Speaker placement and room dimensions

Started by BWO on 20 January 2011. 8 replies.

Originally posted at johnlsayers.com, topic 15517.

First post here. This is not question regarding a studio, but it's about a listening room. Hope that's ok. I'm asking it on behalf of someone else. I may go there and do some measurements in near future. But I wanted to ask firstly about what problems the dimensions of the room may give and if it's possible to say where the better speaker placement would be. The room is 8,2 m in length, 4,7 m wide and the height is 2,42 m. The owner says he can't get a decent result and has tried a lot of different placements, also the 38% rule. He says it lacks bass and the midrange is harsh. Wasn't a problem where he previously lived. He has some absorbents panels placed at first reflection points (sidewall, ceiling and back wall), two Tri-trap bassstraps from GIK I believe and two Helmholtz Varitune V-6 resonators (tunable between 25-60 Hz). It has helped, but not enough. Anyway, here there's lay out of the room and it shows the placement he's using today. The speakers are a pair of Von Schweikert vr4 j.
Image not preserved: 1 (Large).png
welcome! please update your profile with location. is it an option to turn this around and have the speakers on the (top) wall and the seating facing that wall? it would seem the the short distance to the top (back for listener) wall for LF and MF.
gullfo wrote:
welcome! please update your profile with location. is it an option to turn this around and have the speakers on the (top) wall and the seating facing that wall? it would seem the the short distance to the top (back for listener) wall for LF and MF.
Updated! Everything is an option here. Even set up a wall to change the length of the room. He's tried the position you're mentioning and said it sounded very thin and lacking bass. I'm wondering if there's something with the dimensions of the room that are causing really bad standing waves/cancellations.
i think it's time to shift things around to troubleshoot and see if its equipment, position, or room issues. perhaps put a single speaker in the corner and do some room measurements - REW/FuzzMeasure and/or ears to see if you can spot obvious nulls. check all the equipment settings... and finally adjust the positions of speakers and seating to find the best location. if the room measures poorly enough it may be time to change it but it looks like a pretty big room so volume wise it would seem to be less of an issue.
"lot of different placements, also the 38% rule. " You need to visit http://www.hometheatershack.com/ A control room monitor placement suggested starting position that Mr. Lachot uses has no place in a listening environment. Your speakers should be at the wall, with the listening position early reflections being controlled acoustically at the chair position. It really is a matter of spatial enhancement.
I agree with Brien and Glenn: I'd put the speakers right at the top wall (up against it), facing down the length of the room, and set the sofa back far enough that you get good imaging. You also need deep bass traps in all the corners, and deep absorption on all the first reflection points Including the ceiling), as well as on the back wall (bottom of your sketch). There is a problem with the room dimensions as well: Right now it fails two out of the three BBC tests for a listening room, and the dimensions are not close to any known good ratio. You COULD change the wall positions to something like L=6.5, W=4.4, which passes all three tests and gets close to the IEC 60268-13 recommendation for listening rooms, but before even thinking about that I'd suggest taking a look a the construction of the room itself. What are the walls, floor and ceiling made of? I would also suggest running a full acoustic analysis of that room, with the layout suggested above, and posting the result here to see how closely it matches the predicted behavior of that room. You say that you have tunable traps in there, for the range 25-60 Hz, and that makes sense because your first five modes in that room are roughly 21, 36, 42, 56, and 63 Hz., (there's actually a double at 42). However, you can't solve those with tunable Helmholtz resonators alone: they are isolated, first order axials, with nothing else around them. It would be better to go with deep broad-band absorption covering the entire low end, all the way up to 96 Hz, which is the zone where the modal behavior of that room is dominant. However in doing that you'd also have to be careful to not kill the high end with too much absorption. Can you post photos of the room, so we can see what it looks like and what you have in there already, as well as details of the room construction? It might well be that the room materials are doing unpleasant things to the sound. - Stuart -
Dealing with the modes with porous material? Wouldn't that be impossible without using several meters of rockwool?
Dealing with the modes with porous material? Wouldn't that be impossible without using several meters of rockwool?
