John Sayers' Recording Studio Design Forum Archive

Hi all,

First off, I would like to say this website and forum is fantastic! Such a wealth of knowledge and wisdom, shared through love! This is the sort of thing the internet was made for. Thank you all for contributing.

*****

My friends and I took over a "facility" a number of years ago for the purpose of making our own music, but also doing some recording to earn some cash for development of the place etc. Unfortunately, the folks who built the place didnt really have much of a clue about studio design and so we have been trying to make improvements over time.

The real heart of the problem has been the control room. It was a bit small and pokey, poorly insulated, and had the most baffling wall cutting off part of the room:



Apart from the terrible ergonomics (this sketchup is rather flattering! It was a lot tighter in there), there was particular issue with the bottom end. Everything was too boomy and lacked definition, despite our best efforts to soak up the bass. Im not sure if this was to do with the floor (which is "floating" in that the joists are on rubber) but not particularly well made, thus making it resonant. The speakers in the corner probably didnt help the situation, especially given the small space.

We have since removed the wall, taken out the window to the live room and rebuilt it along the specs of the Recording Manual, rebuilt the door frame (again according to John's specs), moved some electrics, and more besides:



There was a further problem of some practice rooms in an adjacent studio causing a fair bit of bleed through the walls, and so we erected a stud partition next to that (50mm air gap, 50mm double-layed plasterboard sandwich, 50mm stud with insulation) in the hope this will reduce the low freqs (we just hear kick and bass). We aim to clad the facing wall with wooden slats to act as a Helmholtz resonator, but need to calculate what freqs to target and thus the size and layout of the slats (using the Helmholtz calculator ).

However, the one immovable problem is the smallness of the room! The dimensions are:

L - 3.445m
W - 2.8m
H - 2.160m

A bit boxy, but still within the acceptable "Bolt-area" parameters according to this website.

Im not sure how, with the new arrangement, the room will sound, but I believe it is workable. We have plenty of acoustic foam, a large cloud to stick back up, and the tools and materials to build more treatment if required. As a final note, we are going to have wood or laminate on the floor (previously thin, horrible carpet).

We are going to do some REW tests over the weekend and will put them up here, but I would be most grateful if some of you wise folk could offer a little insight into what potential pitfalls we might face given the circumstances of the room. As mentioned, we have a whole wall to fix up with slats to act as a resonator, but of course need to tune it to the relevant freqs.

As an addendum, I checked out the room modes via these websites, but got a range of different results:

Hunecke:


Sengpiel:


ModeCalc:


I appreciate any assistance you can provide.

Thanks

Chris

Welcome back, Chris! It's been almost 9 years since your last post!

Im not sure if this was to do with the floor (which is "floating" in that the joists are on rubber) but not particularly well made, thus making it resonant.

What's under that? Do you have a concrete slab down there some place? If so, then rip out that floor, gain yourselves several inches of headroom, and much better acoustics, due to both the increased room volume, and also due to eliminating the drum that your room is sitting on. Floating a floor is a major deal, very expensive to do right, and a terrible thing when done wrong.

We aim to clad the facing wall with wooden slats to act as a Helmholtz resonator, but need to calculate what freqs to target and thus the size and layout of the slats (using the Helmholtz calculator ).

Which wall is that, in your diagram? Slot walls are useful, but only when tuned correctly. If you are hoping to treat low bass with a shallow slot wall, that ain't gonna work, unfortunately. You need a deep cavity to get low frequencies, and you need lots of volume. Slot walls are good for the mid range, but not so much for the low end.

A bit boxy, but still within the acceptable "Bolt-area" parameters according to this website.

Yes, but just being in the Bolt area does not guarantee that you have good modal distribution. Take a look at this one, and plug in your numbers... http://www.bobgolds.com/Mode/RoomModes.htm

Im not sure how, with the new arrangement, the room will sound, but I believe it is workable.

Specs for critical listening rooms call for a minimum floor area of 20m2, and a minimum volume of about 50m3. You have 9m2 and 20m3. Well less than half in both cases. That doesn't mean the room won't work: Just that it wont be optimal, and it will need a ton of treatment. Small rooms needs lots of treatment. Really small rooms need huge amounts of treatment. You are going to need treatment on about 50% of the entire surface area of the room, and considering that the floor takes up 20% of the entire surface area, that means you need over 65% of the remaining surfaces fully covered. Allowing for doors and windows, you will need something close to 70% of coverage. In other words, almost all of the visible area of he room walls and ceiling will need to be covered.

We have plenty of acoustic foam,

What brand and model? How thick? Total surface area?

a large cloud to stick back up,

Describe how the cloud is constructed (what materials, dimensions, etc.)

we are going to have wood or laminate on the floor (previously thin, horrible carpet).

We are going to do some REW tests over the weekend and will put them up here

Please, pleas, please make sure you follow the full calibration procedure laid out in the REW manual! I often waste so much time analyzing REW data for forum members, only to discover at some point that it is all invalid, because the system was never set up correctly, with proper gain structure and accurate calibration using a decent hand-held sound level meter. Calibrate for 86bdBC with all speakers running. (ie, 80 dBC each if you have just two speakers)

Also, make sure you position the mic and speakers correctly, and do three tests: one with just the Left speaker, one with just the Right speaker, and one with both. Do not adjust anything bet wen tests. Make sure that you are not in the room while the test runs. Do full spectrum sweeps, from 17 Hz to 22 kHz.

but I would be most grateful if some of you wise folk could offer a little insight into what potential pitfalls we might face given the circumstances of the room.

It's a very, very small room, so you will have issues well up into the lower end of the mid range: Your Schroeder frequency will be around 170 Hz. There will be serious bass issues, requiring very large amounts of bass trapping to control somewhat. It will be hard to set up the correct geometry for the speakers and listening position: You will have to limit the size of your desk to something very small and low profile, and there will certianly not be room for anything as large as you show, nor for two chairs. No chance! Use only one chair, and a minuscule desk. There will be SBIR issues, and reflections from the ceiling, which is very low. The amount of treatment you need will make that room rather dead, but you have no choice.

