a reasonably accurate stereo image and sound stage.

Do another set of REW tests like that, and also post those results.

This will be an easy jumping-off point except for one thing. I haven't got any 703 handy. Can I get away with 4" of mineral wool, at least for preliminary testing purposes?

What density? That's a key point. You need to know the density of the insulation that you used. Not the R value, which has no meaning for acoustics, but the weight per unit volume. Also, how thick is that insulation?

The insulation in my existing panels is 4". The insulation in the new ones I built is 6" thick. As for its actual density, I'll have to look into that.

Cotton or nylon cambric? I have seen some cheaper chairs with course weave nylon "fabric" underneath, but that's not much use acoustically. Photo? Link? Make and model of speakers?

I believe it's nylon. I got it from my father's upholstery shop. It's high quality stuff with a pretty tight weave. (I used to work for him, so I know what you mean by that coarse weave junk...)

You mentioned that the panels are rather small (only 4 feet high), so they are not tall enough to cover the full floor-to-ceiling distance. How high did you hang them? In reality, your corner traps really do need to cover the entire floor-to-ceiling distance. All room modes terminate in corners, so treating the corners is key, and treating the "tri-corners" es even more key than "key"! The "tri corners" are the eight places in a rectangular room where two walls meet the ceiling, or two walls meet the floor. Are you covering your tri-corners?

The corner panels are hung from the soffits at each end of the room, so they are basically about a foot from the actual ceiling, and about two feet from the floor. From what I've been reading here and elsewhere, I've learned that this is not adequate, which is why I built the new panels to go from floor to ceiling.

What about the back wall? That's the most important wall of all in a small room: What treatment do you have on the rear wall?

Here's the bad news. Other than the corner traps, there's nothing on the back wall except for shelving from floor to ceiling. You'll see in the pictures.

What mic did you use for the REW tests?

More disappointing news. I used the mic on the sound level meter, because it's what I had.

Technically, that's more of a panel trap than a membrane trap, but that's splitting hairs.... Anyway...

We'll leave any discussion of what I was trying to build aside for now, since I was obviously headed in the wrong direction and was doing it bass-ackwards anyway!

Broadband porous absorption is the way to go in a small room.

\

That's what I'm finding out.

But you won't be able to make much music like that...

I guess that depends on your idea of music!

Before we go any further with REW data, what's the story on using the mic in the Radio Shack meter for such tests? If it's utter crap, then I can certainly try to find something better.

Incidentally, I did another test that gave me similar results. Basically just ran pink noise through speakers and then ran the audio from the RS meter to an input on my DAW, into the recording software as a live input. Then put Voxengo's SPAN plugin across that. Same nasty hole around 130Hz.

Oh, and to answer your question about what speakers I'm using: Event 20/20 passives powered by a Crown D-75.

[Soundman2020]

Like drinking from a fire hose!

Welcome to the wonderful world of acoustics, where nothing is ever as simple as it seems! Get used to that "fire hose" feeling....

I fooled around a little with the room simulator in REW,

It's a useful tool, but it doesn't give you the full picture,like Bob Golds's Andy Mel's calculators do. The REW simulator is meant for a different purpose, and is actually dealing more with reflections and phase cancellation issues, based on speaker and listener positions, but the mode calculators are more about the room response itself, as a whole, everywhere, independent of mics and speakers. The room as a unit is a resonant system (more accurately, a reverberant system), and it has multiple manners in which it can reverberate / resonate. That's what the room mode tools deal with, and the REW sim doesn't do so well. On the other hand, the REW sim can do things that the mode prediction tools can't, such as help you figure out where to set up speakers and mix position, and what treatment to put on each wall.... to a certain extent. But all of those are just theoretical predictions: Theory is a nice comfy sofa, but reality is a bed of nails, and sometimes doesn't match theory so well...

