How loud, in SPL, is fff from a large orchestra at front row distance?

[454Casull]

I'm thinking somewhere between 100dB and 110dB... anybody with concrete information?

[Dave Jones]

Split the difference. It's about 105 dB, give or take.

[454Casull]

Is that a guess?

[GM]

Here's an old post of mine:


Somewhere back in the mists of time someone measured the peak power
distribution of a symphony. Don't know who, but my copy is in "How To Build
Speaker Enclosures" / Alexis Badmaieff and Don Davis / 1966. What's not
listed is at what distance these peaks were measured. My own half assed
measurements with an RS SPL meter at the ASO indicates at the front row.

below 63hz = 1.5Acoustic Watt (121.76dB)
63-125hz = 2AW (123dB)
125-250hz = 4AW (126dB)
250-500hz = 10AW (130dB)
500-1khz = 2AW (123dB)
1-2khz = 1.6AW (122dB)
2-4khz = 1AW (120dB)
above 4khz = 0.6AW (117.78dB)

Loud for sure, although these peaks are usually 24-30dB above the average
SPL and as much as 60dB louder than quiet passages. The 10AW in the
250-500hz range in particular implies some seriously loud music, but anyone
who's watched peak level meters knows that these are transients that occur
so rapidly you don't notice much increase in sound level unless it's
sustained for some period of time, dependent on frequency.

GM

[phase_accurate]

GM

Good work, but I have one question:

Aren't your acoustic-power/dB[SPL] relationships off by 10 dB ? AFIK one acoustic watt is 112 dB re 20 up for a point-source radiating into half-space, measured at a distance of 1 meter.

It is doubtful if it is at all possible to get the radiated power from a measurement of SPL in one point of space, given the size and the complex radiation characteristics of an orchestra.

Regards

Charles

[ThorstenL]

Konnichiwa,

Originally posted by 454Casull
I'm thinking somewhere between 100dB and 110dB... anybody with concrete information?

I measured (using a meter measuring "slow averages" from the third row at the London Festival Hall (excellent sounding hall) during one of the bigger warhorses, the Mussorsky/Ravel "Pictures at an Exhibition". The quietest it ever got was around 35db average, the loudest it got was around 95db. However, these where average levels, knowing the ballistics of the meter and symphonic music I would expect peaks to be between 15 & 20db higher, during the tuttis, so 110-115db on the higest peaks seem reasonable.

I found that listening at home I am usually naturally around the same sort of average level for the tuttis, if I can turn up the wick.

Sayonara

[GM]

>Aren't your acoustic-power/dB[SPL] relationships off by 10 dB ? AFIK one acoustic watt is 112 dB re 20 up for a point-source radiating into half-space, measured at a distance of 1 meter.

====
Hmm, nobody is going to accuse me of having any math smarts, but between posting this many times on various forums over the years without anyone correcting me and also seeing speaker system design pros convert AW to SPL (or SPL to AW) and getting the same values I do, I have to assume that I didn't misunderstand what's published on the subject in Sound System Engineering/Don and Carolyn Davis (and that it too is correct):

"total acoustic power can also be expressed as a level dB-PWL (Lw) = 10*log(total acoustic watts/10^-12W)", ergo 1AW = 120dB.

This is derived from a particle density at the surface of an omnidirectional point source with a 1m^2 surface area, which as you note can be translated into ~112dB/1w/1m/half space. Since the orchestra's SPL chart is rated in AW (total acoustic power) at some unspecified point in space of a large venue though, distance and directivity is already accounted for, so if 1AW is defined as 120dB, a measured 10AW at 'x' distance = 130dB/'x' distance, etc.. If we had all the info to work backwards, we could calculate whatever the acoustic power a large orchestra could theoretically generate at the surface of an omnidirectional point source with a 1m^2 surface area.

IOW, it's my understanding that 1AW = 120dB at the surface of an omnidirectional point source with a 1m^2 surface area is only a reference, not an absolute that can't be exceeded. Indeed, at one time it was dB-PWL (Lw) = 10*log(total acoustic watts/10^-13W)", ergo 1AW = 130dB.

Of course none of this means that me or the folks I've compared 'notes' with are right, but my SLM (126dB max scale) pegged several times the one time I tried to get an idea of how this chart was derived, so while I know that the readings I got weren't all that accurate due to using an uncalibrated meter, I don't doubt the calculated peak SPLs. As always though, YMMV.
====
>It is doubtful if it is at all possible to get the radiated power from a measurement of SPL in one point of space, given the size and the complex radiation characteristics of an orchestra.
====
Meters measure the pressure it 'feels', and it's 'feeling' a summed acoustic power of all the direct and reflected sound at that position, so I guess I miss your point. I mean, if you put the mike at different positions in the soundfield, you will get a different SPL chart for each one due to the attenuation over distance and differing reflection patterns.

GM

[Svante]

Maybe time for some room acoustics?

In a room the sound field can be separated in two parts, the direct field and the reverberant field. The direct field follows the distance law ie the level drops by 6 dB per doubling of distance. The level of the reverberant field is essentially constant over the entire room, except for some random fluctuations and close to the walls. At a certain distance from the (point) source these two levels are equal. This distance is called the reverberation radius. Typically, this distance is some 2-5 metres for a concert hall, depending on size, reverberance and frequency. This means that beyond this distance, the level is about the same. The front row will usually be hit by about the same level as the back row.

So, the level of the reverberant field can be estimated by calculating the intensity as it would have been at the reverberation radius under free field conditions. This intensity can then be converted to a level.

For example, if the orchestra manages to produce 10 watts of acoustic power, and the reverberation radius is 4 metres, the surface of a sphere with radius 4 metres is 4*pi*4^2=201 m2. The intensity is then 10/201=0.05 W/m2 and the level is 10*log(0.05/10^-12)=107 dB.

So the acoustic power of the orchestra and the reverberation radius of the hall determines the level in the audience. I think the numbers above are reasonable, possibly 10 watts is a too high number, but it is at least somewhere there.

HTH

[Dave Jones]

Originally posted by 454Casull
Is that a guess?

I read it somewhere, but I don't remember where.

D.