I am designing a speaker, and have decided that 105 db one meter would be an appropriate target. My thinking was I listen at 2 meters, 92 db absolute max. So at 1 meter with dipole, add 3 db for 95 db plus 10 db headroom. So I started by specifying drivers and amplifiers so any tone would play at least 105 db, both power, thermal and Xmax.
I am bi-amping at about 300 Hz, so for pink noise, I can add 6 db.
Also, for 2 speakers, add another 6 db.
Does that bring the max volume up to 117 db for pink noise at 1 meter, two speakers driven?
This seems a bit excessive to me, but I am looking for other perspectives.
Thanks in advance.
Doug
I am bi-amping at about 300 Hz, so for pink noise, I can add 6 db.
Also, for 2 speakers, add another 6 db.
Does that bring the max volume up to 117 db for pink noise at 1 meter, two speakers driven?
This seems a bit excessive to me, but I am looking for other perspectives.
Thanks in advance.
Doug
DougL said:
If you need to play at 92 dB at 2 m, it will be 98 dB at 1 m (a 6 dB loss for every doubling of distance -- ignoring room load & reflections). ie REALLY loud. 10 dB is not all that much headroom.
It doesn't add any dB -- just means you require less total power before overload if you are playing a tone below the crossover & one above the crossover.
For 2 speakers add 3 dB
No. If each speaker is able to play 105 dB without overload at 1 m, 2 speakers will provide 102 dB at 2 m (ignorning the room)
dave.
Re: Re: Max SPL theory Question
That is correct.
As a practical matter, though, with all those reflections, I think you will find in an average room that the average SPL is only a few dB less than the 1 meter@1 watt level-wherever you are in the room. I don't think that you are going to walk 12 feet away from the speakers in a room and find that the SPL is 12 dB down from the 1 meter/1 watt measurement.
I believe that two identical speakers, each powered by a separate amplifier channel playing at 1 watt, will play 6 dB louder than one speaker powered by a single amplifier channel playing at 1 watt.
planet10 said:
That is correct.
As a practical matter, though, with all those reflections, I think you will find in an average room that the average SPL is only a few dB less than the 1 meter@1 watt level-wherever you are in the room. I don't think that you are going to walk 12 feet away from the speakers in a room and find that the SPL is 12 dB down from the 1 meter/1 watt measurement.
Two speakers hooked up to a single channel will increase 3 dB over one speaker hooked up to a single channel.planet10 said:
I believe that two identical speakers, each powered by a separate amplifier channel playing at 1 watt, will play 6 dB louder than one speaker powered by a single amplifier channel playing at 1 watt.
Re: Re: Re: Re: Max SPL theory Question
True, but when you are dealing with excursion requirements, we usually concentrate on the loudest sound.
Bass puts a strain on excursion, but that tends to be mixed equally between channels. Lead vocals tend to be the loudest sounds above bass-but most lead vocals are mixed equally between channels.
There is no guarantee that this need be the case, but it is true more often than not. So on the sounds that tax excursion, most of the time, both channels will be contributing equally.
Gregm said:
True, but when you are dealing with excursion requirements, we usually concentrate on the loudest sound.
Bass puts a strain on excursion, but that tends to be mixed equally between channels. Lead vocals tend to be the loudest sounds above bass-but most lead vocals are mixed equally between channels.
There is no guarantee that this need be the case, but it is true more often than not. So on the sounds that tax excursion, most of the time, both channels will be contributing equally.
-- sorry
Greets!
>I am designing a speaker, and have decided that 105 db one meter would be an appropriate target.
====
OK, but depending on what source material you have, this could limit you to as little as 75dB/m average, or ~ how high folks listen to prime time TV at.
====
>I am bi-amping at about 300 Hz, so for pink noise, I can add 6 db.
====
?? What's pink noise got to do with it?
====
>Also, for 2 speakers, add another 6 db.
====
Uh-uh. We're summing SPLs in this case, which is different from adding a second driver/doubling power to a speaker, so:
10*log10(10^(SPL1/10)+10^(SPL2/10), + etc.) = 3.01dB for two identical SPLs.
Of course for stereo this will be frequency dependent due to all the source/room phase issues, but is close enough for calcing power requirements in the BW above where the room gain curve kicks in.
====
>Does that bring the max volume up to 117 db for pink noise at 1 meter, two speakers driven?
====
So far, with 105 dB/m we're at 108.01 dB/m with two speakers. In free space SPL falls at the square of distance so a 'worst case' at 2m:
108.01 - (20*log10(1/2)) = 101.99 dB/2m
Due to room issues, the average drop in the typical HI-FI/HT room is 3-4 dB, but if using the free space calc then you never have to worry about clipping the speaker and/or amp if you start with a realistic peak SPL requirement.
For most folks the DD/DTS HT reference of 105 dB peaks/channel/listening position is plenty loud so this seems a reasonable goal for most music. For symphonys and/or organ music, then up to 112 dB/channel/listening position; and if ~live levels are desired for these, up to 127 dB/channel/listening position in the 250-500 Hz BW is required if you want to experience ~ a 'front row center' seating position.
