Hi
a couple of months ago i started a topic called ceiling treatment but wasnt able to get REW to work and could not deliver requested info...
BUT
ive done my homework and im writing a short semi scientific paper on a second room in my basement and id need some help:
The room's 374x287x245cm (LxWxH) concrete walls and ceiling, tiled floor, one window ('behind the listener) and one door (see the attached pics) ive used this site to calculate the modes if the room ( http://www.trikustik.at/wissen/rechner- ... m_store=at )
i used REW to proof the modes by measuring having the speaker (ADAM A7X) at the wall and moving the mic (Behringer ECM 8000) through the room (calculating 'dead spots' with the wavelengths of the frequencies and positioning the mic there and near the walls to show mins and maxs)
I want to use the room mainly for mixing and maybe some 'controlling' (guitar or drum recording in an other room) so i put my speakers in it and tried to find the place where listening sounds the less boomy. i found that to be having the speakers 70cm off the wall being 110cm apart (thats where they going to be on my desk) symmetrically in the room - see the attached pictures and drawing. i made measuring with speakers right against the wall and 70cm of it (see REW)
https://www.dropbox.com/sh/2kxcv3vg6fe4 ... pjXT5ojT6a
as far as treating the room:
id start of with some super chunks behind the speakers - are there any formulas for how they should be done (ive read that weight of the rockwool is important - different weights should be use - heavy next to the wall and getting lighter to the room; too heavy might end up in reflecting as the air cant get in?!?!?)
after that id do some more measuring and maybe add some tuned absorber (helmholtz) in the corners behind the listener...
i know that this is not the end but id like to go until here for the paper im writing with as much scientific background as possible
i use the Master Handbook of acoustics (everest) for reference so if u have advices for me on which chapter i should focus on please let me know
thanks for your help!
Kurt
room treatment once more
-
strat_cat21
- Posts: 6
- Joined: Sat May 18, 2013 7:10 am
- Location: Austria
- Contact:
-
Soundman2020
- Site Admin
- Posts: 11938
- Joined: Thu Aug 21, 2008 10:17 am
- Location: Santiago, Chile
- Contact:
Re: room treatment once more
Whoaaa! I don't know how I managed to miss this thread for nearly a month! 
sorry about that! But "better late than never", as the saying goes...
That room is a reverb chamber! 
The overall decay time is well over 4 SECONDS! It's over three seconds all the way up to 2.5 kHz, and even at 10 kHz, it is still way over one second.... That room is going to need major treatment. Big time major treatment!
There are a couple of issues with the way you are doing your tests that need to be fixed, so you can get more accurate readings. The ones you did already tell the basic story about the major problems with the room, but there are some things you need to fix for getting better readings, once you start treating the room. First, the speakers should not be on their sides. Not for the tests, and certainly not for mixing. They should be standing upright, with the tweeter on top (usually). There are many reasons for that, which I won't go into here, but it's a mistake to use those speakers lying down. (Yes, I know the A7X manual says you can do that, and there are many photos on the internet and in magazines of studios laid out like that, but it is not optimal). So first set them upright, then make sure that the acoustic axis of the speakers is about 1.2m above the floor: Adjust your speaker stands as necessary to get that. It is important.
Next, the levels are not correct in REW. That is most likely because you didn't do the calibration process using a hand-held sound level meter, and also because you changed the levels between tests. There are differences of over 30 dB between some readings, but there should never, ever be a difference of more than about 6 dB, or perhaps just a bit more. The correct way to do the tests is to calibrate each individual speaker to produce 80 dB by itself, which then means that wen both speakers are going together the level should be between 83 dB ant 86 dB without you adjusting anything! That's just the natural sum of the two sound fields. So always do your tests with each speaker at 80 dB, then do the combined test (both speakers) without adjusting anything, and the level should automatically be about 86 dB.
Right now, it is impossible to accurately compare the various tests you did, since they were done at different levels.
Finally, you should set REW to run the tests from 20 Hz through to 22 kHz. Even though the A7X's don't do a good job down below about 45 Hz, there's still a lot of stuff going on down there that you need to see. And 22 kHz is no problem at all for the A7X: it goes way above that. Your sound card should also be able to handle that, without any trouble, even with a 44 kHz sampling rate.
So please repeat the tests like that, and keep those levels correct for all future tests!
