Hi there "Bin Salem", and Welcome!
with the following ratio :
1.4 W/H (7/5 diminished fifth )
1.67 L/H (5/3 major sixths )
1.2 L/W ( 6/5 minor third )
I think you've been reading too much Wes Lachot! He's a great designer, and has made some really good studios, but I'm not aware of any good reason why you would want to tune your modal response to the musical scale! I know he promotes that as part of his signature design process, but I'm not aware of any research, paper, or principle of acoustics behind it. My single question about it is: "Why?" If someone can explain to me why it is acoustically or musically significant that a control room should be tuned to a specific chord or set of musically related notes, then I'd be happy to incorporate that into my future designs, but until then, I remain unconvinced. I'm also curious as to how that works out if my room is tuned to C# (for example) and I happen to be mixing a song that is in the key of F...
In my opinion, it is more important to have greater room volume and greater floor area, than it is to tune to a specific musical relationship.
In other words, if you had to reduce the height, width, or especially the length of your room in order to get to that ratio, then forget it: Go bigger.
all within
EBU / ITU recommendation
IEC recommendation
No it is not. All of those (IEC / EBU / ITU) recommend a minimum floor area of at least 20m2 (with some of them it is even higher): Your floor area is only 75% of that (15m2). Therefore it is not within the recommendations. So if you can, make your room bigger.
I angled the side walls little bit by 4.37 degrees to deal with standing waves
I see that frequently, and I have no idea where that wrong information is coming from, but the simple truth is that
angling your walls does nothing at all to "deal with standing waves". Zero. Zilch. Nada: Nothing. Zip. It isn't even logical to think that it would.
Not only that, but by angling the walls, you totally threw away all the work you did in calculating your room ratio!
Room ratios
ONLY apply to rooms that are perfectly rectangular, and have six sides (4 walls + ceiling + floor) that are mutually perpendicular and parallel. That is the ONLY condition where room mode calculators are valid. If your room does not have parallel walls, or has more than 6 boundary surfaces, then room mode calculators do not apply, and all of your predictions are meaningless, and wrong.
Here's the issue: "standing waves", more correctly called "room modes", are simply specific frequencies where the wavelength corresponds to a path around the room that a wave can take, then arrive back at it's starting point in the room, going the same way as before, and in phase with itself. That's all. If there is a path in your room that meets those conditions, then you will have a "mode" or "standing wave" for that specific frequency (wavelength). If you then move one of the walls associated with that mode, the ONLY thing you will accomplish, is to change the frequency! The mode will still be there, the standing wave will still form.... it will just do that at a slightly different frequency. So by changing the position of your walls, all you do is to change the frequency of some of the room modes. This is the basic reason why room mode calculators (room ratio calculators) exist: to ensure that the room modes are evenly spaced, and do not all happen at the same frequency as each other.
So changing the position of the walls does nothing to make the standing waves go away: It just makes them happen at different frequencies. That's all.
And angling the walls a bit also does nothing!
There are three types of room mode (standing wave): There are "axial modes" that form between two opposite surfaces, there are tangential modes that form between four surfaces, and there are oblique modes that form between all six surfaces. You might think that if you change the angle of one wall, so that it is no longer parallel to the wall on the other side of the room, then that would eliminate the modes associated with those two walls. Wrong. It does not do that. If you just change the angle by a small amount, then there's practically no effect: All that happens is that the mode becomes "broader" It now happens across a small range of frequencies, instead of at one specific frequency. Which makes it harder to deal with using tuned treatment (such as a membrane trap).... Think about it.
And if you angle the wall at a very large angle, large enough so that the axial mode disappears, it is simply replaced by one or more tangential modes and/or oblique modes. You cannot "deal with" modes by angling walls.
so at front became 329 at front & 395 back so average is 362 cm
If you angle your wall at a very low angle like that, then you end up "blurring" the axial modes associated with those two walls. So instead of having one single sharp mode at 47.6 Hz, you now have a "broad" mode that covers the range from about 44 Hz to 52 Hz. All frequencies in that range will now trigger the mode, potentially, and it might change frequency as it decays. And now you can no longer deal with that mode using tuned membrane traps... because the mode is no longer at one specific frequency: it is spread out...
