Why do you think there are safety workplace laws regarding excessive noise? Because anything above 85dB Leq-8 (A-Weighted) causes long term hearing damage! Get your facts right.
According to Australian Law:
Part 10 of the Workplace Health and Safety Regulation 1997 states that the employer must prevent risks to the health and safety of workers from exposure to excessive noise at work. Under the regulations, "excessive noise" is a level of noise above:
* an 8 hour equivalent continuous A-weighted sound pressure level of 85dB(A), referenced to 20 micropascals
* a (C)-weighted sound pressure level of 140dB(C), referenced to 20 micropascals
That means that peak noise at or above 140dB(C) is truely damaging, and continuous equivalent A-weighted SPL at or above 85dB(A) is also damaging.
(Note that the A and C weightings have to do with the position of the persons ears with respect to the noise, and how to add periods of noise together)
Although you might not think 85dB is damaging, if it is 85dB at the position of your ears, and you are exposed to it for more than 8 hours continuously, this will contribute to long term deafness. Noise effects are additive, ie each period of time where you have exposure to "excessive noise" (above 85dB(A) for 8 hours) are added together to produce the final A-Weighted exposure.
Every 3dB above that limit doubles the SPL, so you half the acceptable time limit:
8h @ 85dB
4h @ 88dB
2h @ 91dB
1h @ 94dB
30min @ 97dB
15min @ 100dB
7.5min@ 103dB
3.75min@106dB
1.88min@109dB
0.94min@112dB
0.47min@115dB
0.23min@118dB
7.0sec @121dB
3.5sec @124dB
1.8sec @127dB
0.9sec @130dB
0.4sec @133dB
0.2sec @136dB
0.1sec @139db
So no wonder 140dB is dangerous at peak levels, since you only need <0.1 seconds of exposure to contribute to long term deafness.
So to put it lightly, 130dB from speakers in an enclosed space like an apartment is absolutely rediculous, as probably around 120dB(A) will reach your ears. So not a good idea to listen to for more than a few seconds at a time unless you like the idea of becoming deaf in your old age.
In actual fact, going to a club where loud music is always playing at around 100dB is VERY damaging to your ears regardless of the frequency. Every time you come home with ringing in your ears, or temporary deafness from a powertool, is doing serious damage to your hearing! The human ear is a very sensitive device and was never intended to be subjected to such dangerous levels. Effects from excessive noise are irreversible and actually destroy the tiny hair-like sensors in your ear.
If you want more information regarding noise levels and deafness, ask your boss for a free copy of the noise regulations. He is obliged by law to provide them.
Take care of your ears, you only get 1 pair
According to Australian Law:
Part 10 of the Workplace Health and Safety Regulation 1997 states that the employer must prevent risks to the health and safety of workers from exposure to excessive noise at work. Under the regulations, "excessive noise" is a level of noise above:
* an 8 hour equivalent continuous A-weighted sound pressure level of 85dB(A), referenced to 20 micropascals
* a (C)-weighted sound pressure level of 140dB(C), referenced to 20 micropascals
That means that peak noise at or above 140dB(C) is truely damaging, and continuous equivalent A-weighted SPL at or above 85dB(A) is also damaging.
(Note that the A and C weightings have to do with the position of the persons ears with respect to the noise, and how to add periods of noise together)
Although you might not think 85dB is damaging, if it is 85dB at the position of your ears, and you are exposed to it for more than 8 hours continuously, this will contribute to long term deafness. Noise effects are additive, ie each period of time where you have exposure to "excessive noise" (above 85dB(A) for 8 hours) are added together to produce the final A-Weighted exposure.
Every 3dB above that limit doubles the SPL, so you half the acceptable time limit:
8h @ 85dB
4h @ 88dB
2h @ 91dB
1h @ 94dB
30min @ 97dB
15min @ 100dB
7.5min@ 103dB
3.75min@106dB
1.88min@109dB
0.94min@112dB
0.47min@115dB
0.23min@118dB
7.0sec @121dB
3.5sec @124dB
1.8sec @127dB
0.9sec @130dB
0.4sec @133dB
0.2sec @136dB
0.1sec @139db
So no wonder 140dB is dangerous at peak levels, since you only need <0.1 seconds of exposure to contribute to long term deafness.
So to put it lightly, 130dB from speakers in an enclosed space like an apartment is absolutely rediculous, as probably around 120dB(A) will reach your ears. So not a good idea to listen to for more than a few seconds at a time unless you like the idea of becoming deaf in your old age.
In actual fact, going to a club where loud music is always playing at around 100dB is VERY damaging to your ears regardless of the frequency. Every time you come home with ringing in your ears, or temporary deafness from a powertool, is doing serious damage to your hearing! The human ear is a very sensitive device and was never intended to be subjected to such dangerous levels. Effects from excessive noise are irreversible and actually destroy the tiny hair-like sensors in your ear.