Take a look at the SPL curves for the "against" and "70cm" tests that you did:
The yellow curve is with the speakers up against the wall, and the blue curve is with them at 70 cm: You can clearly see that the entire bottom end, below about 80 Hz, is severely attenuated in SPL level: There's a loss of over 12 dB between those two, which is only explainable by phase cancellation. There's also a huge dip (nearly 23 dB)at 85 Hz, which was not there with the speakers against the front wall, so that must be SBIR. There are other smaller dips at 129 Hz and 175 Hz, and a peak at 229Hz, which are probably also SBIT or some other form of phase-related issue, since the are onl there for the 70cm location. But you can easily see that the overall response with the speakers up against the front wall is much smoother, overall.
You don't say where the mic was positioned for these tests, but if you do the math, I'm betting you'll find that the 85 Hz dip at the very least is directly related to the difference in path length between the direct and reflected sound.
I would suggest you do another test with the speakers about 10cm away from the front wall, located about 156 cm apart, and with the tip of the mic located at about 145 cm from the front wall, 1.2m above the floor, and on the room center line. That will be roughly where the everything should be, for that room, once it is set up. Post the results for that test, so we can take a look at them.
You do not need to mix different densities. It just makes it more complicated, for very little benefit (if any). It's not that "too dense stops the air getting in", but rather that "too dense is worse for low frequencies". Denser absorption is better for high frequencies, and less dense is better for low frequencies.
So forget tuned devices for that room: What you need is massive broad-band absorption. Do those four vertical superchunks to start with, as well as putting 10cm panels of OC-703 between the speakers and the front wall, then covering the entire real wall with 15 cm of OC-703, another panel of OC-703 at the first reflection points on the side walls, and an angled hard-backed cloud hanging form the ceiling, also filled with OC-703.
Do a REW test after you install each of those devices, so you can see the progress. And it is vitally important that you place the test mic in the exact same position inside the room for each test! It MUST be precisely located, accurate to within a few mm, in the same position for all tests. If not, then you cannot validly compare the readings.
So I'd suggest you start by doing all that, with careful REW tests each time you install something new, then post the results (and photos!) each time, so we can take a look and help you analyze the results.
- Stuart -
sorry about that! But "better late than never", as the saying goes...That looks a lot like Andy Mel's calculator: He's a member of this forum, and his thread about the calculator is here: http://www.johnlsayers.com/phpBB2/viewt ... =3&t=18613 That's one of the best calculators out there, so you are in good hands if you used that.ive used this site to calculate the modes if the room
The ratio is not fantastic, but it is also not terrible. It's on the edge of the Bolt area, which is good. It's a very small room, though, so it will need a LOT of treatment.The room's 374x287x245cm (LxWxH)
I downloaded your REW files, and: WOW!!!!i made measuring with speakers right against the wall and 70cm of it (see REW)
That room is a reverb chamber! The overall decay time is well over 4 SECONDS! It's over three seconds all the way up to 2.5 kHz, and even at 10 kHz, it is still way over one second.... That room is going to need major treatment. Big time major treatment! That would explain the very high level, very long decays times.concrete walls and ceiling, tiled floor,
There are a couple of issues with the way you are doing your tests that need to be fixed, so you can get more accurate readings. The ones you did already tell the basic story about the major problems with the room, but there are some things you need to fix for getting better readings, once you start treating the room. First, the speakers should not be on their sides. Not for the tests, and certainly not for mixing. They should be standing upright, with the tweeter on top (usually). There are many reasons for that, which I won't go into here, but it's a mistake to use those speakers lying down. (Yes, I know the A7X manual says you can do that, and there are many photos on the internet and in magazines of studios laid out like that, but it is not optimal). So first set them upright, then make sure that the acoustic axis of the speakers is about 1.2m above the floor: Adjust your speaker stands as necessary to get that. It is important.
Next, the levels are not correct in REW. That is most likely because you didn't do the calibration process using a hand-held sound level meter, and also because you changed the levels between tests. There are differences of over 30 dB between some readings, but there should never, ever be a difference of more than about 6 dB, or perhaps just a bit more. The correct way to do the tests is to calibrate each individual speaker to produce 80 dB by itself, which then means that wen both speakers are going together the level should be between 83 dB ant 86 dB without you adjusting anything! That's just the natural sum of the two sound fields. So always do your tests with each speaker at 80 dB, then do the combined test (both speakers) without adjusting anything, and the level should automatically be about 86 dB.
Right now, it is impossible to accurately compare the various tests you did, since they were done at different levels.
