In order to provide a reproducible parameter, a standard reverberation time has been defined as the time for the sound to die away to a level 60 decibels below its original level. The reverberation time can be modeled to permit an approximate calculation. . Then, My question is as follows. Where the listener/observer must be located in a enclosing room while measuring reverberation time? In contast to definite sound source, different observer location must induce different sound intensity accepted, then produce different reverberation time parameter.Then, where can be the exact location to obseve the sound intensty? It is not designed definitely in defination of reverberation time. It is the same location as source ? HyeeWang
reverberation time measurement question
Started by ●June 1, 2009
Reply by ●June 1, 20092009-06-01
"HyeeWang" <hyeewang@gmail.com> wrote in message news:04435c29-2394-4ff3-b49f-73b13a301fa9@l28g2000vba.googlegroups.com...> In order to provide a reproducible parameter, a standard reverberation > time has been defined as > the time for the sound to die away to a level 60 decibels below its > original level. The reverberation > time can be modeled to permit an approximate calculation. . > > Then, My question is as follows. > > Where the listener/observer must be located in a enclosing room while > measuring reverberation time? > > In contast to definite sound source, different observer location must > induce different sound intensity accepted, > > then produce different reverberation time parameter.Then, where can be > the exact location to obseve the sound intensty? > > It is not designed definitely in defination of reverberation time. > > It is the same location as source ? > > HyeeWangI look at reverberation time simply as an indicator, it helps people get a better or quicker mental image of the sound they're making. Listener locations may effect this, but that's a different parameter. You should download a few FX plug-ins and simply see how they represent the information yourself.
Reply by ●June 1, 20092009-06-01
HyeeWang wrote:> In order to provide a reproducible parameter, a standard reverberation > time has been defined as > the time for the sound to die away to a level 60 decibels below its > original level. The reverberation > time can be modeled to permit an approximate calculation. . > > Then, My question is as follows. > > Where the listener/observer must be located in a enclosing room while > measuring reverberation time? >Unless it is a pathologically bad space, the dense reverb ("tail") should be the same everywhere. The direct sound will be different of course, and the early reflections to a greater or lesser extent, but the RT60 measurement is not concerned with either. Richard Dobson
Reply by ●June 1, 20092009-06-01
On Jun 1, 10:58�am, Richard Dobson <richarddob...@blueyonder.co.uk> wrote: ...> > Where the listener/observer must be located �in a enclosing room while > > measuring reverberation time? > > Unless it is a pathologically bad space, the dense reverb ("tail") > should be the same everywhere.after the early reflections, the envelope of the reverb tail should be about the same everywhere. the room as a whole gets charged up with acoustic energy and energy is getting absorbed in surfaces pretty much all over in the room. at least those are the assumptions in the Sabine equation: RT60 = (c*V)/(a*S) c is the speed of sound V is the room volume a is the *mean* absorbtion coefficient of the room surfaces (i *don't* believe that this coef is directly sabines/ft^2 because there would have to be a conversion factor living in that equation, perhaps that 0.057 number. a sabine is the amount of absorbtion of 1 ft^2 of open window.) S is the inside surface are of the room. with enough reasonable assumptions, you can show that the energy in the room (and the smoothed intensity of reverberating sound) must be decreasing exponentially. the RT60 is really just proportional to the reciprocal of the "alpha" in the exponential.> The direct sound will be different of > course, and the early reflections to a greater or lesser extent, but the > RT60 measurement is not concerned with either.the RT60 is concerned with the tail. if you can use a sound editor (or even an old fashioned storage scope with a logarithmic rectifier), you can get the negative slope of the log decaying envelope and the RT60 is the reciprocal of that (with the appropriate scaling factor). r b-j
Reply by ●June 1, 20092009-06-01