So as you can see, you have not "dealt with the standing waves" at all! All you did was make it harder to predict where they will be, and "smear" them out.
In addition, by angling the walls like you did, you have reduced the floor area considerably. If you were to leave the entire room at 392 cm wide, then you would have 17m2 of floor area, instead of the 15 m2 you have right now... You would have 10% more floor area by not splaying the walls. It would also be much easier to build, as you would not be trying to cut your building materials at the rather strange angle of 4.37°!
the finished walls is 3 layers of drywall with construction adhesive glue them
Why???? Gluing the layers of drywall together is a really bad idea: it makes them act as one solid block, instead of allowing them to act as individual panels. That reduces your isolation in the lowest frequencies, and also increases the size of the coincidence dip. Do not ever glue your layers together (except with Green Glue, but that isn't glue anyway!).
with 10 cm of rock wool 100 kg/m & 2 cm of air gap at front to deal with first reflections
100 kg/m3 is much too dense for low frequency absorption. It should be no more than about 50 kg/m3 if you use mineral wool, or about 30 kg/m3 if you use fiberglass insulation. If you use a higher density, then you get worse absorption in the low frequencies.
what that can reach in the low absorb coefficient below 200 hz ???
Don't just look at the coefficient of absorption: It's just one of many aspects of the asborber. Look at
all of the acoustic properties of the treatment device. Coefficient of absorption does not tell you everything you need to know...
then the front wall with the 10 cm of rock wool 100 kg/m & 2 cm of air gap to deal with 1/4 wavelength
Once again , 100 kg/m3 is way too high. I'm not sure what you mean by "deal with 1/4 wavelength": Porous absorption does not absorb just one frequency at one specific wavelength: Rather, it absorbs a wide range of frequencies, some better than others. Yes, there is a relationship between wavelengths and the depth of the absorber (front surface to wall surface), but that is not a sharp cut-off point. It is a gentle roll off. The absorber still works down to much lower frequencies.
but what about back wall
The back wall should have either acoustic hangers across the entire width but larger in the corners, or it should have large "superhcunk" traps in the corners with the rest of the surface being covered with at least 15cm of porous absorption similar to OC-703.
I am thinking of membrane absorber
Why?
When you read about them in text books, it all sounds so easy and so effective. But when you try design them and build them in real-life, the results are rather different... In general, tuned absorption is not a good idea, unless you really know what you are doing, and have a method for measuring and adjusting the actual tuning of the device. Things never work out the way you think they will, or the way the equations predict. Equations assume that everything is perfect: real life isn't perfect. Materials are porous, don't have the exact same density / thickness assumed by the equations, the air might be at a different temperature in your room, or at a different pressure / density... etc.
then QRD diffuser above it,
Your room is way too small to be able to use numeric-sequence diffusers. You say that your design meets Trevor Cox recommendations, so I'm assuming you have read some of his publications.... but if you read his books and papers on diffusers (mostly written together with D'Antonio) you will find that he does NOT recommend diffusers for this situation. He says that you should never use diffusers if the distance from the diffuser to the ears of the people in the room who are doing "critical listening", is less than ten feet (3m). So forget about using numeric diffusion in your room: it just is not big enough.
I feel better choice for me is to use membrane absorber
Why? They are just as hard to tune as diaphragmatic absorbers, and you still need a method for measuring the actual tuning of the device after you build it, and for changing that tuning if it did not turn out correctly.
Also, what frequencies would you tune your devices too? You have 18 room modes below the Schroeder frequency, so you would need to build 18 devices, each tuned to a specific frequency. Each of those devices would need to be large enough to have at least 1% of the entire room volume, in order to have a useful effect, so you would need to fill up nearly 20% of the entire room volume with membrane traps...