As an addendum, I checked out the room modes via these websites, but got a range of different results:

You can safely ignore any room mode calculator that does not show all three types of modes. Any calculator that only shows axial modes is basically useless. Tangentials and obliques are also important, especially in a room that small where you only have a grand total of five modes below 100 Hz, and only about 60 under 200 Hz! So forget the bottom two in your list. No use at all.

The best two calculators are the one you already found first (Amroc), and this one, by Bob Golds: http://www.bobgolds.com/Mode/RoomModes.htm

Before you do your first REW test, make sure that you have the geometry set up correctly. If your speakers and listengin position are not correctly positioned, then the REW results won't be much use.


- Stuart -

Many thanks Stuart.

Yes it has been a long time! Looking back, my first foray on here was for a different place. Times change and people get wiser! Hopefully I have too!!!

Soundman2020 wrote:What's under that? Do you have a concrete slab down there some place? If so, then rip out that floor, gain yourselves several inches of headroom, and much better acoustics, due to both the increased room volume, and also due to eliminating the drum that your room is sitting on. Floating a floor is a major deal, very expensive to do right, and a terrible thing when done wrong.

It is concrete underneath. I think the walls are built on the floor, so that might be problematic to removing the floor, but ill have a look. Incidentally, I bought a shed load (60msq!) of hardwood (teak?) parquet floor tiles today that we intend to use in the control room. Would this change the matter with regard to keeping the floor as is?

Soundman2020 wrote:Which wall is that, in your diagram?

The "north" wall in the new room image, i.e. behind the mixing desk, will have the slats. We did realise the resonator panels would only affect mid-freqs, and hope the REW test will give us an indication of what to target. When building such slats, should the whole wall be covered, or just part? Perhaps the outer thirds, leaving the middle third for a large screen to be mounted? (we intend to mount a large screen there).

Re: acoustic foam, we have Auralex pads. I think we have this set: D108L-DST™ Roominator Kit, though some have been used up elsewhere, or are too ripped for use (they were all up when we took over the place and had to take them down).

Soundman2020 wrote:Describe how the cloud is constructed (what materials, dimensions, etc.)

The cloud is 1500x1500x170. It has 100mm rockwool, with the remaining depth between that and the ceiling. The surface is covered with carpet and has some auralex pads stuck on too.

Soundman2020 wrote:Please, pleas, please make sure you follow the full calibration procedure laid out in the REW manual!

Will do!

Soundman2020 wrote:It's a very, very small room, so you will have issues well up into the lower end of the mid range: Your Schroeder frequency will be around 170 Hz. There will be serious bass issues, requiring very large amounts of bass trapping to control somewhat. It will be hard to set up the correct geometry for the speakers and listening position: You will have to limit the size of your desk to something very small and low profile, and there will certianly not be room for anything as large as you show, nor for two chairs. No chance! Use only one chair, and a minuscule desk. There will be SBIR issues, and reflections from the ceiling, which is very low. The amount of treatment you need will make that room rather dead, but you have no choice.

Oh dear The desk in the image is kind of indicative of what we have! We have a Soundtracs Megas 40ch, 12 buss desk. Approx. 1m deep, 1.8m long! The room will be used as both for production and mixing, hence the keyboard and rack images. We will need two seats in there too, and were also going to make a small "pew" on the back wall for guests.

I do appreciate this makes it far from ideal and puts the listening spot close to the centre of the room, but it is what it is and we have to make the best of it for now.

With regard to room modes, ive run a spec on Bob Gold's, as you suggested. I will try to attach a screenshot to show.

From what i can tell, there seems to be a problem around the 79-80hz area, and 100hz. This seems to be corroborated by the Amroc calcs too. This suggests we will have to take direct action on these in particular. Hopefully the REW tests will shed some light on the rest.

I hope this sheds a little more light on our situation. Hold tight for the REW report!

Thanks

Chris

It is concrete underneath. I think the walls are built on the floor, so that might be problematic to removing the floor, but ill have a look.

If it is concrete under there, then it's a no brainer! I can't tell you what to do, of course, but I can tell you what I would do if that were my room: If I had a room with a low ceiling and a "floated" floor that isn't floated at all, and instead is basically a drum and a panel trap at the same time, I would not hesitate. You'd find me down on my knees with a sawzall and a hammer, ripping that sucker out, with a huge grin on my face!

Assuming it is up on 2x4's, and there's an couple of layers of OSB, ply, or some such, plus the rubber pucks, you will gain something like 15cm of head room, and you will eliminate the drum head and the panel trap, all in one go! What's not to like about that? Your room height goes from 216 cm to 231 cm in one huge jump, adding 10% to your room volume: It is now 22 m3, instead of 20 m3. Your Schroeder frequency dives from 170 to 155, your Bonello plot is a bit smoother... Lots of good things happening.

Incidentally, I bought a shed load (60msq!) of hardwood (teak?) parquet floor tiles today that we intend to use in the control room. Would this change the matter with regard to keeping the floor as is?

It would not change much. You'll still have a "floated" floor that doesn't float (which makes it worse than a non-floated floor), and you still have a resonant cavity inside that floor, and it is still a drum head. And that flooring would look really cool, laid directly on the concrete slab. It would be a great acoustic surface, as well...

I think the walls are built on the floor,

... and the floor is resting on rubber pucks???? Not only is not floating, but that's also likely not legal. How did that even pass inspection?

But it's not such a big deal as it might seem: It is possible to pull out the floor re-do the bottom part of the wall correctly, so it rests on the slab. You'll need the help of a structural engineer, probably, but its not as hard as it sounds to support each wall with bracing / jacks / whatever while you pull out the bits of incorrect floor from under it, and add the missing parts to the bottom of the wall. It might sound scary, but it is feasible.

The "north" wall in the new room image, i.e. behind the mixing desk, will have the slats

The wall at the top of the page? Where your speakers will be? That's the front wall of the room. It's never a good idea to put tuned devices on the front wall, as they can change the frequency response and phase response of the sound reaching your ears. In any event, that wall will need 10cm of OC-703 directly between the speakers and the hard, solid, rigid actual wall surface, to deal with the SBIR and edge diffraction artifacts coming back at you after bouncing off the actual wall. Those are very low frequency things, and they won't "see" your slats at all: they will behave as though your slats were not even there.