These tiles are made of similar material to drop ceiling panels but are designed to be flush mounted to either bare joists or existing drywall ceilings. They're relatively smooth as far as surface texture goes, and far lighter than drywall. When I measured 85", that was from the carpet to the tiles.

That's pretty much what I suspected. Those tiles are not doing much at all, and the actual acoustic height of your room is really to the underside of the sub-floor above you. The reason I say that is simply because "light weight" materials do not stop sound. Only massive, heavy, solid, rigid materials do. So the low frequency waves (which are the most critically important for room acoustics), are going straight through those tiles, and what they "see" is the hard, solid, rigid, massive sub-floor above you. That's your real room height, as far as room modes go. In fact, you could take off those tiles and probably improve the room acoustics considerably, since that would expose the insulation to the room, as well as the joists, so you'd be getting diffusion, more volume, and more absorption. That's not what "acoustic tiles" are meant for, and they aren't really doing you any favors, apart from making the room look nice. But you could make it look equally nice by putting nice fabric up there, like this:

Inside-out-ceiling-04.jpg

Inside-out-ceiling-06.jpg

Inside-out-ceiling-10.jpg

That's from the studio of one of my customers in Australia, showing how he did his ceiling.

I had them away from the wall because I thought that's what you're supposed to do with them.

If the room is large enough, yes, that's the best position by far. But it has to be rather large! Depending how low your speakers go down to in frequency, you need to get them at least 7 feet from the front wall, and probably more like 10 feet if they go down really low. So you'd need a room about 30 feet long to do that... That's the part that the "helpful" websites and basic guidelines don't tell you! They mention the "rule" that it's better to have your speakers away from the walls, but without checking the facts to find out how it REALLY works! As the saying goes: That's fine in theory, but how does it work in practice?

SBIR? I had to look that up! This is the only place so far that I've even come across that term!

Sorry, I should have made that a bit easier for you by explaining it! Yes, you won't find that mentioned in many places: it only comes up on forums and websites that actually discuss real acoustics, rather than fantasies involving carpet on the walls and egg crates on the ceiling....

SBIR is Speaker Boundary Interference Response. It's what happens when the low frequency sound wave from the speaker spreads out, hits a wall, and comes back again, then interferes with itself, creating patterns of nulls and peaks in the frequency response inside the room. At the points where the wave cancels itself out, there's obviously a null where the sound level takes a rapid dive at a specific frequency that is related to the location of the speaker and the location of the "listener". That's what I'm suspecting in your case. The two biggest causes of SBIR are the front wall and the back wall, followed by the side walls. Depending on the speakers and the room, the front wall might have a larger effect, or the back wall might have larger effect. I have developed a method that helps to figure out what the cause is, and I call that the "walking mic" test. When you have the previous testing done and analyzed, I'll let you know how to do that, so we can take a closer look at your problem.

This will be an easy jumping-off point except for one thing. I haven't got any 703 handy. Can I get away with 4" of mineral wool, at least for preliminary testing purposes?

That's probably fine for the first approximation, but for the final treatment you'd likely be better off with 703. It depends on the GFR of your mineral wool, of course: If that happens to be similar to the GFR of 703, then you'll be just fine!

GFR = "Gas Flow Resistivity", which is the actual technical characteristic of the insulation that governs how it attenuates sound. It's a measure of the acoustic impedance of the insulation. Basically, it's related to the changes in pressure and the flow rate of air moving through the insulation, and since sound waves are sort of like "air moving", it tells you a lot about how the acoustics of that material work. It is measured in the rather cryptic and unlikely units of "MKS rayls", and most manufacturers of insulation never bother measuring it, since it has little meaning for the primary reason they make the stuff: thermal insulation. It's only useful to us folks that want to use their products from the wrong reason: acoustic treatment. Some do measure it, and there are tables of the GFR of various products that can be very useful., but mostly it's really hard to find that information for any given material. However, there is a very rough non-liner relationship between the density of each specific type of insulation, and the GFR, so as long as you know the density and what type of insulation it is (fiberglass, mineral wool, etc.), you can roughly estimate the GFR. That's the reason why I asked you for the density: so I can estimate the GFR, and see what results you should be expecting. Unfortunately, there's no relationship at all between the R rating of the insulation, and GFR.