On Jun 1, 11:44�am, robert bristow-johnson <r...@audioimagination.com> wrote:> On Jun 1, 10:58�am, Richard Dobson <richarddob...@blueyonder.co.uk> > wrote: > > ... > > > > Where the listener/observer must be located �in a enclosing room while > > > measuring reverberation time? > > > Unless it is a pathologically bad space, the dense reverb ("tail") > > should be the same everywhere. > > after the early reflections, the envelope of the reverb tail should be > about the same everywhere. �the room as a whole gets charged up with > acoustic energy and energy is getting absorbed in surfaces pretty much > all over in the room. �at least those are the assumptions in the > Sabine equation: > > � �RT60 = (c*V)/(a*S) > > c is the speed of sound > V is the room volume > a is the *mean* absorbtion coefficient of the room surfaces > � (i *don't* believe that this coef is directly sabines/ft^2 because > � �there would have to be a conversion factor living in that > � �equation, perhaps that 0.057 number. �a sabine is the amount > � �of absorbtion of 1 ft^2 of open window.) > S is the inside surface are of the room. > > with enough reasonable assumptions, you can show that the energy in > the room (and the smoothed intensity of reverberating sound) must be > decreasing exponentially. �the RT60 is really just proportional to the > reciprocal of the "alpha" in the exponential. > > > The direct sound will be different of > > course, and the early reflections to a greater or lesser extent, but the > > RT60 measurement is not concerned with either. > > the RT60 is concerned with the tail. > > if you can use a sound editor (or even an old fashioned storage scope > with a logarithmic rectifier), you can get the negative slope of the > log decaying envelope and the RT60 is the reciprocal of that (with the > appropriate scaling factor). > > r b-jSmall correction -> c needs to be the inverse of the speed of sound if the units are to work out. I have never heard of the Sabine equation - just heard of Sabine Women. Clay
Reply by ●June 2, 20092009-06-02
On Jun 2, 3:31�am, c...@claysturner.com wrote:> On Jun 1, 11:44�am, robert bristow-johnson <r...@audioimagination.com> > wrote: > > > > > > > On Jun 1, 10:58�am, Richard Dobson <richarddob...@blueyonder.co.uk> > > wrote: > > > ... > > > > > Where the listener/observer must be located �in a enclosing room while > > > > measuring reverberation time? > > > > Unless it is a pathologically bad space, the dense reverb ("tail") > > > should be the same everywhere. > > > after the early reflections, the envelope of the reverb tail should be > > about the same everywhere. �the room as a whole gets charged up with > > acoustic energy and energy is getting absorbed in surfaces pretty much > > all over in the room. �at least those are the assumptions in the > > Sabine equation: > > > � �RT60 = (c*V)/(a*S) > > > c is the speed of sound > > V is the room volume > > a is the *mean* absorbtion coefficient of the room surfaces > > � (i *don't* believe that this coef is directly sabines/ft^2 because > > � �there would have to be a conversion factor living in that > > � �equation, perhaps that 0.057 number. �a sabine is the amount > > � �of absorbtion of 1 ft^2 of open window.) > > S is the inside surface are of the room. > > > with enough reasonable assumptions, you can show that the energy in > > the room (and the smoothed intensity of reverberating sound) must be > > decreasing exponentially. �the RT60 is really just proportional to the > > reciprocal of the "alpha" in the exponential. > > > > The direct sound will be different of > > > course, and the early reflections to a greater or lesser extent, but the > > > RT60 measurement is not concerned with either. > > > the RT60 is concerned with the tail. > > > if you can use a sound editor (or even an old fashioned storage scope > > with a logarithmic rectifier), you can get the negative slope of the > > log decaying envelope and the RT60 is the reciprocal of that (with the > > appropriate scaling factor). > > > r b-j > > Small correction -> c needs to be the inverse of the speed of sound if > the units are to work out. I have never heard of the Sabine equation - > just heard of Sabine Women. > > Clay- Hide quoted text - > > - Show quoted text -Sabine equation is a empirical formula, which facility the computation of reverberation time. But it disregard the topic, that is my question. And I wanna know the essence of reverberation time . So I disregard Sabine equation,behaveing myself as a learner. I would describe my question again by a exapmle. Let us suppose a sound source, 2 listeners, A and B, located in a room somewhere differently. The sound source is in 200db energy and stop suddenly. Differnet location should accept the sound from source in different intensity.So at the exact time point that sound source disappear, say ,listener A accept sound in 180db, in contrsat to 160db with listener B simultaneously. The time period used by A and B to drop 60db ,that is 180db - 120 db for A and 160db - 100b for B, must be different. Then, which period is the right reverberation time RT60? According to definition of reverberation time -- the time for the sound to die away to a level 60 decibels below its original level, which is original level? It is 200db,180db or 160db? Cheers HyeeWang