I would suggest not trying to tune the front wall. There won't be any treatable mid-range issues on the front wall in any case, so it would not be very effective. Leave that wall either solid, or make it fully absorbent, depending on which room design concept you are going to follow. I have only ever done slats on the front wall in one control room, and that was for some rather unusual and very specific reasons. Even then, it was not the center section of the front wall, but rather way out, beyond the speakers, and in a much, much bigger room than yours. Here's a link: http://www.johnlsayers.com/phpBB2/viewt ... =2&t=20471

For your room, personally I would leave it as hard, solid, and massive as I possibly could. I would out superchunks in both front corners, panels of 703 between the speakers and the wall, then leave the small section in the middle bare.

and hope the REW test will give us an indication of what to target.

REW will show you the major issues, yes, and after you fix those, then it will show you the less major issues that were hidden and masked by the major ones. And after you fix those, it will show you yet smaller issues... and after you fix those... etc. Ad infinitum. I would not expect that REW will reveal much that can be fixed by slot walls on the front of the room. I would expect REW to show you that you have SBIR issues with the rear wall, and perhaps with the front wall too, but you can't fix SBIR with slot walls. REW will show you some modal issues too, very probably, but you can't fix those with slot walls either. REW will doubtless also highlight reflections that are arriving at the mix position way too early, or way to powerfully, and slot walls can't fix those. It will also show you some frequency response issues that MIGHT be fixable with slot walls, but I don't think you'll have enough volume inside your slot walls to make a huge difference.

Helmholtz resonators need to encompass a fairly large percentage of the entire room volume in order to be effective, and they have to be deep enough that they actually can be resonant at the frequencies that need treating, and also not so deep that the wavelength is shorter than the cavity depth and the slat width. And the device needs to be located at a point in the room where there is a pressure mode of the frequency that needs to be treated. For example, if you are wanting to treat an issue that is related to the ceiling/floor axis, then putting your devices on the side walls would do nothing, since it would never "see" that particular problem.

Also, Helmholtz resonators are notoriously difficult to tune: Text book equations are one thing (and many of those equations are wrong anyway, perpetuating the same old misprinted "+" sign instead of "x" sign, from decades ago!), but the real world is another thing. Materials are not perfect, workmanship is not perfect, the actual air temperature might be different from what you used in your prediction equation, there might be measurement in inaccuracies... trying to hit a specific frequency is nowhere near as easy as it sounds from reading a text book. If you do use slot walls, then tune them broad-band, low Q, across the entire mid-range, and be content with whatever small effect they give you. It really isn't worth the effort of trying to tune them perfectly to hit exact frequencies.

Considering all of the above, I would suggest not bothering with them at all. The room is tiny, there's not enough available surface areas (maybe 25%) where you could put them, and I don't see them being very useful. Stick with absorption. That's abotu all you can do in a small room.

Perhaps the outer thirds, leaving the middle third for a large screen to be mounted? (we intend to mount a large screen there).

You can, indeed, hang a large screen in the middle of the front wall, between the speakers. As long as it is not larger than the distance between the speakers (that will be determined when you do the calculations for the correct room geometry), then that's fine. Hanging a screen would effect the slot wall behind it for some frequencies, but not for others....

Re: acoustic foam, we have Auralex pads

That's fine. At least you have proper foam, from a reputable manufacturer.

or are too ripped for use (they were all up when we took over the place and had to take them down).

Don't throw away those damaged scraps! You can still use them. Keep all the bits and pieces of your left-over foam and insulation, and dump them inside your superchunks. They will work quite well in there.

The cloud is 1500x1500x170.

That's rather small.

It has 100mm rockwool, with the remaining depth between that and the ceiling.

No hard back?

You would get a better effect by making it thinner (120mm, for example, instead of 170), then putting a thick, rigid, massive, heavy hard back on top of it (eg. 19mm MDF), and angling it slightly (lower over the speakers, higher over the desk), then stuffing some more of your foam craps on top. That would be hard to do with a low ceiling, but much easier to do once you gain back the lost height that the "floated" floor is robbing you of....

The surface is covered with carpet and has some auralex pads stuck on too.

The carpet isn't doing much, except making the room sound dull: Ditch that, and just cover the cloud with thin plastic (such as painter's drop cloth) to keep the mineral wool fibers from filtering down all over you and your gear, as time passes, then cover that with nice looking fabric of your choice. Take a look at the clouds in the link I gave you above.

The desk in the image is kind of indicative of what we have! We have a Soundtracs Megas 40ch, 12 buss desk. Approx. 1m deep, 1.8m long!

You are not going to like what I am about to say, and even though that's a great console, it is just way too big for that room. Putting a large consoles in a small room is just begging for problems. I can start listing some of them, if you want . . .

1) Mid-range mud. The console surface will be very close to both your speakers and your ears. That's a sure-fire, guaranteed recipe for major reflections, comb-filtering, and phasing artifacts. The sound coming from the speakers will be greatly affected by the surface itself (lows and mid-lows) as well as by the control knobs and buttons (high-mids). The mid-range frequency response curve will resemble the mountains of some distant alien planet, and there is absolutely nothing you can do about that: no treatment is possible, because the issue is not related to the room at all. The mangled sound will reach your ears long before it reaches the room walls, where it could be treated, but that would be far too late....

2) Low-end "flummph". That's a very technical term that I just invented to describe the way the low end falls apart when there are large objects destroying these nice long waves that would otherwise have got through cleanly. In other words, your low end will not be tight. It will be muffled, unclear, but also over-represented. Your kick and toms will sound like wet cardboard boxes: No clean, hard "thump!" Just a waffly baggy "flumph".

3) Phase cancellation. You will get multiple reflected versions of the same sound hitting your ears with slightly different timing, and therefor slight delays. Those will create phase cancellations for some frequencies, but not for others, but that will change as you move your head to different positions relative to the console....

4) Early reflections: A cardinal rule fo control room design is that no reflections should reach the engineer's ears until at least 20ms after the direct sound has arrived, and even then should be 20 dB lower in intensity. With a massive hard flat surface just a few inches from the speakers, and a few inches form the engineer's ears, that is impossible. Those reflections will be arriving with just a couple of ms, or less, and will be practically the same intensity, maybe a dB or two lower, if that. The result is a psycho-acoustic mess! That destroys all of the clues that the highly precise design of your ears and your brain use to determine frequency response and directionality. Your brain will interpret the resulting grunge as being a different tone that came from a different direction. In other words, you lose your sense of directionality and of tonality. In simple terms, blurred sound-stage, unclear stereo image, messed up sense of musical notes.