If you want more information regarding noise levels and deafness, ask your boss for a free copy of the noise regulations. He is obliged by law to provide them.
Take care of your ears, you only get 1 pair
Do you know how the A-curve looks like?
Get your facts right! Hehe like you said!
http://hyperphysics.phy-astr.gsu.edu/hbase/sound/acont.html
So, 100 dBA at 20 Hz is equal to ~150 dB unweighted. I wasn't that far off hehe!
So again, I wouldn't worry much about low bass.
Since Fletcher and Munson curves compress when the loudness goes up, you should stay under 130 dB unweighted at 20 Hz according to those curves to stay under 100 phons ( = 100 dB @ 1 kHz ).
Get your facts right! Hehe like you said!
http://hyperphysics.phy-astr.gsu.edu/hbase/sound/acont.html
So, 100 dBA at 20 Hz is equal to ~150 dB unweighted. I wasn't that far off hehe!
So again, I wouldn't worry much about low bass.
Since Fletcher and Munson curves compress when the loudness goes up, you should stay under 130 dB unweighted at 20 Hz according to those curves to stay under 100 phons ( = 100 dB @ 1 kHz ).
damo21,
Sure excessive "noise" does damage to our ears, but not to the extent you exaggerate it to be. The published papers in Australia are done by dumbass idiot researchers looking to make a buck. I'll tell you something for sure, let him build his 130dB because he wants to, it is his choice. I/We don't want to hear your stupid talk about how it's not good in the "workplace".
As for your assumptions, our hearing is non-linear and we are less sensitive at the lower frequencies. Which makes all your statements about halving the acceptance limit invalid. There comes a point where low frequencies no longer affect the small body of mass such as a human, and begin go radiate through buildings. At this point I would worry more about the building, any building that this subwoofer is. Because there arent many houses that can widthstand those low frequencies indefinetly.
I personally believe that 130dB+ can be achieved within those ranges by using multiple subs. However you could make the best subwoofer in the world and have it in a room thats not large enough or too large that it wouldnt matter. You have to take into consideration the room's performance before you can design any speaker and many of you are too caught up in dribble rather than investing some time in actually understanding. "Quick I'll just run a simulation" -- Get a clue!
Edit: and it seems simon5 has just added a reply with factual evidence that your claims are incorrect.
Sure excessive "noise" does damage to our ears, but not to the extent you exaggerate it to be. The published papers in Australia are done by dumbass idiot researchers looking to make a buck. I'll tell you something for sure, let him build his 130dB because he wants to, it is his choice. I/We don't want to hear your stupid talk about how it's not good in the "workplace".
As for your assumptions, our hearing is non-linear and we are less sensitive at the lower frequencies. Which makes all your statements about halving the acceptance limit invalid. There comes a point where low frequencies no longer affect the small body of mass such as a human, and begin go radiate through buildings. At this point I would worry more about the building, any building that this subwoofer is. Because there arent many houses that can widthstand those low frequencies indefinetly.
I personally believe that 130dB+ can be achieved within those ranges by using multiple subs. However you could make the best subwoofer in the world and have it in a room thats not large enough or too large that it wouldnt matter. You have to take into consideration the room's performance before you can design any speaker and many of you are too caught up in dribble rather than investing some time in actually understanding. "Quick I'll just run a simulation" -- Get a clue!
Edit: and it seems simon5 has just added a reply with factual evidence that your claims are incorrect.
Oh, ok no i didn't realise the A-curve was like that... cool
Well, 130dB at 2000Hz would be really damaging!
I guess the A-curve is derived from the frequency response of the human ear.
So why do people design loudspeakers to have a flat response?? Really the perfect loudspeaker system would have a response curve reciprocal of the A-curve.
ok no i didn't realise the A-curve was like that... coolWell, 130dB at 2000Hz would be really damaging!
I guess the A-curve is derived from the frequency response of the human ear.
So why do people design loudspeakers to have a flat response?? Really the perfect loudspeaker system would have a response curve reciprocal of the A-curve.
damo21 said:Oh,
Well, 130dB at 2000Hz would be really damaging!
I guess the A-curve is derived from the frequency response of the human ear.
So why do people design loudspeakers to have a flat response?? Really the perfect loudspeaker system would have a response curve reciprocal of the A-curve.
Speakers reproduce music, they do not enhance it or customise it to the A-curve or to an individual's hearing. You also can't design such a speaker, the reciprocal of the A-curve would be implemented using something like Linkwitz-Riley Transform on an already flat response system, which would require tremendeous amounts of power and also this boost would also be unrealistic since the reproduction of say an organ note at 23Hz gone up from 93->120dB does not accomodate the record->output ratio of 1:1.