I believe it's nylon. I got it from my father's upholstery shop.

Hmmmm..... they key point is that it must be "breathable", meaning that if you hold a piece of that fabric over your mouth and nose, you should be able to breathe almost normally, without noticing too much extra resistance. That nylon stuff usually isn't very breathable. You might want to switch to a different fiber type. The reason is rather simple: you want sound to get through that stuff so it can be treated by the device behind it. If air can't get through easily, then sound has a tough time getting through as well! In fact, if it isn't breathable at all, then the device acts more like a membrane trap than an absorber, and reflects back high frequencies as well.... That might or might not be a good thing, depending on the device. For a bass trap, that's fine, but for a cloud or first reflection point, that's not so fine!

It's high quality stuff with a pretty tight weave.

Hmmm.... makes me think "not very breathable"....

The corner panels are hung from the soffits at each end of the room, so they are basically about a foot from the actual ceiling, and about two feet from the floor.

Yup, that's what I suspected: they won't be working at maximum effect there. They'll do something, for sure, since they are straddling the wall corners, but they are not as effective as they could be for full floor-to-ceiling traps. In simple theory, a trap up against the middle of a flat wall can be a maximum of 6 dB more effective than the same trap out in the middle of the room, away from the walls. That same trap straddling the corners is a maximum of 12 dB more effective, and a trap straddling a "tri-corner" is a maximum of 18 dB more effective. So you were losing at least 6 dB of effectiveness by not going up to the ceiling and down to the floor. In fact, you could leave out the middle part and still be more effective....

From what I've been reading here and elsewhere, I've learned that this is not adequate, which is why I built the new panels to go from floor to ceiling.

Exactly. For the reasons above, and other reasons too.

I'd also suggest that you hold off on building those new units until we get the analysis done: I'll probably suggest that you build them as "Superchunk" traps, which are the most effective of all, and I'll also probably suggest that you use OC-703 to build them, but let's first see the tests, and then we'll use that to come up with the details of how big to make them, where to put the, and how to build them.

Here's the bad news. Other than the corner traps, there's nothing on the back wall except for shelving from floor to ceiling.

OK, now I'm REALLY suspecting that SBIR is a big issue in your room. That back wall is the most important of all the surfaces in the room, followed by the front wall, ceiling, side walls, and floor, in that order. You'll DEFINITELY need to put about 8 million tons of treatment on the back wall.... but once again, hold off on building anything until I can analyze your REW data.

More disappointing news. I used the mic on the sound level meter, because it's what I had.

That's OK for basic measurements, but you'll need something better for more precision. In the instructions that I linked you to yesterday there's a list of mics that I recommend and some I do NOT recommend for this. A good one will cost you about US$ 100, plus or minus. Take a look at that, and order one of the ones on the list for your future tests. If you want to go to the limit tweaking the room as far as it will possibly go, then you will DEFINITELY need one of those. I can use the REW data from your RS mic for now, but I'd really urge you to get a good mic.

We'll leave any discussion of what I was trying to build aside for now, since I was obviously headed in the wrong direction and was doing it bass-ackwards anyway!