Want me to go on?

I would seriously consider getting rid of that monster, even though it is a great console, and replacing it with a DAW, or a minimalist console, and a minimalist desk. That room is just too small to be able to successfully fit such a big console in. Either get a much bigger room, or get a much smaller console.

The room will be used as both for production and mixing, hence the keyboard and rack images.

Ummmm. Nope! Just nope. It's a tiny room! You cannot have big things in it! Each of those will take up a large percentage of the entire room volume, messing up a large section of the room response. You MIGHT be able to fit a small keyboard on a slide-out tray under the desk, but that's about it.

We will need two seats in there too,

Nope. Not in that room. If you do, nobody will be in the sweet spot. It is a tiny room, the sweet spot will bin minuscule, and there is no way in hell that two heads could be in it. It's going to be tight to even get ONE head in it! You will most certainly hear large changes in frequency response as you move your head around, even like it is, even if you set it all up and treat it perfectly. For critical listening, the engineer will need to keep his head fairly still, since moving it will give him a different "image". Yet another reason for not having a console where the engineer needs to move side to side to reach distant faders.....

and were also going to make a small "pew" on the back wall for guests.

The back wall will have about 15 to 20cm of absorption on it, to deal with the bass issues, modal problems, and help some with SBIR. You can put a "pew" in front of that if you want, but nobody on that pew will be able to do any critical listening. They will be right at the point where all the modal issues and SBIR issues are at their peak.

and puts the listening spot close to the centre of the room,

That's the worst possible location in the room. All the first-order modal peaks and second-order modal nulls are at their maximum intensities there, in all three axes...

we have to make the best of it for now.

Unfortunately, from what you describe, that "best of it" is actually somewhere between "worst possible" and #even worse than that"!

From what i can tell, there seems to be a problem around the 79-80hz area, and 100hz. This seems to be corroborated by the Amroc calcs too. This suggests we will have to take direct action on these in particular. Hopefully the REW tests will shed some light on the rest.

Here's what your low-end response will look like, at the mix position in the center of the room, assuming optimal geometry, basic treatment, no console, and no racks:
Not a pretty picture! I would not like to try mixing in that room.... That grapsh is showing only the predicted frequency response from the modal issues. It does not show rear-wall SBIR issues. There will be an SBIR dip at around 100 Hz, which is a real fun situation since it is combined with both an axial mode and also a tangential mode that are already reinforcing each other! Which of course implies that there will also be a corresponding 6 dB peak at around 200 Hz, then a dip at 300 Hz, peak at 400, dip at 500... all the way up the scale... Can you say "major comb filter"?

Once again, I can't tell you what to do, but know what I know about acoustics, if that were my room I would absolutely not put any furniture in it, except for a very small desk with my DAW keyboard, mouse and monitor on it, perhaps with very small, low.profile "wings" to either side with my key pre-amps, interface, and a couple of other critical rack-able items in them (Sterling desk style...), and maybe a very small single-seat sofa towards the rear, as far away from the wall as I can get it while still leaving space for my own chair. And that's it! Nothing more. I would have massive bass trapping across the entire rear wall, with a couple of very broad slats across it at strategic locations, large superchunks in each vertical corner and the wall/ceiling corner, superchunks in both front vertical corners and both front horizontal corners, thick absorption in the first reflection points on the side walls, 10cm OC-703 panels between the speakers and the front wall, a large, hard-backed, angled cloud with extra insulation on top, and another panel of 703 on the rest of the ceiling, behind the cloud. I might add a couple of other items, if REW showed me something really bad, but not much more than that. That would give me border-line usable acoustics, but if I tried to put anything more in the room, that would make things worse, so it would just be me and my DAW, and acceptable acoustics.

Hold tight for the REW report!

Looking forward to it! Let's see how closely reality matches prediction! I'm assuming that you already have your speakers set up correctly, with the optimal geometry for that room?

Please also post a couple of pictures of how the room looks at the time you do the REW tests.


- Stuart -

Hi Stuart,

Apologies for the delay in getting back to this.

After some thought and discussion, we have reluctantly decided not to pull up the floating floor

Our reasoning for this is that we need to get some money back into the place ASAP (and also start making our own beats! Its been over a year now!), and tearing out the floor was going to be a serious undertaking and delay us even further. I am kicking myself for not having thought of this before when we first started, but such is life. We will have to deal with it and make the best of what isnt the best situation.

Anyhow, we've ended up laying a laminate down (taking another 10mm off the room height!) but ive taken some REW readings since.

I think i calibrated the soundcard levels correctly (Alesis ION 2), found a cal file for my Behringer ECM8000, and did the SPL reading too. The level on the SPL seemed a little hot (+- 100dB?), but i still had headroom according to REW and nothing seemed to clip.

Speakers were set up 485mm from each side wall and 370mm from the back. Distance between tweeters was 1440mm and about the same distance to the mic (pointing upwards).

A couple of in progress shots: Hopefully you can see the unfinished wall in the background, waiting for either resonator slots or more plasterboard.

The soundcard (which isnt great tbh) behaved a bit better in ASIO, although had the undesirable side-affect of only allowing running tests in either left or right, rather than stereo. As such, results are for left and right speakers:

Left: Right: From what I can tell, the room aint great! But in some respects that was to be expected given A) the floor, B) the shoddy build generally, C) there being no treatment at all. There are some nasty looking peaks on the waterfall for both sides around 40hz, 75hz, and 135hz. The peak on the left around 315hz could be due to its close proximity to the window. Above that doesnt look too bad, though its the first time ive ever really had a look at these things.

Overall, it looks pretty clear the room is not flat, especially below around 200hz, which is kind of what I/we expected. I cant really understand much from the RT60 and SPL graphs, although the latter looks rather alarming!

Hopefully Ive run the tests correctly (perhaps too hot?) but if you are able to offer any insight into what these are saying I would be most grateful.