Finally, you should set REW to run the tests from 20 Hz through to 22 kHz. Even though the A7X's don't do a good job down below about 45 Hz, there's still a lot of stuff going on down there that you need to see. And 22 kHz is no problem at all for the A7X: it goes way above that. Your sound card should also be able to handle that, without any trouble, even with a 44 kHz sampling rate.
So please repeat the tests like that, and keep those levels correct for all future tests!
The reason the sound less boomy like that, is because you have set them and yourself up in what seems to be a combined modal null and SBIR null!so i put my speakers in it and tried to find the place where listening sounds the less boomy. i found that to be having the speakers 70cm off the wall
Take a look at the SPL curves for the "against" and "70cm" tests that you did:
The yellow curve is with the speakers up against the wall, and the blue curve is with them at 70 cm: You can clearly see that the entire bottom end, below about 80 Hz, is severely attenuated in SPL level: There's a loss of over 12 dB between those two, which is only explainable by phase cancellation. There's also a huge dip (nearly 23 dB)at 85 Hz, which was not there with the speakers against the front wall, so that must be SBIR. There are other smaller dips at 129 Hz and 175 Hz, and a peak at 229Hz, which are probably also SBIT or some other form of phase-related issue, since the are onl there for the 70cm location. But you can easily see that the overall response with the speakers up against the front wall is much smoother, overall.
You don't say where the mic was positioned for these tests, but if you do the math, I'm betting you'll find that the 85 Hz dip at the very least is directly related to the difference in path length between the direct and reflected sound.
I would suggest you do another test with the speakers about 10cm away from the front wall, located about 156 cm apart, and with the tip of the mic located at about 145 cm from the front wall, 1.2m above the floor, and on the room center line. That will be roughly where the everything should be, for that room, once it is set up. Post the results for that test, so we can take a look at them.
The speakers should NEVER be on the desk! The should be BEHIND the desk, on stands, close to the front wall. If you put them in the desk, you create a whole additional set of acoustical issues that could be avoided.being 110cm apart (thats where they going to be on my desk)
You will need superchunks in at least four corners of the room, and probably six, or even more corners. Fortunately, the shape of your room allows for that. So start by making four superchunks for the four vertical corners. Make them BIG: about 90 cm across the front (diagonal) face, which should make them about 60 cm deep down the walls. Floor to ceiling. Use ordinary OC-703 for that, or something similar. If you use fiberglass, then look for something with a density of around 30 kg/m3, or if you use mineral wool, then about 50 kg/m3 is right.id start of with some super chunks behind the speakers - are there any formulas for how they should be done (ive read that weight of the rockwool is important - different weights should be use - heavy next to the wall and getting lighter to the room; too heavy might end up in reflecting as the air cant get in?!?!?)
You do not need to mix different densities. It just makes it more complicated, for very little benefit (if any). It's not that "too dense stops the air getting in", but rather that "too dense is worse for low frequencies". Denser absorption is better for high frequencies, and less dense is better for low frequencies.
I would not do that: Tuning Helmholtz devices is very complicated: it's nowhere near as easy as the text books make it sound! Modal resonance is very narrow band (just a few Hz), and tuning a resonator that precisely is hard. Then there's also the problem of locating the mouth of the resonator at the correct position in the room: The mouth MUST be at the pressure peak for the specific mode that you are trying to treat. For example, if you decide to tune your device to 70.3 Hz, which is your first vertical mode (0,0,1), then there would be no point at all in putting that device on the back wall: it would have to go on the ceiling, since that's where the pressure node is for vertical modes. It would look pretty strange having a huge resonator hanging from the middle of your ceiling!after that id do some more measuring and maybe add some tuned absorber (helmholtz) in the corners behind the listener...
So forget tuned devices for that room: What you need is massive broad-band absorption. Do those four vertical superchunks to start with, as well as putting 10cm panels of OC-703 between the speakers and the front wall, then covering the entire real wall with 15 cm of OC-703, another panel of OC-703 at the first reflection points on the side walls, and an angled hard-backed cloud hanging form the ceiling, also filled with OC-703.
Do a REW test after you install each of those devices, so you can see the progress. And it is vitally important that you place the test mic in the exact same position inside the room for each test! It MUST be precisely located, accurate to within a few mm, in the same position for all tests. If not, then you cannot validly compare the readings.
So I'd suggest you start by doing all that, with careful REW tests each time you install something new, then post the results (and photos!) each time, so we can take a look and help you analyze the results.
- Stuart -