It happens! There's a LOT of misleading, half-baked, and just plain incorrect information on "how to treat your room" all over the internet. Some of it is well-meaning but wrong, just from ignorance. Some of it is tin-foil hat crazy stuff. And some of it is pure snake-oil, trying to convince you that you need to buy a certain product that only they can sell you... You probably came across some of those in your travels, since there's just so much of it! That's probably where you came across panel traps and membrane traps. Those are in the first category: they really do work, when designed and built right, but getting them "right" is a huge big deal that those places don't tell you about. For example, I did use some small membrane traps in this studio: http://www.johnlsayers.com/phpBB2/viewt ... =2&t=20471 but just a couple, not very big, and carefully tuned for specific purposes, and carefully placed at specific locations. I hope I didn't give you the idea that membrane traps are nuts! They are not: they are useful tools in the studio designer's toolbox, but they are not needed anywhere near as much as some places try to tell you, and they are much harder to tune and build than they want to admit. I seldom use them, to be honest. The only reason I used them in that room, is because the owner challenged me to make the room as good as it could possibly be, with response as flat as a ruler, and I really needed them to deal with some minor but stubborn issues that I could not tame any other way. And took days to figure them out and make them work.... I would NOT recommend them for the majority of rooms, and even then only as a last resort. So don't feel bad about yours not working the way you expected! Designing those things is not for the feint of heart. It is possible that your room could benefit from them, but I doubt it, and it's too early to tell for sure anyway.

I guess that depends on your idea of music!

True! There's nothing like a little explosive percussion and fast-burning hot stuff trapped inside a metal case, to provide a powerful musical experience! On the other hand, it's not so good for your ears...

Before we go any further with REW data, what's the story on using the mic in the Radio Shack meter for such tests? If it's utter crap, then I can certainly try to find something better.

I covered that above, but basically it's OK for what you did already, to get a rough idea of the room: however, you'll need something better from here on, if you want to get your place sounding good.

Incidentally, I did another test that gave me similar results. Basically just ran pink noise through speakers and then ran the audio from the RS meter to an input on my DAW, into the recording software as a live input. Then put Voxengo's SPAN plugin across that. Same nasty hole around 130Hz.

Yup, but that's just telling you about frequency repose: how loud each specific frequency is at the mix position. That's just a small fraction of the actual acoustic information that you need to look at to understand what is REALLY happening in the room. Frequency response is useful, to a point, but it tells you nothing about how the room behaves in the time domain, nor reflections, nor phase response, nor a whole bunch of other important things. All of that is hidden in the REW data: you just need to coax it out and look at it in meaningful ways, so it can tell you the full story.

Also, pink noise is fine for setting levels roughly, and for simple analysis, but the log sine sweeps of REW are far more accurate, and give you a mountain of extra data about the room....

Take a look at that thread I linked to above, to see some of the many ways that the data can be analyzed, to reveal what's going on in there. Only when the total acoustic picture is fully clear does it make sense to design and build treatment devices. I often use this analogy: Trying to treat a room without analyzing it and making a plan, is like trying to go on vacation without knowing when or where you are going, or how you are going to get there. Can you imagine someone saying "I'm taking a vacation soon, but I have no idea where I want to go, or if I'll walk, drive, fly, or ride a bike to get there, I don't have a clue if it will be summer or winter there, no idea if it rains or not, nor what language they speak, or what currency they use, I wouldn't have a clue where I'm going to sleep, or what I'll eat, and I'll be blindfolded with ear plugs the whole time!" That's sort of like trying to treat your room without first understanding what is wrong with it: no map, no plan, no way of even knowing what your destination is, or how to get there!

Oh, and to answer your question about what speakers I'm using: Event 20/20 passives powered by a Crown D-75.

Nice! Those should work well in your room. They do gown pretty low, around 35 Hz if I recall correctly, so you are going to need a lot of treatment in there.

- Stuart -

[CurtZHP]

Those tiles are not doing much at all, and the actual acoustic height of your room is really to the underside of the sub-floor above you....

Yes, those tiles were pretty much installed for purely cosmetic reasons. That, and I was building this room alone and didn't fancy trying to hoist full sheets of drywall over my head. And doing mud work on drywall is a b**ch to begin with, never mind ceilings! As far as the subfloor above, I'd have to measure to be sure, but a wild guess would put that about 8-10" above the tiles. I certainly wouldn't be averse to the idea of a fabric ceiling.

GFR = "Gas Flow Resistivity"....

I'll look into this.