Many thanks

Chris

The level on the SPL seemed a little hot (+- 100dB?),

That's rather surprising with those speakers in that room! I would not expect they'd be able to accomplish that. And 100 dBC in that room would have been deafening! Are you CERTAIN that you had your hand-held sound level meter set up correctly ("C" weighting and "Slow" response)? Is it a decent one? Some of the cheap Chinese ones are basically junk, hardly better than toys... What brand and model of meter do you have? Any chance you can beg, borrow, or buy another one to double-check?

but i still had headroom according to REW and nothing seemed to clip.

Also rather surprising, given the speakers, room, mic and circumstances... I'd suspect that your hand-held meter is flawed, damaged, not calibrated, or just plain lousy. Better repeat the calibration with a decent one. It's not good to rely on REW data if you are not confident that REW has been calibrated correctly with an accurate meter.

I'm also wondering why you ran the tests at 100 dB when I said this: "Please, pleas, please make sure you follow the full calibration procedure laid out in the REW manual! I often waste so much time analyzing REW data for forum members, only to discover at some point that it is all invalid, because the system was never set up correctly, with proper gain structure and accurate calibration using a decent hand-held sound level meter. Calibrate for 86bdBC with all speakers running. (ie, 80 dBC each if you have just two speakers)

100 dB ≠ 86 dB

After some thought and discussion, we have reluctantly decided not to pull up the floating floor ... tearing out the floor was going to be a serious undertaking and delay us even further.

To be brutally honest, you probably could have had it out and disposed of in less time than it took for the "thought and discussion"!!!.. The entire floor area is barely 9 m2, for crying out loud! One man with a half-decent set of tools could have had that out in half a day. I speak from experience: Not too long ago I was part of a six-man team that ripped out 200 m2 of wooden flooring in under four hours.... All we used were crowbars and our bare hands.... Draw your own conclusions.... So your walls are resting on the edges of the floor? Big deal. Any half-decent framing carpenter could have fixed that in the remaining few hours of the day...

(and also start making our own beats! Its been over a year now!),

... and a couple more days was too much? Sorry for being a bit sarcastic, but that type of decision just doesn't make a lot of sense. If you've been out of production for, say, 400 days, I just don't see the logic in rejecting the biggest single thing you could do to improve the room drastically because it might take another day or two...

Speakers were set up 485mm from each side wall and 370mm from the back.

Ummmm.... Why? It's a small room. You should have it set up as such. With that setup, you'll have a large SBIR null at around 220 Hz.... Whoops! Yup, there it is, in your graphs, clear as daylight... Massive dip at around 220, in all of your graphs.

370mm from the back.

... I think you mean "front" ... The speakers should be up against the front wall: the one you are facing while you sit and mix. If they were up against the "back" wall, they would be behind you...

Theoretically optimum geometry for that room is to have your speakers tight up against the front wall (rear corner of speaker almost touching the insulation), and 79 cm from the side wall, thus setting them 122 cm apart. Your mix position should be 131 cm from the front wall (the location of your ears), and the speakers should be angled inwards to aim at a spot that is 165 cm from the front wall (in other words, about 35 cm behind your head).

Distance between tweeters was 1440mm and about the same distance to the mic .

Forget the mythical "equilateral triangle". It only works for people who have their ears located in their eye-sockets. Think about it.... What matters is symmetry, not exact 30° angles. Why is it necessary to have exactly the same distance between the speakers, as there is from speaker to ear? Answer: Nobody has a clue!

... mic. (pointing upwards)

Right, but try to get it at an angle of about 60° upwards, not 90°. 90° tends to over-emphasize the ceiling, and straight forward tends to over-emphasize the speakers. 60° is a good compromise.

A couple of in progress shots:

How high do you have your speakers? That looks way too high, to me. The speaker axis should be at 120cm above the floor, or maybe a little higher. Yours looks more like 145 or so... Get them down to the correct height.

I realize that this is just a very temporary set-up, but I did want to point out that those speaker stands are no use at all. Speaker stands need to be massively heavy, very rigid... those are just thin wobbly light-weight gig stands. They are fine for these initial baseline measurements, but no use for later measurements, where you'll need precision.

Hopefully you can see the unfinished wall in the background, waiting for either resonator slots or more plasterboard.

To me, it looks pretty good just as it is! Put some thin plastic over the insulation, put the fabric of your choice over that (with a layer of cheap black fabric under it), build superchunks in the corners, and you are done!

For that size room, I would normally and unequivocally recommend soffit-mounting the speakers, but since you are in a major rush to get the place finished, and "making it the best it possibly can be" is not high on your list of priorities, I'd say to just stack up some concrete blocks, wrap them in fabric, and use those for speaker stands, cheap, fast, simple...

behaved a bit better in ASIO, although had the undesirable side-affect of only allowing running tests in either left or right, rather than stereo. As such, results are for left and right speakers

That's weird! So if your system only works one channel at a time, then how on earth are you going to make music in there???? You better get that fixed, fast. Without seeing how both speakers together interact with the room, it will be pretty hard to plan the treatment, and then figure out if the treatment is working or not...

There are some nasty looking peaks on the waterfall for both sides around 40hz, 75hz, and 135hz.

To be honest, I'd be far more concerned about the dips at 60, 75, 105, and between 220 and 280. Peaks are relatively easy to deal with, but there's not a lot you can do about dips... especially if they are caused by SBIR.

The peak on the left around 315hz could be due to its close proximity to the window.

I don't see any major peak on the left channel at 315...

Overall, it looks pretty clear the room is not flat, especially below around 200hz, which is kind of what I/we expected.

Right. That's normal for a small room with no treatment.

Hopefully Ive run the tests correctly (perhaps too hot?)

Very much so! I did say that you needed to run them at 80 dBC per side, which is automatically 86 dBC with both running.

However, I strongly suspect that you did not actually run the tests at 100 dBC. I doubt that your speakers would be able to do that. I suspect that you did not calibrate REW correctly, or that your hand-held sound level meter is very inaccurate. Get a better meter, re-calibrate, then do another set of tests.

but if you are able to offer any insight into what these are saying I would be most grateful.

I would need the actual MDAT file, so I can analyze it in detail. The scale you used for your graphs is way to small, and hides all of the important detail. You didn't show the impulse response curve either, nor the spectrogram. Please upload the MDAT file to some place like DropBox, then post the link here so I can download it.


