The Cottage Studio Build

Whoa! Thank you so much Stuart! I will try to answer each question best I can below..

Soundman2020 wrote:

Here is the MDAT File...

Assuming you did the calibration correctly, then the tests were done at a level that is way too low. The graphs are showing an average level of around 65 dB: It should be about a hundred times louder: 85 dB. At such a low level, there's not enough dynamic range left for REW to be able to give accurate results. So I'd suggest repeating the calibration process for setting the SPL level (using your own hand-held meter), then do the tests again with levels of around 80 dB for the individual speakers, which should give you around 86 dB for both speakers, as seen on both the REW meter and also your own hand-held meter ("C" weighting, Slow response").

My spl meter no longer works, So I went to radio shack and bought a new one and it was defective. I am waiting for a new one to come via mail order. The measurements were not calibrated, I just set the test levels to a level that would be equivalent of extremely loud playback. probably close to 95-100db at mix position.

possibly turning the speakers on their sides

I'm trying to download the manual for those things, to find out how they are supposed to behave, but all I'm getting is blank pages!. But they do seem to be somewhat large for that room.

I am attaching the only 2 data sheets I could find on the monitors. Chris Pelonis makes a speaker using the same dual concentric driver, and I have seen the sub speaker oriented all sorts of different ways. But who is to say if that is ok. The data sheets say that the dispersion is 90 degrees conical.

so that the drivers are stacked on top of each other,

Does the manufacturer mention that it is possible to do that? The manual should tell you. Not all speakers can be turned sideways, and indeed even the ones that can be, seldom should be. Your NS-10s, for example, are lying on their sides, which is a common error with that type of speaker, especially with a console that is reasonably wide, such as yours. Even small head movements totally distort the perception of the sound field. The NS-10s should be standing up vertically, as they were designed to do. It might be possible to turn those Tannoys sideways, but I would check with the manufacturer to find out for sure.[/quote] I know it's wrong to put them on their sides, but I'm so used to hearing them like that, it's hard for me to stand them up.

this would also bring the listening position significantly closer to the front wall.

Not necessarily, and with the design of those speakers, probably not. Where is the acoustic axis of those things? How are those speakers aimed? Where is the intersection of the axes, in relation to your ears? Is that diagram you posted of the studio layout back on August 10, accurate and correct? If so, please add the speaker axes to that, and re-post it. Or better still, post the actual SketchUp file, with the axes added.

I will post the latest diagram. However, looking at the diagram, I had had the mid/HF driver on the inside and the sub on the outside. putting the listening position much farther forward.

I'm suspecting that the 116 Hz dip is probably modal. IF those dimensions you gave are correct, then this is the area around that frequency: 111.1 hz 3.9% 10'2", 5'1", 2'7" (2,2,1 Oblique) 113.7 hz 2.2% 9'11", 4'12", 2'6" (3,0,1 Tangential) 118.0 hz 3.6% 9'7", 4'9", 2'5" (3,2,0 Tangential) 118.1 hz 0% 9'7", 4'9", 2'5" (0,3,1 Tangential) 118.7 hz 0.5% 9'6", 4'9", 2'5" (3,1,1 Oblique) 120.0 hz 1% 9'5", 4'8", 2'4" (0,0,2 Axial) I'm guessing that it's actually the 0,0,2 axial mode you are seeing there, which is related to the vertical axis of the room (height), so modifying the rear wall is not going to help much. All of those (except 118.0) involved the ceiling, and all of them as the first or second harmonic. I suspect the ceiling.

I have never been so happy to be wrong! :)

At my mix position it is around 114hz, and as I roll my chair back to the rear of the room, it goes up to around 200.

That isn't evident on the graphs at all. 200 Hz doesn't change at all for any of the readings for the 9 and 10 foot positions as measured by REW. I loaded the impulse response files into REW as well, but they are highly suspect: they claim to have been taken at a level of over 150 dB! :shock: Somehow, I doubt that is true!!! I had to adjust them down by 84 dB to get them to the same level as what was measured by REW. They also suddenly crash down to nothing at 2 kHz, which is also highly unlikely. How were those measured?

But even though they are suspect, even they do not show any changes at 200 Hz. In fact, 200 Hz is about the most stable and consistent point across all of the graphs! So I don't understand why you have the impression that 200 Hz has a null at any of those locations. Maybe you did not have the mic at the same height as your ears when you measured?

I should not have used 200hz as an exact number. I used that number as a high range number, because I assumed that if I kept going closer to the rear wall, eventually there would be a cancelation at 200 if it was caused by the dip. The cut off at 2k was some sort of default setting, that I have now changed.