Reply by ●June 2, 20092009-06-02
On Jun 1, 3:31�pm, c...@claysturner.com wrote:> On Jun 1, 11:44�am, robert bristow-johnson <r...@audioimagination.com> > wrote: > > > > � �RT60 = (c*V)/(a*S) > > > c is the speed of sound > > V is the room volume > > a is the *mean* absorbtion coefficient of the room surfaces > > � (i *don't* believe that this coef is directly sabines/ft^2 because > > � �there would have to be a conversion factor living in that > > � �equation, perhaps that 0.057 number. �a sabine is the amount > > � �of absorbtion of 1 ft^2 of open window.) > > S is the inside surface are of the room. >...> > Small correction -> c needs to be the inverse of the speed of sound if > the units are to work out.that'll teach me from going to Wikipedia for reliable information. a better rendition of it is at http://www.tonmeister.ca/main/textbook/node278.html . that 55.26 factor is 4*ln(10^(60/10)).> I have never heard of the Sabine equation - > just heard of Sabine Women.dunno who Sabine Women are. but you *have* heard of the acoustician named Wallace Clement Sabine, no? anyway, a couple decades ago, i rederived this two different ways and the two derivations resulted in very similar, but not exactly the same results. it throws together a lot of assumptions. one is that sound is flying around the room in all directions. the amount of Intensity of sound impinging upon a little differential volume is proportional to the solid angle from whence it comes. so the measure of mean sound intensity in the room is not watts/m^2 but (watts/m^2)/steradian. if you know that, you know how much energy there is per cubic volume everywhere. then, assuming all of the room surfaces are homogeneous, then you can calculate how much energy is being absorbed per unit time per square meter of room surface. from that you can get a 1st-order diff eq for the mean energy per unit volume or mean sound intensity (per steradian). there's another method involving figgering out the mean free path which is the reciprocal of the mean number of reflections per meter of sound travel. r b-j
Reply by ●June 2, 20092009-06-02
On Jun 1, 11:08�pm, HyeeWang <hyeew...@gmail.com> wrote:>...> Sabine equation is a empirical formula, which facility the computation > of reverberation time.not entirely. given the assumption (and it *is* an assumption) of acoustic homogeneity of the volume of the room (every cubic meter has the same energy density as the others) and of the surface of the room (every square meter of room surface has the same absorbtion as any other square meter), there is a solid theoretical basis for it. and, what's cool, if the room is big enough and diffuse enough, it fits well with experiment.> I would describe my question again by a exapmle. > Let us suppose a sound source, 2 listeners, A and B, located �in a > room somewhere differently. The sound source is in 200db energy and > stop suddenly. > > Differnet location should accept the sound from source in different > intensity.but it's the rate of decay that RT60 is an associated parameter.> So at the exact time point that sound source disappear, > say ,listener A accept sound in 180db, in contrsat to 160db > with listener B simultaneously. > > The time period �used by A and B to drop 60db ,that is 180db - 120 db > for A and 160db - 100b for B, �must be different.why? in a homogeneous room? do you need for me to dig out how this gets derived for a rectangular room with consistent homogeneous absorbtion with the walls and show you? (lotsa assumptions in that, but it turns out that neither the volume nor surface area care precisely what the X, Y, and Z dimensions of the room are, only the products V=XYZ and sum of products S=2(XY+YZ +ZX)). that can be derived. r b-j
Reply by ●June 2, 20092009-06-02
On Jun 1, 11:34�pm, robert bristow-johnson <r...@audioimagination.com> wrote:> On Jun 1, 3:31�pm, c...@claysturner.com wrote: > > > > > > > On Jun 1, 11:44�am, robert bristow-johnson <r...