And on another note we do already have methods of accomodating for something like the a-curve, usually called a loudness filter it helps where enclosures are too small or have trouble reproducing deep notes this adds an extra 3dB or so to the bass based on frequency...
No, I did not know the A-curve was frequency dependent you are correct.muhy3 said:
I was about 50 decibels too exaggerated for 20Hz sounds. At least I learnt from my mistake, thanks to simon5!
This comment is incorrect, I quoted from Australian law.
I quoted the workplace laws because they are the highest scientific authority on noise and sound pressure. Unfortunately, I did not find out about the A-weighting.
If my values above refer to A-weighted noise, which is what i really meant, then the halving argument is valid. It's just that 130dB at 20Hz isn't comparable to 130dB(A) which I didn't realise. Actually, 130dB at 20Hz is around 80dB(A) which is safe for your ears for any duration. But if 135dB at 20Hz is equivalent to 85dB(A) then it is unsafe for periods of more than 8 hours, according to the equations.
So at 20hz the acceptable time values are:
8h @ 135dB
4h @ 138dB
2h @ 141dB
1h @ 144dB
......
I guess I owe an apology to the dude building this sub, sorry. But keep in mind, your apartment will not like such sound pressures at low frequencies and your ears wont like it for extended listening periods.
"So why do people design loudspeakers to have a flat response?? Really the perfect loudspeaker system would have a response curve reciprocal of the A-curve."
Many people use EQ to get a so-called "house curve", which IIRC is a sloping upward response that results in about +10dB boost at 20 Hz.
Many people use EQ to get a so-called "house curve", which IIRC is a sloping upward response that results in about +10dB boost at 20 Hz.
damo21 said:
Greets!
No, it wouldn't. A flat in-room response is just one of the criteria required to ~accurately reproduce the signal. If you listen to/record a live acoustic event, then played it back on a system with this EQ curve it would sound nothing like it, ergo neither would one on a CD, etc..
That said, all media is compressed to a greater or lesser extent so that it can be recorded/played back at a reasonably high level on less than optimum equipment, so like vinyl's RIAA curve, I 'uncompress' it as much as practical using a processor that stores several EQ curves and select whichever one I believe works best for each media type.
GM
I rethought my original idea, and I realized that it doesn't really matter how the human ear perceives sound. The fact is, loudspeakers aim to reproduce all frequencies as recorded by the sound studio. The flatter the response of a loudspeaker system the better, because any audible frequency present in a recording has an equal weighting relative to any other frequency.
The whole point of equalizing is to offset any non-linearity of drivers which accentuate some frequencies better than others. Nothing new here.
But it is interesting how the human ear works:
We have a tonotopic map of frequencies in the audio region of our cortex, ie, a certain frequency is "heard" by mechanical resonation of a particular set of sensor s in our brain. By the particular geometric layout of these sensors which evolved, we are actually performing a fourier decomposition of frequencies without knowing it! Each narrow frequency band resonates a specific set of sensors, thus we are able to distinguish different pitch of sound. It's pretty cool! I studied a little about it last semester.
It's not surprising that we are very sensitive to sound at around 1-2kHz because this is about the frequency of our voices.
The whole point of equalizing is to offset any non-linearity of drivers which accentuate some frequencies better than others. Nothing new here.
But it is interesting how the human ear works:
We have a tonotopic map of frequencies in the audio region of our cortex, ie, a certain frequency is "heard" by mechanical resonation of a particular set of sensor s in our brain. By the particular geometric layout of these sensors which evolved, we are actually performing a fourier decomposition of frequencies without knowing it! Each narrow frequency band resonates a specific set of sensors, thus we are able to distinguish different pitch of sound. It's pretty cool! I studied a little about it last semester.
It's not surprising that we are very sensitive to sound at around 1-2kHz because this is about the frequency of our voices.
If I remember correctly the use of a loudness filter is to boost the high and low end when the system is at a low level and slowly flatten the response as it goes up in volume. This is used because our ears are especially insensitive to the highest and lowest frequencys at low volumes and the sensitivity of our ears "flattens" out a bit as volume increases. I think its kinda rediculis myself though from looking at equal loudness curves
Attachments
damo21 you can use a off the shelf RatShack SPL meter and set it on the A weighting. It's not perfect, but close to the truth. You can use a compensation chart to boost the lower bass and the upper treble, but I think that when used with the A weighting, the RatShack meter is quite close to the A curve. We usually use it on the C weighting and compensate heavily to get the dB unweighted readings.
I have not read the entire post, but has anyone suggested to you an Infinite baffle sub? that would be as close as possible to being flat from 10 to 40hz.....of course if you want outragous spl levels you would need outragous watts and wooferage, but there is no way around that whatever you do.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.