Just checking: So the highest level of 103 dB occurs at that frequency of 138 Hz?Lowest loud frequency to isolate is 138hz (saxophone). Maximum possible volume is 103db. Normal playing 74+db
Please refresh my memory: I'm sure it's in the thread somewhere, but no time to read through it all again... That's what you actually measured with a meter? That you are no getting 50 dB of isolation? And I assume that you calculated the resonant frequency, rather then measure it?The brick cavity wall is giving us attenuation of about 50db. F = 26.83
It's better to do all your measurement son "C" weighting, as it more closely models the way the human ear perceives loud sound. "A" weighting is less sensitive to low frequencies, so a sound that hardly even moves the meter needle on "A" at all can still be heard clearly and will show up well on "C".Ambient noise at neighbour's house is about 34dba, when playing this goes up to maximum 47.8dba (52dbc).
If your goal is to get then 103 down to 34, then yes... well, sort of.... hmmm... OK, it's not so simple.... Think of it this way: if you do manage to get the 103 down so that it sounds like 34 on the other side of the wall, well then the TOTAL sound level on the other side of the wall is 34 dB + 34 dB = 37 dB. That's your 34 plus the 34 ambient that was already there. So your sax is sounding 3 dB louder than ambient. As it turns out, the human ear can still pick out individual sounds that are at or even below the ambient level: just like you can pick out a specific conversation happening across the room at a cocktail party, or pick out one instrument in the orchestra even though the orchestra as a whole is louder than that one single instrument. Our brains are good at that. So just getting a specific sound down to the same level as ambient does not guarantee that it will not be heard at all. It just means that it will be heard very, very quietly, as part of the background noise.The brick wall therefore is doing 50-55db attenuation. Does this mean we need 14-18db more attenuation? Just checking.
So, in other words, if possible it is good to aim for isolation a bit beyond what the numbers show by themselves, if you are concerned that your noise might still disturb your neighbors.
For a three leaf wall, there are two numbers because there are two fundamental resonant frequencies. Those are called "F-" (for the lower of the two) and "F+" for the higher one. With a two leaf wall there is only one fundamental, and that is called simply "F0".2. Air gap size relation to f+.
Soundman 2020 - when you were given F numbers of 17.2 (stud wall 250mm gap) and 26.83 (brick wall) said as it was more a 3 leaf wall, around f+ = 30 was more likely. How did you calculate this f+ number?
For a 3-leaf, F- is given by: F+ / SQRT (2)
This is probably more than you really wanted to get in to, but the full equation for resonance in a 3-leaf system is:If the two F numbers were 22.57 (stud wall 144mm gap) and 26.83 (brick wall) what would the f+ number be (approximately)?
This question is because 250mm was a number picked out of thin air and 144mm is a large timber stud size here in the UK. We need to know whether a single 144mm stud is sufficient or if a wider gap is needed.
F+ = (PI/2) * SQRT( ( 3.6 * Air_Density * (Speed_Of_Sound^2) ) / (((M1 + M2 + M3)/3)*(D1 + D2)) )
M1=mass of first leaf (kg/m^2)
M2 mass of second leaf (kg/m^2)
M2 mass of third leaf (kg/m^2)
D1=spacing between first and second leaves (m)
D2=spacing between second and third leaves (m)
And F- is as above.
But as I said, that's probably way more than you really need to be getting in to, and at the same time it is also not enough: After you figure out F+ and F-, (or just F0, if two-leaf), then you also need to figure out the approximate isolation from mass law alone at F+ * 1.414, and project isolation rising by 18 dB per octave from that point until the coincidence dip, then about 12 dB per octave from there on up. Then you need to weight your resulting curve by applying the "C" weighting curve to it, to give you the actual expected isolation from the wall, roughly as you will perceive it.
It's probably a lot easier to just browse through some of the documents that show the results of actual walls tested in acoustic laboratories, and see what makes the most sense for you.
That's the problem! There's no simple formula for predicting single-number isolation values. It's quite a bit more complicated than that. I've been trying to develop a spreadsheet to do it, for the past couple of years, and it's getting reasonable results, but still nowhere near ready for prime time.3. Frequency/isolation/decibels
If F (or f+) is say 30, how much isolation in decibels is there for a noise at 60hz? At 120hz? Or what is the formula?
That depends on how you do your wall and ceiling framing. If the framing of your new inner-leaf walls will touch the existing brick walls, or the existing ceiling above you, at any point, then you should use resilient channel. If the framing is entirely separate, and touches nothing but your floor, then you don't need the resilient channel.4. Resilient bars and frequency
Should we use resilient bars to seperate the the timber studs and the plasterboard? I assumed so but read in the FAQ thread that for low frequencies not to use them. What counts as low? Bear in mind it is mostly 138hz+ to isolate.