@audioimagination.com> > > wrote: > > > > � �RT60 = (c*V)/(a*S) > > > > c is the speed of sound > > > V is the room volume > > > a is the *mean* absorbtion coefficient of the room surfaces > > > � (i *don't* believe that this coef is directly sabines/ft^2 because > > > � �there would have to be a conversion factor living in that > > > � �equation, perhaps that 0.057 number. �a sabine is the amount > > > � �of absorbtion of 1 ft^2 of open window.) > > > S is the inside surface are of the room. > > ... > > > Small correction -> c needs to be the inverse of the speed of sound if > > the units are to work out. > > that'll teach me from going to Wikipedia for reliable information. �a > better rendition of it is athttp://www.tonmeister.ca/main/textbook/node278.html > . that 55.26 factor is 4*ln(10^(60/10)). > > > I have never heard of the Sabine equation - > > just heard of Sabine Women. > > dunno who Sabine Women are. �but you *have* heard of the acoustician > named Wallace Clement Sabine, no? > > anyway, a couple decades ago, i rederived this two different ways and > the two derivations resulted in very similar, but not exactly the same > results. > > it throws together a lot of assumptions. �one is that sound is flying > around the room in all directions. �the amount of Intensity of sound > impinging upon a little differential volume is proportional to the > solid angle from whence it comes. �so the measure of mean sound > intensity in the room is not watts/m^2 but (watts/m^2)/steradian. �if > you know that, you know how much energy there is per cubic volume > everywhere. �then, assuming all of the room surfaces are homogeneous, > then you can calculate how much energy is being absorbed per unit time > per square meter of room surface. �from that you can get a 1st-order > diff eq for the mean energy per unit volume or mean sound intensity > (per steradian). �there's another method involving figgering out the > mean free path which is the reciprocal of the mean number of > reflections per meter of sound travel. > > r b-j- Hide quoted text - > > - Show quoted text -Hello Robert, There is a little about Sabine Women here: http://en.wikipedia.org/wiki/Sabines I have to plead my ignorance about Wallace Sabine. I'll go study him. I do however have Hemholtz's book. The assumptions about the energy being spread all throughout the volume with absorption causing exponential decay (caused by rate of absorption being proportional to the amount of energy) is a common one even outside of acoustics. An example that almost everyone has seen is phosporescence. The initial "charging light" is like the initial sound in the echo chamber and then the decaying afterglow is like the decaying reverberation. Of course the mechanisms behind the two case are somewhat different with the optical case using quantum mechanical forbidden transitions to delay the release of the energy. Clay
Reply by ●June 2, 20092009-06-02
On Jun 2, 12:51�pm, robert bristow-johnson <r...@audioimagination.com> wrote:> On Jun 1, 11:08�pm, HyeeWang <hyeew...@gmail.com> wrote: > > > > ... > > Sabine equation is a empirical formula, which facility the computation > > of reverberation time. > > not entirely. �given the assumption (and it *is* an assumption) of > acoustic homogeneity of the volume of the room (every cubic meter has > the same energy density as the others) and of the surface of the room > (every square meter of room surface has the same absorbtion as any > other square meter), there is a solid theoretical basis for it. �and, > what's cool, if the room is big enough and diffuse enough, it fits > well with experiment. > > > I would describe my question again by a exapmle. > > Let us suppose a sound source, 2 listeners, A and B, located �in a > > room somewhere differently. The sound source is in 200db energy and > > stop suddenly. > > > Differnet location should accept the sound from source in different > > intensity. > > but it's the rate of decay that RT60 is an associated parameter. > > > �So at the exact time point that sound source disappear, > > say ,listener A accept sound in 180db, in contrsat to 160db > > with listener B simultaneously. > > > The time period �used by A and B to drop 60db ,that is 180db - 120 db > > for A and 160db - 100b for B, �must be different. > > why? �in a homogeneous room? > > do you need for me to dig out how this gets derived for a rectangular > room with consistent homogeneous absorbtion with the walls and show > you? �(lotsa assumptions in that, but it turns out that neither the > volume nor surface area care precisely what the X, Y, and Z dimensions > of the room are, only the products V=XYZ and sum of products S=2(XY+YZ > +ZX)). �that can be derived. > > r b-jThank you all. I have learned that in an ideal chamber (homogeneous material decorated), the paramter RT60 is exactly the same. It is measured in a relative DB scale, so it has nothing to do with the location observer located. The relative measurment is of course at the same location. Another questions about RT60 are here. 1. In a rectangle ,not sphere, room, the reverberation time at corner and center of that room are the same? Why? 2. ---after the early reflections, the envelope of the reverb tail should be about the same everywhere. Here , why stress the assertion " after the early reflections" ? HyeeWang