Hmmmm..... they key point is that it must be "breathable"

Yes, this material is definitely breathable. I made sure of that before using it. (I did SOMETHING right!)

If you want to go to the limit tweaking the room as far as it will possibly go, then you will DEFINITELY need one of those. I can use the REW data from your RS mic for now, but I'd really urge you to get a good mic.

I guess I'm going shopping. From the other posts I've seen here, you seem to recommend the Presonus PRM1 as an economical option, but I also noticed you recommend the Dayton EMM-6, which I can get for about half the price. What say you?

Stay tuned!

[CurtZHP]

Here are the Dropbox links to requested files....

Baseline REW data:
https://www.dropbox.com/pri/get/Public/ ... Al_TXIpEQg


Basic drawing of studio:
https://www.dropbox.com/pri/get/Public/ ... z7zwmJq6iw

Photo of front of studio:
https://www.dropbox.com/pri/get/Public/ ... tLbS44X8tw

Photo of rear of studio:
https://www.dropbox.com/pri/get/Public/ ... 1_--FZ45Fg

[Soundman2020]

Here are the Dropbox links to requested files....

Great!!! Except that you made the files private, instead of public, so I can't download them...

- Stuart -

[CurtZHP]

Dropbox doesn't appear to allow me to do that. Trying to figure out how, but not having much luck. Seems I have to put in someone's email address to give them full access.

[CurtZHP]

I'm going to see if I can "borrow" the ftp server at work. I can't believe I'm having this much trouble with Dropbox.

In the meantime, I'd like to get the measurement mic. Looking to get the Dayton. Thoughts?

[CurtZHP]

Meanwhile...

To answer your question about the density of the mineral wool, I found 2lb/cubic foot, or about 32kg/cubic meter.
From what else I could find, that puts its GFR a little lower than that for 703 fiberglass.

How am I doing so far? (Keep in mind, I suck at algebra!)

[CurtZHP]

Since Dropbox is being stupid, I created an ftp site to get you the files.

PM'ed you the login details.

[Soundman2020]

Not sure what the issue is with DropBox: It's not hard to make a file or folder public. Google that, or check out their help section.

Anyway, I downloaded the files, and to be honest, your REW data doesn't look too bad, considering the size of the room. It's not fantastic, of course, but not as bad as I expected.

It's a pity that you only have LR data in there, as I suspect that there might be significant difference between the channels, based on looking at the photos you sent. So please do follow the instructions carefully when you do the set of tests with the speakers moved up to the front wall.

Anyway, from what I see in the REW data, I'm still suspecting that your problem is SBIR, or perhaps floor bounce. After you've done the test with the speakers moved, I'd also like to try a "walking mic" test while the speakers are still up front against the wall, as that should prove it on way or the other. I'll send you the instructions for that.

- Stuart -

[CurtZHP]

Great! Thanks for taking a look.

I do have separate left and right REW graphs if you'd like to see those. I can upload them tonight.

Do you still think I need the measurement mic? Sounds like a better option for the "walking mic" test.

[Soundman2020]

I do have separate left and right REW graphs if you'd like to see those. I can upload them tonight.

OK, cool. That would help.

Do you still think I need the measurement mic?

Yep! There's a very steep drop off in your data above 10 kHz, and I'm pretty sure your speakers are not that terrible! The range of the RS meter mic is only up to about 8 kHz, so I have a strong suspicion that's what you are seeing on your graphs, which means the entire top half of the spectrum (1 kHz upwards) is suspect.

When you get the new mic, do one test with the RS mic and another test with the new mic, in the exact same position (the tip of the mic), without changing anything else, so we can compare the two, and see what parts of the original tests are valid, and what parts can be ignored.

In the meantime, I'd like to get the measurement mic. Looking to get the Dayton.

The EMM-6? That should be fine. I don't have one myself, but I've read good reviews, and the specs are fine.


- Stuart -

[CurtZHP]

Going shopping! Stay tuned!