No, that's a common myth but it simply isn't true, for several reasons. First, according to the "theoretical-mythical" case, you need 4.5 m because you assume that sound waves only ever approach the treatment exactly head on, and that you want 100% absorption of the 20 Hz. quarter wave and all others as well. In reality, none of those is true for the real world. Only a very small percentage of the waves are actually incident at 90° to the treatment. Practically all waves are at far lesser angles, meaning that they go through the material at an angle, and are therefore exposed to a much longer path through it than if they had hit head-on. It is the path length that counts, not thickness of the material. Axial modes might hit head-on more frequently, but bass traps are angled across the corners for this reason. Tangential and oblique modes are pretty random, and almost always at very high angles. So even a very thin piece of material is good enough, under most circumstances. 10% of the wavelength is more than enough, 5% is good for most cases, and even 2% or 3% has considerable effect. I think it was Eric Desart who pointed this out, perhaps on another forum. So even 8" of material is good enough to get down fairly deep. Second, you don't want it right up against the wall: that is not very efficient, since at the air-wall boundary the wave has zero speed anyway, and fibrous material acts on the velocity component of the wave, not the pressure component. (You can't stop a wave at the point where it already isn't moving anyway!) So it is better to move the material away from the wall, where the velocity component is higher: Once again. mythical theory says that you have to place your material exactly one quarter wave away from the wall. Once again, reality says that this is only the case for waves incident at 90°. Since the vast majority of waves are incident at much lower angles, they "see" their wavelength much closer to the wall. So placing 8" of good absorption (EG, 703 or 705) mounted 16" off the wall, will give you pretty good coverage down to very low frequencies. That is less than half a meter of treatment. You do not need "several meters" of material to do this. You can if you want, but it simply is not necessary. Third, the "5 meters thick against the wall" myth also assumes that you need to absorb 100% of the wave the first time it hits. Not only is that impossible, it isn't even desirable! Even if you had a room big enough that you could put 4.5m over all the walls, ceilings and floors, and "theoretically" absorb 100% ( :shock: ) the room would sound so terribly disgustingly dead as to be useless! :!: It would be VERY unpleasant. Go sit in an anechoic chamber for a few minutes, to see just how unpleasant it sounds. Or wrap your head with several pillows on all sides. Same effect, roughly. Not nice. In fact, you do not want to absorb 100% of any wavelength on the first bounce. You WANT the room to have some character, some liveliness, even though you do want it to be neutral and fairly flat response. You WANT to have a diffuse field as well as a direct field. you just want it to be controlled. That's the key. The goal is not to murder the room in cold blood, or even to slowly strangle it in agonizing torture: you just want to tame it, so that it behaves well. You do not want to beat it to a pulp: just get it in line, and let it know who is the boss. :) (Excuse the dumb illustration, but I think you get the point). In other words, "large deep bass traps" in the vertical corners can be simply 6 or 8 inches of 703 running 36 inches diagonally across the corners, and it will do wonders. That is a reasonably large surface area, plenty deep enough. A few large panels of 8 inch thickness of 703 held 12 to 16 inches off the rear wall will also do wonders. And for even better effect you can angle it or tilt it, so that for example it starts 4 inches away at the bottom and goes out to 24 inches at the top. Or is 18 inches away in the center, angling to6 inches away at the edges. The exact measurements are not critical, since this is broadband bass trapping, which is the best kind for an unruly room that refuse to behave. There is another design of bass trap. Called a "superchunk" which is even more effective. Just cut large triangles of 703, at least 36" across the long side, and stack them in all the vertical corners, floor to ceiling. That is even more effective than just diagonal panels. But I'm not convinced that the problems you are describing are caused only by the modal activity due to the dimensions of the room, and the lack of sufficient trapping. I still suspect that there might be something in the construction itself that is sapping energy at some frequencies, and/or reinforcing it at others. The only real way to tell is to use REW to do a full analysis on that room, with the speakers set up as described previously, and the mic in the listening position that sounds best in that room. If you can post the graphs from that, plus photos of the room, then we can try to figure it out and offer better advice. - Stuart -
Soundman2020 wrote:
Dealing with the modes with porous material? Wouldn't that be impossible without using several meters of rockwool?
No, that's a common myth but it simply isn't true, for several reasons. First, according to the "theoretical-mythical" case, you need 4.5 m because you assume that sound waves only ever approach the treatment exactly head on, and that you want 100% absorption of the 20 Hz. quarter wave and all others as well. In reality, none of those is true for the real world. Only a very small percentage of the waves are actually incident at 90° to the treatment. Practically all waves are at far lesser angles, meaning that they go through the material at an angle, and are therefore exposed to a much longer path through it than if they had hit head-on. It is the path length that counts, not thickness of the material. Axial modes might hit head-on more frequently, but bass traps are angled across the corners for this reason. Tangential and oblique modes are pretty random, and almost always at very high angles. So even a very thin piece of material is good enough, under most circumstances. 10% of the wavelength is more than enough, 5% is good for most cases, and even 2% or 3% has considerable effect. I think it was Eric Desart who pointed this out, perhaps on another forum. So even 8" of material is good enough to get down fairly deep.
I sort of knew that, but needed the reminder. Very informative!