You did mention that the cavity is uninsulated, so that would explain at least some of the "missing" isolation. Lack of damping in the cavity can cost you anywhere from 5 dB to 15 dB, give or take. And there's also the issue that you are not really measuring the isolation of "just the wall": rather, you are measuring the overall isolation between the rooms, which includes walls, floors, ceilings, windows, doors, flanking paths, ventilation paths, electrical conduit, and a few other things too. So the wall itself is probably performing better than you are measuring, and the difference is due to all the other paths.Actual transmission loss achieved is 50-55db (68-75% of the theoretical result)
My still wobbly spreadsheet predicts a realistic value of about 72 dB.Achievable transmission loss ??? At 68-75% (from double panel results) = 74.51-82.19db
It is a lot, but that's the magic of MSM resonance.Is this correct? A 19+db increase seems like a lot.
That said, 72 dB seems to be on the high side, since you are very likely around the "flanking limit" for the entire building at that sort of level. There's a limit to what you can achieve realistically in any building, simply because at some point the very small amount of vibration that has gotten into the building structure itself starts to be significant, notably the floor. Flanking is always there, and there's not much you can do about it, but at low levels of isolation it makes no difference at all, since you are comparing something the size of a grain of sand to something the size of a bowling ball. But when you get the bowling ball out the way, and move down to baseballs, ping-pong balls and then marbles (figuratively speaking!), as you get better and better isolation, the all of a sudden at the size of ball bearings, well, grains of sand no start to be very significant, and noticeable. Not sure if I'm getting the point across here! But when you compare very quiet sounds to very loud ones, you don't notice the quiet one. But when you compare a very quiet sound to one that is only a bit louder, then you do notice it. So you don't even notice the clock ticking when you have the TV on, or a jet is flying overhead, but late at night when it is the ONLY sound, then you sure do notice it, and it becomes very loud, even though it is actually at the same level as always.
I'm still not doing a good job of trying to explain, I think... Let me try again: going from 20 dB of isolation (single sheet of drywall) to 30 dB of isolation is dead easy: just build a simple stud wall with drywall on each side. Going from 30 to 40 is a bit harder: you need to decouple one side and add more mass. Going from 40 to 50 is seriously harder: decouple both sides, add more mass, and damping. Going from 50 to 60 is a Really Tall Order: Large amounts of mass, large air gaps, and very careful sealing. Going from 60 to 70 is a major undertaking: Oodles of mass, even bigger gaps, amazing attention to sealing every possible last nook and cranny, even sealing the wall surfaces. So what is left to do if you need to go from 70 to 80? There simply is nothing more that you can reasonably do, since you have hit the dreaded "law of diminishing returns". To get to the next level, you now need to start doing amazingly expensive things, like floating your entire room on acoustic isolation springs, and increasing the mass yet again. And going from 80 to 90? More of the same: even MORE massive floors, walls, ceilings, incredibly complex air vents, and we are now talking millions of dollars. The very best isolated studio in the planet measures about 100 dB, and they went to major extremes to achieve that.
Not sure if you noticed that at each stage in this progression, we only went up by 10 dB. But because dB is a log scale, each extra "ten dB" costs you exponentially more than the "ten dB" just below it. The first 10 dB step from 20 to 30 dB only cost a few tens of dollars to accomplish. The final 10 dB step, from 90 to 100, costs tens of millions of dollars.
So there's a point at which your budget and practical reality just stop you from getting any higher on the scale.
I'm really not trying to dishearten you or throw a wet blanket on your plans, but realistically I would expect that you won't be able to get much beyond 65 dB total isolation, perhaps 70 with a lot of luck, based on your current plans.
Now to put that in perspective: if you get 65 dB, then your 103 dB absolute highest peak is down to 38, which will be barely audible next door. Your average sax level of about 75 to 80 is down to just 10 or 15 on the other side: that is absolutely inaudible. It is more than 15 dB below ambient, so it won't be heard. Your piano peak at 90 is down to 25: perhaps on the very edge of being vaguely perceived on the other side, if you listen very, very carefully. And your average piano level, like your average sax level, is also totally inaudible. on the other side.
So if the above meets your needs, then realistically it is probably doable. But once again, this isn't just about the walls: you also need to get your doors, windows, ventilation and electrical systems to that same level of 65 dB. The room is a system, and everything needs to be at the same level. Sound is like water: it takes the easiest path. If you want to build a dam, it's no use just putting up a short wall in the middle of the riverbed. While it might be a wonderful wall, it won't stop the water unless you extend it across the entire river, bank to bank. Only then will it hold water. Your room is the same: you need to "extend the wall" around the entire room to get those high levels of isolation. And the higher your isolation level is, then more a tiny crack matters. That small gap under your door is a major canyon, as far as sound is concerned. Your doors and windows need just as much attention as your wall does.
- Stuart -