You would need traps on every single square inch of wall, ceiling, and floor space to do that, and even then you would not get the effect you want. You would be creating more problems than you solve.
Porous absorption is a LOT more efficient. It is broadband, so it covers the entire frequency range that you need to damp, so one single device (or a small number of them) can cover all 18 of those frequencies.
I see some designer use plywood for the whole back wall but how calculate the thickness ? the depth ?
The equation is not hard to find:
Equation for membrane traps:
d = 28900 / (M * f^2)
Where:
d = depth of airspace in inches
M = surface density of panel, lb / ft^2
f = peak absorbing frequency
(You will have to do the conversion to metric)
I just find John sayers sluts but it's different & expensive
I think you mean "slats" not "sluts". Slat walls (or "slot walls") are not membrane traps! They are Helmholtz resonators. They don't work well for modal frequencies either: they are much more suited to mid range and higher frequencies.
couldn't find any absorb coeff. for rockwool below 125 hz, how I can find it ?
You can't. Because they don't measure it! It is very difficult to accurately measure most acoustic properties of materials for very low frequencies. You might find some cases where it has been measured down to maybe 63 Hz, but below that it is really, really hard to do, so nobody bothers. It isn't necessary either. If you know the GFR (Gas Flow Resistivity) for the material, then you can roughly predict how it will perform.
Your diagram shows the entire rear wall is one huge membrane absorber. You have at least 18 frequencies that you need to deal with (actually, more like about 30). So which one of those frequencies would you tune your wall to, and what will happen to the other 17 frequencies that you DON'T treat?
I have to treat the side walls to to achieve 0.6 abs coeff. to solve 96 & 195 peaks & 175 dip
Why did you chose only those frequencies? What about the tangential and oblique modes associated with the side walls? Why are you ignoring those? And why did you want a coefficient of absorption of 0.6? Why not 0.7, or 0.8, or 1.0, or even 1.2, for example? Why do you want to use an absorber at less than maximum efficiency. When I design acoustic treatment, I look for materials that provide the highest possible coefficient of absorption at the frequency I am trying to treat. If I designed something specifically to use only absorption that has a coefficient of absorption of 0.6 at the frequency I want to treat, then it would have to be twice as big as a device that had a coefficient of absorption of 1.0 at the frequency I want to treat. I don't understand why you want to have a device that is less than 100% efficient. Why only 60%
I'm also not sure what you mean be "achieve 0.6 abs coeff. to solve ...". Coefficient of absorption only tells you how effective a device is for treating
normally incident sound. The tables show only a few specific points for the material, but there is actually a curve that covers the entire spectrum. It is more effective at some frequencies, less effective at other frequencies, and there's no reason to choose a material that shows 0.6 for a specific frequency. Also, that is only about normal incidence, but most of the sound in the room is not normal incidence: it arrives at the surface of the absorber at many angled, not just 90°.
achieve 0.7 absorbing at ceiling to treat 132 hz peak
achieve 0.7 in back & front walls to treat 40,80,117,200 hz
If you cover 70% of your walls, and 70% of your ceiling, with one specific porous absorber type and depth, then you would have too much absorption at some frequencies and not enough at others. The room would be very dead, too. You probably don't need more than about 55% coverage for that room, but it has to be of the correct type and in the correctly location: 100% on the rear wall, maybe 30% on the front wall, 90% on the ceiling, and the rest on the walls, to get the total number of sabins for your room.
Also, you are using a very simple calculator to predict a very complex result, and you are not even using it correctly. For example, why did you choose that specific position for the speakers and listening position when you set up the prediction? If you move those speakers and the listening position by even a small amount, the curves will look very different....
Also, the REW simulation assumes that your room is rectangular, with parallel walls. Your design is not rectangular, and the walls are not parallel, so the result predicted there is not valid.
the problem if we treat it we can't change later, because we have to cover the walls with fabric then no way back,
then don't do that!
Do not treat it in such a way that you have no way of changing it! Rather, treat it in ways that do allow you to change it.