Soundman2020 wrote:
Second, you don't want it right up against the wall: that is not very efficient, since at the air-wall boundary the wave has zero speed anyway, and fibrous material acts on the velocity component of the wave, not the pressure component. (You can't stop a wave at the point where it already isn't moving anyway!) So it is better to move the material away from the wall, where the velocity component is higher: Once again. mythical theory says that you have to place your material exactly one quarter wave away from the wall. Once again, reality says that this is only the case for waves incident at 90°. Since the vast majority of waves are incident at much lower angles, they "see" their wavelength much closer to the wall. So placing 8" of good absorption (EG, 703 or 705) mounted 16" off the wall, will give you pretty good coverage down to very low frequencies. That is less than half a meter of treatment. You do not need "several meters" of material to do this. You can if you want, but it simply is not necessary.
Yes. A challenge here though is the fact that we are talking about a living room. It doesn't look very nice to place absorbents that far out. But it will be considered. Buying a bunch of RPG Modex Plates would probably do the trick, but they are not cheap. I have question. Isn't the best aircavity for the porous material the same as the thickness of it? You're mentioning spacing a 8" panel 16" off the wall. How does that compare to mounting ut 8" off the wall?
Soundman2020 wrote:
Third, the "5 meters thick against the wall" myth also assumes that you need to absorb 100% of the wave the first time it hits. Not only is that impossible, it isn't even desirable! Even if you had a room big enough that you could put 4.5m over all the walls, ceilings and floors, and "theoretically" absorb 100% ( :shock: ) the room would sound so terribly disgustingly dead as to be useless! :!: It would be VERY unpleasant. Go sit in an anechoic chamber for a few minutes, to see just how unpleasant it sounds. Or wrap your head with several pillows on all sides. Same effect, roughly. Not nice. In fact, you do not want to absorb 100% of any wavelength on the first bounce. You WANT the room to have some character, some liveliness, even though you do want it to be neutral and fairly flat response. You WANT to have a diffuse field as well as a direct field. you just want it to be controlled. That's the key. The goal is not to murder the room in cold blood, or even to slowly strangle it in agonizing torture: you just want to tame it, so that it behaves well. You do not want to beat it to a pulp: just get it in line, and let it know who is the boss. :) (Excuse the dumb illustration, but I think you get the point).
Funny you should mention that. I've just been discussing this with a guy. I'm saying the same as you're, but he's saying a room with no reflections is better as long as the decay is even at all frequencies. His point is that music outdoor sounds good. And he will only agree that listening in an anechoic space sounds unatural if the are studies that indicate that listening outdoor is claustrofobic. Any comment to that?
Soundman2020 wrote:
In other words, "large deep bass traps" in the vertical corners can be simply 6 or 8 inches of 703 running 36 inches diagonally across the corners, and it will do wonders. That is a reasonably large surface area, plenty deep enough. A few large panels of 8 inch thickness of 703 held 12 to 16 inches off the rear wall will also do wonders. And for even better effect you can angle it or tilt it, so that for example it starts 4 inches away at the bottom and goes out to 24 inches at the top. Or is 18 inches away in the center, angling to6 inches away at the edges. The exact measurements are not critical, since this is broadband bass trapping, which is the best kind for an unruly room that refuse to behave. There is another design of bass trap. Called a "superchunk" which is even more effective. Just cut large triangles of 703, at least 36" across the long side, and stack them in all the vertical corners, floor to ceiling. That is even more effective than just diagonal panels. But I'm not convinced that the problems you are describing are caused only by the modal activity due to the dimensions of the room, and the lack of sufficient trapping. I still suspect that there might be something in the construction itself that is sapping energy at some frequencies, and/or reinforcing it at others. The only real way to tell is to use REW to do a full analysis on that room, with the speakers set up as described previously, and the mic in the listening position that sounds best in that room. If you can post the graphs from that, plus photos of the room, then we can try to figure it out and offer better advice. - Stuart -
I have not been to this room yet that we are talking about. But I have a friend who's been there and did some measurements. I've just talked briefly to him, but he believed something was wrong with the construction of the walls and that they were dampening/stealing too much bass. If that's case, what's the solution? I'm a bit hesitant to share pictures with the room before I ask permission from the owner. I'm going to visit him probably next week and I plan to bring several basstraps and absorbents. Unfortunately I don't have a mobile soundcard at the moment. There is so much good info here that I would really like to start a thread to ask for aid with my own listening room. I also need improvements. I can give you both pictures and measurements. It's an incredible difficult room and I'm not sure it's possible to get an even response without using several subwoofers/midbasses placed randomly. I will share the link for a new thread to keep it separate from this one. Really appreciate the help and tips!