- Stuart -

Thats quite a condemnation! But i guess it is to be expected. We arent doing a great job, I apologise and appreciate that. It must be frustrating for you to give your time up for people let us who dont heed your advice. However, it is greatly appreciated. Thank you. Sometimes a kick up the backside is better than a gentle word

Nonetheless, ill carry out some new measurements again given your advice. I did think that 100dB reading was a bit suspect, and indeed i only had a pair of unplugged headphones keeping me safe, which certainly wouldnt have been enough if it was at that volume! SPL meter was set at C weighting and slow, and the measurement was taken adjacent to the mic.

The SPL meter we have is one of these. It was cheap but and guessing not up to scratch, although i do remember seeing it mentioned as up to the job here or perhaps elsewhere (SoundonSound?). Ill check it again though, as we i did some tests before the floor went in and they were around the 75-80dB mark.

With regard to the speaker stands, they are indeed temporary. We have some heavy duty steel beasts that i thought were too unwieldy to take in for the sake of a test, especially given as they will need to be taken out soon after so we can carry on with finishing the room (they take 2 people to carry!). The speakers are Event TR6 monitors btw. I positioned them relative to where i thought we would have them when working, i.e. on both horizontal and vertical orientations.

It is interesting you mention soffitt mounting. Is this a genuine possibility given our circumstances? I thought this was only possible in larger rooms and with larger speakers. It certainly is an option still as we havent done anything with the FRONT wall ( ill get my orientation right from now on!). Are there any considerations/rules/principles we would need to follow regarding height, distance from side walls etc?

The soundcard used was my cheap-n-cheerful Alesis, not the one we will actually use in there! (Motu 24io). Not sure why it wasnt allowing stereo output in ASIO but ill look into that.

Thanks for you help again, Stuart. I HOPE we will arrive at a happy place soon!

Chris

Thats quite a condemnation! But i guess it is to be expected.

Yeah, sometimes I do go a bit overboard, and get possibly too much "in your face". But I do believe in just telling it like it is, calling it the way I see it, and hopefully getting your attention! I'm not much into PC, trying to soften things, pussy-foot around the issues, etc. Most people seem to prefer the blunt approach. Hopefully I wasn't too harsh!

It must be frustrating for you to give your time up for people let us who dont heed your advice. However, it is greatly appreciated. Thank you. Sometimes a kick up the backside is better than a gentle word

True that! Yes, it is frustrating, when I see what could be done to make a room great, knowing that it isn't terribly hard to do.... Sigh! As you noticed, I don't do "gentle words" very well...

The SPL meter we have is one of these. It was cheap but and guessing not up to scratch,

Right! I'd agree with "not up to scratch"! For 18 quid, you won't get much of a meter. A decent one will cost you around US$ 100, or a bit less. Look for a Galaxy, or Extech, or something reputable like that. Try ebay or Amazon.com. I'm also going to PM you the instructions that I give to my paying customers on how to do the calibration and measurements correctly.

i did some tests before the floor went in and they were around the 75-80dB mark.

To get an accurate "picture" of how the studio is behaving, you need to do the tests at the "standard" level of 86 dBC with all speakers running. If you do it at a lower level, it's likely that you won't be triggering all the room modes that will be active when you are mixing normally. And if you do it a level that is too high, you will probably be seeing stuff that won't there normally, as well as potentially leaving insufficient headroom in the measurement. 86 dBC is a very good reference level, which is why organizations such as the AES, ITU, EBU, Dolby, etc. all use it as their calibration level. To get 86 dBC with both speakers running, then you need to calibrate each speaker individually to 80 dBC, without the other speaker on. That's assuming plain stereo with just two speakers. If you also have a sub, or 5.1, or 7.1, etc., then the level of each individual speaker is set lower (there's an equation for that), such that the combined level of all spakers playing the same coherent sound, is 86 dBC. That's the reason why I use 86, rather than the 75 stated in the REW manual. I'm not sure why they have 75 in there, since 86 is the standard (or as close as you can get to a "standard" in this industry!). Maybe it has something to do with the original purpose of REW being for home theater enthusiast, not studios. But 86 is the level you want at the listening position with all speakers on.

We have some heavy duty steel beasts ... they take 2 people to carry!

Sounds about right, though maybe a little on the light side!! (Just kidding...)

The speakers are Event TR6

I thought they might be. That's why I made the comment about the 100 dBC SPL level... those are nice speakers, but they can only put out 106 dBC according to the manual. That's absolute maximum, pushed to the limit, on a good day, going downhill, with the wind behind them.... The real limit is probably more like 100-102, so if you had a level of 100 in the room, you were pushing them pretty hard.

I positioned them relative to where i thought we would have them when working, i.e. on both horizontal and vertical orientations.

There's two ways of setting up the room geometry: optimally, using acoustic theory... or by guessing... Not meaning to be harsh, but as with most things technical, there's usually one right way and about seven hundred and fourteen wrong ways, give or take a few dozen (not counting the Murphy's law variations... )

The measurements I gave you are for the optimum geometry for a room of that size. If you set up the system like that, and set your chair down exactly where I mentioned, with your ears centered on that point, then listen to some of your favorite music (stuff you know really well) at a level of about 80 dBC, I think you'll find that it sounds pretty darn good, even in the empty, untreated room. It puts your speakers and head in the best relationship, psycho-acosutically, which gives your brain the best chance of interpreting the frequencies, phases, levels and directionality correctly. You should hear things clearly, with decent bass, good stereo imaging, clean, accurate sound stage, reasonably smooth response, etc. If you then re-arrange the room the way you had it and listen again, I reckon you'll notice the difference!

Having said that, there's sometimes a difference between theoretically optimum geometry, and absolute best geometry, but it wont be far off. So it can pay to spend some time making small adjustments to that basic layout: Move the speakers slightly closer together or further apart (just a cm or so at a time) then re-aim them for that point behind your head. Try moving your chair a couple of CM back or forward, once again re-aiming the speakers. You might find a better spot, but it will be pretty close to what I gave you.

It is interesting you mention soffitt mounting. Is this a genuine possibility given our circumstances?

Absolutely!

I thought this was only possible in larger rooms and with larger speakers.