The angled grey sections of the ceiling are .5" mdf panels filled with ultratouch and pink fluffy.

Why? What is the purpose of that? 1/2" wood over a 1" - 4" gap will, indeed, give you major absorption in the range of about 150 Hz to 80 Hz, give or take a bit. That's exactly in the range where you are seeing a major null... Hmmm....

So this is where things get confusing for me... If the panels are not big enough to break up room modes, how would they cause a null? would they really be able to absorb so much that they cause a null?

said that it would serve 2 purposes. 1. breaking up the floor to ceiling modes in that area,

It's way too small for that, and in the wrong location too. Modes are at low frequencies, meaning large wavelengths. Waves pretty much ignore objects that are significantly smaller than their own wavelength. At 116 Hz the wavelength is 9.7 feet long, yet that device seems to be only a few inches across: hence, it is doing nothing to the modes. Modes are big problems: they need big solutions.

The panels are 24" x 24" x .5" there are 2 rows of 8 across the ceiling. You just tell me what to do, and I will build it and then report back with measurements! :)

2. serving as a low mid frequency absorption. I think it functions as a dampened membrane?

Yup. It sure is doing that, apparently! And is probably tuned rather tightly, perhaps to the problem frequency you are seeing.

I can dig up some pictures of the construction if it helps.

That would be great, yes! ANd if you have photos of how the rest of the room was built too, that would also help. Especially the soffits.

The original plan was to use those panels to mount these diffusers.

It's not clear at all what theory those are intended to operate on, and there's no useful test data on the web site, so I would DEFINITELY skip those! Whenever a manufacturer of acoustic devices does not publish test data from a reputable independent acoustic laboratory, one has to wonder why.... The question comes to mind: What are they afraid of? If they were confident in their products, then they should have no problems in getting them tested. The fact that they don't, leads one to wonder.... Don't get me wrong! I have no doubt that those things do diffuse: But then again, so does ANY piece of randomly warped material. The question is not "Do they work", but rather "How do they work?". If there is no theory, no peer-reviewed paper, and no independent test results, then I would just walk away from such a product.

When I first started this project, I did not know that the acoustician was planning to build a LEDE room. As things unfolded, I realized I did not want that kind of room. So now I am trying to get as close to a Newell NE type of room as possible. From reading his books and seeing how much volume is required for a proper NE room, I know it is not possible, but I must do whatever it takes to get my decay times even across the spectrum.. right?

Assuming that my problem is being caused by low frequency reflexions from the rear wall coming back to the listening position.

I am not at all convinced that your problem at 116 Hz is related to the rear wall. I suspect the ceiling, and most of all I suspect those wooden angled thingies...

That is great. What would be the next step? taking down the wood panels and then measuring again?

it seems like in order for the hangers to be effective, they need to be combined with some sort of limp membrane absorber on the rear wall?

Not really. Hangers work down to fairly low frequencies, and they do not need membrane absorbers in order to work. They work fine all by themselves. A membrane absorber may help, by providing additional absorption at a specific frequency, but it won't help the hangers do their job.

And for hangers, would this sound board material be appropriate for my application?

No. Way too heavy. What you need is something called "homasote". It is a very light weight, low density fiber board product, of the type used to make bulletin boards for use with push-pins (thumb tacks). It is soft and a bit flaky.

Great. I will look to find a local distributer. - Stuart -

I am posting some photos of the steel plate absorber that was installed on the rear wall. I will post the calculations on it later. I started doing measurements after it was installed. So I plan on removing the plate, and taking more measurements to see exactly how it is affecting the room. The null was perceivable before the plate was installed. The rear wall consists of 2 sheets drywall, stud framing with 3.5" cotton insulation, steel plate trap, pink fiberglass insulation, stud frame with 3.5" cotton insulation, fabric covering. It seems the first order of business would be to remove the steel plate, take a measurement. Remove the ceiling panels, take another measurement, and then go from there? Thanks again Stuart!

Stuart, Here is a link to the Homasote website for their pinnacle tack board. it is 1.2lbs / sqft. is this the right stuff? http://www.homasote.com/products/PinnacleBoard.aspx Beau

I am planning on treating these problems when I return home from vacation. Are you able to give me a directions list on how to treat the problems properly and an order I should perform them and take measurements? Here is what I am thinking. 1. Take calibrated measurements with the room in it's current state. 2. Take down rear ceiling treatment. Measure again. 3. Take down rear wall treatment. Measure again. 4. Take down front ceiling treatment. Measure again. 5. Measure walls, corners and ceiling to find areas where the frequency of nulls have the most energy. 6. Treat those areas first. Now for some questions. When it comes to corner trapping, it seems like most people agree that a super chunk design performs better than straddling absorption across a corner, at the expense of much more material. Does this apply to trapping an entire wall? So my question is if I were to treat the entire ceiling with 6" of roxul safe n sound with a 6" air gap above it, would there be a gain of filling the air gap with light weight fiberglass insulation? or even more roxul safe n sound?