---------
It seems to me that you are putting way too much importance in trying to predict how your room will perform. There's nothing wrong with that (I do it all the time when I'm designing rooms), but you should also take into account that the prediction is only valid for a very precise set of assumptions, and one of those is that the room is built perfectly, with perfect materials, perfect seals, perfect dimensions, perfect uniformity, etc. Real life is not like that. Wood is not perfectly straight, so your walls wont be either. The density of wood / drywall etc. is not perfectly constant: it changes across one sheet, and between sheets. You cannot measure and build with perfect accuracy: Your walls will vary slightly in thickness, straightness, vertical "plumbness", density, angles, etc. Your speakers will not be perfect, and will not produce perfectly linear response: The room itself places an acoustic load on the speakers, which affects how they perform. That alone is a major issue: there are very large differences between the way a speaker performs in an acoustic free field (no walls around it) or an anechoic chamber, compared to the way it performs in a real room with a floor, walls, and a ceiling. Even moving the speakers by a few cm can change the way they respond, and also the way the room responds to them.
Etc.
So it's fine to use the prediction tools to get a rough idea of how the room will perform, but don't assume that all the predictions will be correct: They won't!
I NEVER use prediction software to design acoustic treatment before the room has been built. First, I build the room, then I analyze how it is REALLY performing when it is empty, and I design treatment to deal with the specific problems that are show by the actual data, not the problems that were predicted before. Sometimes they are similar, sometimes they are not.
I do this sequentially: First I design and install basic treatment for the biggest problems, which is always low frequency modal and SBIR problems. The I analyze again, because that treatment will change the room! The treatment I install in the first stage will deal with the problems that I designed it deal with, but it will also change the rest of the response in other ways! so even though I designed bass traps, they will have an effect on the mids and highs too. Sometimes the effect is good, and sometimes it is not. So basically I now hve a different room to deal with! It has changed, because of the treatment: And it will change again when I install the next set of treatment. So I do it in stages: analyze - treat - analyze - treat - analyze - treat - analyze - treat - ... I do that as many times as necessary to get to the result the owner wants. Here's an example:
http://www.johnlsayers.com/phpBB2/viewt ... =2&t=20471 .
Am I in the right track ? not sure if what I am doing is right, or this absorbing coefficient can be achieved or the reality far from what this simulation
The coefficient of absorption is just a number. It just tells you something about the relative effectiveness of the absorption of a specific material at a specific frequency, when measured in a specific manner in an acoustic testing laboratory. That's all it is. It's a useful parameter to take into account when choosing specific materials to deal with specific problems, but that's all. It is just one of many parameters.
what do you think about ratios ?
I think they are interesting, but that's all. I might use ratios to get a rough idea of how the basic room will perform, but then I'll modify the shape and size and other aspects of the room, as needed. As long as your ratio is far away from the bad ones, and roughly close to one of the good ones, that's all you need to be worried about.
how much coefficient with 10 cm of 100 kg/m rockwool + 2 cm air gap ?
I think you are not understanding what a coefficient is! You seem to be misusing the term. I'm not sure what you think it means, but I do know that it doesn't mean what you thin kit means...
how I can treat back wall ?
I already mentioned that above: Either do superchunks in the corners and absorption in between (but also take care to not kill the high end), or use acoustic hangers across the entire rear wall, making them larger in the corners. Both will work. In both cases you will probably need additional treatment to return some of the high end to the room, or it will be over-absorbed.
how I can design membrane absorber for full back wall ?
Don't. It's that simple. Don't do it.
how I can achieve 0.7 absorb ceiling ?
Simple! You just cover 70% of it with suitable absorption! I think you don't understand what you are seeing in the REW simulation. The numbers in the "Surface absorption" fields have nothing at all to do with coefficients of absorption! Those numbers refer to how much of the surface area is covered with absorption. "1" means that 100% of the surface is covered, "0.5" means 50% is covered, "0.1" means that 10% is covered. That's all. It is not related in any way to the coefficient of absorption. Rather, it assume Sabine absorption.
- Stuart -