Actually, from my point of view, it's the other way around! Smaller rooms benefit from soffits more than larger rooms do. If the room is very large, you can get pretty decent results without any soffits, with the speakers on stands far away from the walls. But the room needs to be large enough that the speakers can be about 3m away from the front wall and side walls, and still have the mix position in the correct location, and also have a client couch at the back in another good location... If the room is about 20m long (or more) and about 12m wide (or more) then you can do that.... But not many rooms are that big!

Personally, I feel that the smaller the room, the better it can benefit from soffit mounting. Yes, soffits do take a lot of space, but the benefits still outweigh the downside.

ill get my orientation right from now on!

Are there any considerations/rules/principles we would need to follow regarding height, distance from side walls etc?

Yes! Lot's of them!

The soundcard used was my cheap-n-cheerful Alesis, not the one we will actually use in there! (Motu 24io).

Any chance you can try it with the Motu next time?

Let me know if you get the PM...

- Stuart -

Soundman2020 wrote: Hopefully I wasn't too harsh!

Not at all, chap!

In fact, you'll be pleased to know, it gave us the required slap in the face and a couple of mates spent last night tearing up the floor! There was a load of old conveyor belt rubber (from those big old road breaking machines, i think!) that the joists were on top of, but i think they took that out too.

After a clear up, we'll lay the flooring back down. Figuring out what to do with the doors and lining will take a little bit of effort, but we are gonna build some soffitt mounting and super chunks, as recommended (my mate wanted to do that - more because he's a chippy than for acoustic reasons! - but we held back due to not thinking it possible, as mentioned previously).

However, as you can see in one of the photos, there isnt a huge amount of room between the front wall and edge of the window frame. As such, any superchunk/soffitt frame will be at quite a shallow angle (ill have to get the measurements). Would this be problematic for soffitt mounting? Is there a specific angle or formula for calculating the optimum angle for the speakers/soffitts?

Many thanks again for all your assistance (and persistence!?), and for your REW setup notes.

Chris

it gave us the required slap in the face and a couple of mates spent last night tearing up the floor!

Smart move! And I'll bet it didn't take more than a few hours...

Did you find out if the walls were sitting on that "floated" floor, or not? If "not", and they are really sitting down on the concrete slab, then you are fine. If they are, then did you manage to deal with that?

There was a load of old conveyor belt rubber (from those big old road breaking machines, i think!) that the joists were on top of

Hoo boy! That was doing nothing useful at all, and the floor most certainly was NOT floating on that! It was fully coupled to the slab, and creating a wonderful resonant box...

but i think they took that out too.

I certainly hope so!!!! You can't lay laminate flooring over "old conveyor belt rubber from a road-breaking machine"... It needs to be on the slab itself, with suitable underlay, of course. (underlay that is approved and recommended by the manufacturer of your flooring).

Figuring out what to do with the doors and lining

"Lining"? What lining?

we are gonna build some soffitt mounting and super chunks,

Another smart move!

as recommended (my mate wanted to do that - more because he's a chippy than for acoustic reasons! - but we held back due to not thinking it possible, as mentioned previously

Your mate sounds like a smart fellow! Let's hope he's up to the task...

However, as you can see in one of the photos, there isnt a huge amount of room between the front wall and edge of the window frame.

Bad news time again, I'm afraid. And you ain't gonna like this one: That window is going to end up partially covered by the soffit. Not all of it, but part of it will have soffit blocking it.

As such, any superchunk/soffitt frame will be at quite a shallow angle

Soffits cannot go at a shallow angle! They have to go at the angle that is needed in order to get the room geometry correct.

Is there a specific angle or formula for calculating the optimum angle for the speakers/soffitts?

Yes, but it's complicated.... so I did it for you....

Many thanks again for all your assistance (and persistence!?), and for your REW setup notes.

- Stuart -

Hello there Stuart! I'm Donk, I'm the chippy mate with vested interests in this project. my mothers maiden name was "Smart " im probably only half smart, so we will see.

Your mate sounds like a smart fellow! Let's hope he's up to the task..

After Removing the floor cavity, we actually found that the walls are sitting on the rubber, as was the floor joists. but the ceiling joists of the floating room have directly been fixed to the external floor joists above, wich in turn are bolted to the external concrete block wall. before chris gets too obsessed with the internal acoustic treatment, i think its a wise move to de-couple, this would be fairly easy. but will mean after the plasterboards on the ceiling are down and the 2x2 that is directly attached is removed, i will have to suspend another ceiling below effectively decoupling. after taking the room apart this became apparent. Talk about an acoustic can of worms

the whole room is a god damn resonator

Hello there Stuart! I'm Donk, I'm the chippy mate with vested interests in this project.

Hey there Donk, and Welcome to the forum too! Please fill in your profile details... Forum rules...

After Removing the floor cavity, we actually found that the walls are sitting on the rubber, as was the floor joists

That's good news. The rubber isn't doing anything useful, of course, but at least the walls were not sitting on the "floating" floor. I'd suggest cutting away the rubber along the edge of the sole plates, so you can remove it from the rest of the room and just leave that strip under the wall. Also, I'm assuming that the wall sole plates are properly bolted through the rubber into the concrete? If not, you'll need to do that too. You can use expanding anchor bolts for that.

but the ceiling joists of the floating room have directly been fixed to the external floor joists above,

Hooooo boy! And again.... HoooOOOOO BOOOy! So nothing about the room was floated at all? It was all mechanically connected o everything else! No isolation going on there....

before chris gets too obsessed with the internal acoustic treatment, i think its a wise move to de-couple,

Yes, absolutely. Put Chris on the back burner for a while, and turn him down to simmer.

The room acoustic is linked to the isolation plan to a certain extent, so it all needs to be done together, but the actual treatment can wait.

this would be fairly easy. but will mean after the plasterboards on the ceiling are down and the 2x2 that is directly attached is removed, i will have to suspend another ceiling below

Yes, but no. And also maybe, possible, definitely, undoubtedly, and never!

What I mean by that is that you have alternatives, and may be able to improve things more than you expected.

First question: How is that existing room ceiling built? I mean, after you take of the useless "fake" ceiling and the 2x2 furring, what is next? Is that more plasterboard on other joists? If so, what is above that? Please explain in as much detail as you can...

after taking the room apart this became apparent.