Here is a link to the Homasote website for their pinnacle tack board. it is 1.2lbs / sqft. is this the right stuff?

Yup! That's the stuff. - Stuart -

1. Take calibrated measurements with the room in it's current state. 2. Take down rear ceiling treatment. Measure again. 3. Take down rear wall treatment. Measure again. 4. Take down front ceiling treatment. Measure again. 5. Measure walls, corners and ceiling to find areas where the frequency of nulls have the most energy. 6. Treat those areas first.

#1 and #2 are great ideas, but I don't see the need for #3. You basically have a couple of feet of pure absorption there, and it certainly seems to be working fine for the modes that are associated with it, so I don't think it is necessary to take it out. #4 is a possibility: it might be better to replace that with a hard-backed cloud with thick insulation both above and below it. #5 is not necessary: All modes terminate in corners: it's just the way standing waves work, due to the laws of physics. So there's no need to go hunting for them: they will be there in the corners. That's why bass trapping is most effective in corners, especially the "tri-corners", where two walls and the ceiling meet, or where two walls and the floor meet.

When it comes to corner trapping, it seems like most people agree that a super chunk design performs better than straddling absorption across a corner,

Right.

Does this apply to trapping an entire wall?

Yes.... sort of!

So my question is if I were to treat the entire ceiling with 6" of roxul safe n sound with a 6" air gap above it, would there be a gain of filling the air gap with light weight fiberglass insulation? or even more roxul safe n sound?

Yes, there is a benefit to doing that, since all sound waves striking that wall will have to move through much more absorption both on the way in (towards the wall) an then again in the way out (after bouncing off the wall). So that increases efficiency of absorption, and a deeper overall depth of insulation (wall surface to insulation surface) also absorbs down to lower frequencies. - Stuart -

Thank you Stuart! :) Ok, all this seems like good news. What is the benefit of putting a hard backing behind the ceiling in the front of the room? I am going to sketch up the treatment to make sure I understand everything and repost. Thanks again, Beau

What is the benefit of putting a hard backing behind the ceiling in the front of the room?

It doesn't go behind the ceiling: The ceiling is already in place. What I'm talking about is a "cloud", which is a large thing that hangs a bit below the ceiling. A basic cloud is a wooden frame with insulation in it, above the mix position and console, hanging on chains from above, but not covering the entire room with, wall to wall: there's usually a large gap (several inches) around the cloud. It is also frequently angled (lower over the speakers, higher over the mix position). A "hard-backed" cloud also has a thick, heavy wood backing above the frame and insulation (eg, 3/4" plywood), which helps to break up modes to a certain extent. And the space on top can be put to good use by putting some more insulation up there, which helps even more. If it is designed and built right, the angle can then be adjusted at any time simply by lengthening or shortening the chains. There are pictures and designs on the forum for clouds like this. If you do a search for "hard backed cloud" you'll probably find plenty of examples that you can get ideas from. - Stuart -

Yes, I understand about the cloud. I think I may have worded it wrong. The way I understand it, my existing angled front ceiling is a cloud without rigid backing. The isolation ceiling is flat from front to back. 1 layer osb then 2 layers 5/8" sheetrock. The angled front section of the ceiling is 2x4 studs with ultratouch insulation between each of the studs, and above that is low density fiberglass (pink fluffy stuff) Below, the image shows the isolation ceiling studs running from side to side, and the angled "treatment" ceiling studs run from front to back. Will it be better to make the ceiling evenly absorptive across the whole ceiling? I can get away with about 16" of absorption and still have 8' (visual) ceilings. Or would the angled cloud in the front section be better with the hard backed mdf as you suggested. Ok, Regarding #5 about treatment....

#5 is not necessary: All modes terminate in corners: it's just the way standing waves work, due to the laws of physics. So there's no need to go hunting for them: they will be there in the corners. That's why bass trapping is most effective in corners, especially the "tri-corners", where two walls and the ceiling meet, or where two walls and the floor meet.

In this diagram here, it shows an idea for a flat treatment ceiling, (but feel free to disregard that if the angled cloud is better) Once the existing treatment is removed, will these pink areas be the first order of treatment? The yellow represents rear wall insulation already in place. The remaining space would all be filled with insulation. It would seem the 4 rear tri corners are already trapped by trapping the entire rear wall, right?