That's quite common when renovating! There are always "surprises" to be found that you weren't expecting, and normally they are not pleasant surprises...

Talk about an acoustic can of worms

Yup. You guys are doubtless swearing and pulling your hair out round about now, but it's actually good that you are finding this stuff now. If you find it, you can fix it, and the final studio will be a much better place for it. In the end, you'll be glad that you did. Fortunately, you are not in quite as bad a state as this poor guy was: http://www.johnlsayers.com/phpBB2/viewt ... =2&t=17363 He had an even sadder time, but if you read that through to the end, you'll see how happy he was with the final result, after he tore his entire room apart and re-did it properly. You guys are not quite that bad, if that's any consolation for you... But the good thing after you read through that thread, is that you'll see there really is light at the end of the tunnel, and it isn't an oncoming freight train!

the whole room is a god damn resonator

Yup! But it can be fixed, because you found it.

It might be a good idea to peek a little more into what you have there, to make sure there are no more "surprises" . . .

One issue that I didn't see addressed yet: How are you planning to handle HVAC? What is there already for HVAC, and what additional measures are you taking? HVAC is a major part of any studio, but is often neglected until it is too late. You guys are at the perfect stage now where you ca do it right.


- Stuart -

Hi Stuart. when you say

I'm assuming that the wall sole plates are properly bolted through the rubber into the concrete? If not, you'll need to do that too

I was of understanding that i was supposed to be eliminating mechanical / structural ties between the two independent structures. isolating them? this is why im asking with regards to the fixings on the sole plates.

Cheers donk!!

I was of understanding that i was supposed to be eliminating mechanical / structural ties between the two independent structures. isolating them? this is why im asking with regards to the fixings on the sole plates.

That's the thing: It must be anchored to the floor. No two ways about it. If not, then it will MOVE! There's a lot of air pressure changes and vibrations going on with studios, as the doors open and close, music is played, etc. Also, structures "settle" over time, entire houses do too. You don't have much of a problem with earthquakes, like I do where I live, but it's still a possibility. You don't want your walls sliding around all over the place as "all of the above" stuff happens...

So they have to be anchored. It's probably not even legal to have walls that just sit on top of the floor without being anchored. I'm surprised that it passed the framing inspection!

So your question is doubtless: "How in hell can I anchor something and also have it decoupled at the same time?". Simple: Like this:
With an isolated bolt or screw! It still holds things firmly in place, but it is resilient. That's the correct way to float a wall.... but there's a bit of math involved in doing that correctly! You have to make sure that there is just the right amount of mass in the wall to make that "float". If the wall is too heavy, then the isolation pad underneath is over-loaded, squashed, and it does not float: it "flanks". By the same token if there is not enough weight on the wall then the pad is "under-loaded", not deflected at all, and does not float: it flanks in that case too. You have to get the load just right, properly calculated, such that the pad is deflected (compressed) the right amount to make it float. There's a fairly broad range of acceptable loads, fortunately, but you still have to do the math to make sure.

Think of it this way: It works exactly like a shock absorber in a car. There has to be enough weight on the spring to get it roughly in the middle of it's travel, or if not the car wont "float". There's a reason it is called "suspension": it "suspends" the car on the spring! If you put too much weight in the car (the proverbial four elephants, for example!), you mash the spring all the way down to the limit, the suspension stops suspending, and the ride is terrible. On the other hand, if you put the spring form a Mack truck in a Mini Minor, obviously you'll never, ever compress it at all, so it won't float. And if you put the Mini Minor spring in a Mack truck, it will be pancaked down to the ground, and won't float.

OK, so those are extremes, but you get the picture. Any suspension system must have the correct spring for the weight on top of it, or it will not "suspend".

That's the situation right now with your wall: the "old conveyor belt rubber from a road making machine" has totally unknown properties, but I can pretty much guarantee you that the weight of your wall is nowhere near enough to make ti float. The wall would need to deflect it (compress it) very noticeably, pushing down the surface of the rubber by a large amount, in order to float, and I'm betting that is not the case. I'm betting that it barely makes any impression at all, perhaps a slight indentation... So your wall is not floating. With or without anchor bolts, it is not going to float. I'd put good money on saying that the guy who built that monstrosity did no calculations at all, and would not be able to tell you what the Compressive Modulus was for that setup he created, nor the shape correction factor he should have applied, nor the static deflection, or shear modulus, not any other of the factors that go into the calculation. I'll bet he couldn't even tell you the Durometer rating for that rubber if his life depended on it, no which Shore scale it was measured on! If you ask him about those and just get a blank stare in response, then I guarantee that your wall is not floating, never did, and never will

And that actually doesn't matter! You have a perfectly good concrete slab on grade under you, and that's all you need. That's about the very best floor you can get for a studio. It is very solid, rigid, massive and thick, with a smooth acoustically reflective surface, and it is resting on the ground, so it has the entire planet to damp it! It's hard to beat that....

You are not looking for extreme levels of isolation, so you don't need to float your floor or walls anyway. If you told me that you have Led Zepplin rehearsing in one room, Grateful Dead in the other, and there's a neo-natal ward for new born babies on the other side of your wall that you cannot disturb, then you might be able to convince me that you need to float your floor and walls. Otherwise, you don't. In all the studios I have built over the years, I have never needed to float a floor. In March I'll be starting on a project where I will need to do that, for the first time, abut only because the studio is located two blocks away from a major sports stadium that holds massive rock concerts regularly, in addition to large sports events with tens of thousands of people yelling and screaming, and a PA system that will knock your socks off: the ground vibrates whenever there's a big match on, and the level outside is unbearable when there's a concert or show. There's also a couple of major roads running really close by (obviously: stadium traffic), and a subway running down under it. We will be floating that one, but only because of the extreme conditions. You have nothing like that, so you do not need to float.

Just go ahead and anchor your walls the way they should be: they are not floating anyway, so putting a few bolts through them won't make much difference, and you don't nee them to be floated to start with. There's no point in trying to float that at this stage.


- Stuart -

Soundman2020 wrote:I'd put good money on saying that the guy who built that monstrosity did no calculations at all

That would be quite an understatement. I think id get kicked off the board if I told you what he/they were like, and I dont mean from my opinion of him/them. Lets just say building a good studio facility wasnt at the top of their agenda.