So... after some more thought... (time on vacation, but all I can think about is fixing the CR issues) A CR revamp seems like it could be a nice idea. Here is what I am thinking. PInk areas in diagram below are designated for absorption. 1. Removing existing ceiling treatment and using either 18" of absorption or hangers. I am trying to find white papers on how to calculate hangers, but I am having no luck.. I thought Boggy or Andre had written them, but I am still searching. 2. Rear wall treatment. A- Removing the steel plate from the frame and replacing it with a limp mass absorber tuned to whatever my lowest frequency needed. B- Hangers or all absorption on rear wall. I just don't know which to use for a fully absorptive rear wall. I am allocating 2 feet in front of my existing wall frame. They are inside out walls, so for example, I could do a newell limp mass rear wall type of thing so that from the outside in would be 2 sheets 5/8" sheetrock, 2x4 studs with ultratouch insulation in between studs, then limp mass across whole rear wall. Then 2 feet of absorption in front of that. Redoing the flush mount wall, so that ........ 1.) the speakers are in vertical alignment on top of each other with the mid / tweeter on top at ear level, roughly 4 feet from the ground. This would be more in line with the Newell thought process of making the mid and bass drivers more in phase as I move from left to right. 2.) bringing the speakers into a tighter triangle, pushing my mix position closer to the front of the room. Then using a 25 degree angle rather than 30 to widen the listening position. I usually prefer to be quite off access. I know 30 is supposed to be the proper angle, but I think I have seen people do 25 degrees all the way to 40 degrees without messing up the stereo field too much, right? Assuming a revamp of the control room would not be a total waste of time and money, I have a feeling the proper procedure would be the following. 1-take out all the ceiling treatment, 2-Remove flush mount wall. 3-Place speakers on heavy stands and experiment with different positions and take measurements to find the best position. 4-Build the flush mount wall based on the best position. 5-Measure again.. 6-Design treatment accordingly.. What do you think? B

1. Removing existing ceiling treatment and using either 18" of absorption or hangers. I am trying to find white papers on how to calculate hangers, but I am having no luck..

As far as I know, there isn't much research on hangers out there! But from the empirical results that John and others have been getting with them, they do work very well. If you have 18" to play with, then I would put 6" of 703 directly on the ceiling, then 12" of hangers.

A- Removing the steel plate from the frame and replacing it with a limp mass absorber tuned to whatever my lowest frequency needed.

I wouldn't bother with the membrane absorber: You don't seem to have a problem with length modes.

B- Hangers or all absorption on rear wall. I just don't know which to use for a fully absorptive rear wall.

You have all absorption there right now, and it seems ot be working just fine. As the saying goes: "If it ain't broke, don't fix it!"

Redoing the flush mount wall, so that ........ 1.) the speakers are in vertical alignment on top of each other with the mid / tweeter on top at ear level, roughly 4 feet from the ground. This would be more in line with the Newell thought process of making the mid and bass drivers more in phase as I move from left to right.

It's not just Newell: it's common sense! Even so, there's an awful lot of studios out there, without too much of this "common sense" being evident! You even see this in magazine articles and photos of "top" studios, where there are a pair of NS-10s or some such directly o top of the meter bridge ( :shock: ), lying on their sides ( :shock: ), and either above or below the ears of the engineer ( :shock: ). You'd think that they would have figured it out by now.... However, I'm not sure that I would do that with your speakers, since they may not have been designed for it. I would first contact Tannoy and ask them if it is OK to mount your speakers like that. And to be very honest, even though those are nice speakers, I think they are just too big for your room. I would seriously consider replacing them with something more suitable.

bringing the speakers into a tighter triangle, pushing my mix position closer to the front of the room. Then using a 25 degree angle rather than 30 to widen the listening position.

If you move your listening position closer to the front wall, you will need a LARGER toe-in angle for the speakers, not a smaller one! Probably more like 35° or so. You can actually go up to 45° if you need to, but that's drastic. The higher the angle, the more you broaden your sound-stage, but you also narrow your sweet spot.

I usually prefer to be quite off access. I know 30 is supposed to be the proper angle, but I think I have seen people do 25 degrees all the way to 40 degrees without messing up the stereo field too much, right?

You should be slightly off-axis, but not much. Most speakers only give you their most accurate frequency response on-axis, and it becomes more and more distorted the further you get off-axis. Ideally, you should be no more than about 5° to 10° off axis. In other words, the acoustic axis of the speakers should be aimed an inch or two beyond your ears, such that the intersect about 12" or so behind your head (or more).

What do you think?

Sounds about right, but with the above caveats! :) - Stuart -