For 'headbangers', I was at one outdoor 'rockabilly' concert back in the mid '70s where I measured >130 dB peaks (kept pegging the meter)/~100yds from the stage, so factoring in all the compression they use(d) to increase average SPL and low loss over distance in the humid night air, I figure it was at least 115 dB/m average! For sure it was loud enough that my eardrums compressed and I wound up wearing plugs to keep the upper mids from painfully 'drilling' me.
GM
>I am designing a speaker, and have decided that 105 db one meter would be an appropriate target.
====
OK, but depending on what source material you have, this could limit you to as little as 75dB/m average, or ~ how high folks listen to prime time TV at.
====
>I am bi-amping at about 300 Hz, so for pink noise, I can add 6 db.
====
?? What's pink noise got to do with it?
====
>Also, for 2 speakers, add another 6 db.
====
Uh-uh. We're summing SPLs in this case, which is different from adding a second driver/doubling power to a speaker, so:
10*log10(10^(SPL1/10)+10^(SPL2/10), + etc.) = 3.01dB for two identical SPLs.
Of course for stereo this will be frequency dependent due to all the source/room phase issues, but is close enough for calcing power requirements in the BW above where the room gain curve kicks in.
====
>Does that bring the max volume up to 117 db for pink noise at 1 meter, two speakers driven?
====
So far, with 105 dB/m we're at 108.01 dB/m with two speakers. In free space SPL falls at the square of distance so a 'worst case' at 2m:
108.01 - (20*log10(1/2)) = 101.99 dB/2m
Due to room issues, the average drop in the typical HI-FI/HT room is 3-4 dB, but if using the free space calc then you never have to worry about clipping the speaker and/or amp if you start with a realistic peak SPL requirement.
For most folks the DD/DTS HT reference of 105 dB peaks/channel/listening position is plenty loud so this seems a reasonable goal for most music. For symphonys and/or organ music, then up to 112 dB/channel/listening position; and if ~live levels are desired for these, up to 127 dB/channel/listening position in the 250-500 Hz BW is required if you want to experience ~ a 'front row center' seating position.
For 'headbangers', I was at one outdoor 'rockabilly' concert back in the mid '70s where I measured >130 dB peaks (kept pegging the meter)/~100yds from the stage, so factoring in all the compression they use(d) to increase average SPL and low loss over distance in the humid night air, I figure it was at least 115 dB/m average! For sure it was loud enough that my eardrums compressed and I wound up wearing plugs to keep the upper mids from painfully 'drilling' me.
GM
Doug,
What are you really asking? What volume do you require? You seem to say 105 dB target for headroom then 92 dB max listening volume. Not sure what you are looking for. What is the system for?
80 dB = nice volume
90 = having to talk louder to hear each other
100 = yelling to be heard
110 = ears starting to buzz
120 = girls leave the party
130 = increased chance of nose bleed
140 = knocking the feathers off a chicken
150 = deboning the same chicken
My outdoor system with multiple drivers and amps has a theoretical top end of 130dB. Absolutely insane volume.
My indoor gear rarely exceeds 100 dB and that's only when someone puts on the Jimmy Buffett and passes me a margarita.
Cal
What are you really asking? What volume do you require? You seem to say 105 dB target for headroom then 92 dB max listening volume. Not sure what you are looking for. What is the system for?
80 dB = nice volume
90 = having to talk louder to hear each other
100 = yelling to be heard
110 = ears starting to buzz
120 = girls leave the party
130 = increased chance of nose bleed
140 = knocking the feathers off a chicken
150 = deboning the same chicken
My outdoor system with multiple drivers and amps has a theoretical top end of 130dB. Absolutely insane volume.
My indoor gear rarely exceeds 100 dB and that's only when someone puts on the Jimmy Buffett and passes me a margarita.
Cal
guys pleazz, correct me if I'm wrong..
SPL is a not a measure of energy but presure ... right?
So SPLs are denoted in 20*LOG .. and power is denoted in 10*LOG .. right?
2 speakers have 3dB SPL (20*LOG ) increased efficientcy..due to the more efficient coupling to the air's impedance...?
2 speakers have 2*more max power dissipation capability, so 2*more power into the air= 6dB more SPL?
greetz,
Thijs
SPL is a not a measure of energy but presure ... right?
So SPLs are denoted in 20*LOG
.. and power is denoted in 10*LOG .. right?2 speakers have 3dB SPL (20*LOG
) increased efficientcy..due to the more efficient coupling to the air's impedance...?2 speakers have 2*more max power dissipation capability, so 2*more power into the air= 6dB more SPL?
greetz,
Thijs
Thanks everyone for the responses.
planet10
GM
Since that would model better with pink noise rather than single tones, it seemed reasonable that the power was additive across drivers in a biamp’ed system. probably 10*log10(10^(SPL1/10)+10^(SPL2/10), + etc.) = 3.01dB for two identical SPLs.
Cal Weldon
The Score card so far:
105 db 1 meter 1 speaker single tone.
108 db 1 meter 2 speakers single tone.
102 db 2 meters 2 speakers single tone.
105 db 2 meters 2 speakers multiple tones. (Still under discussion)
planet10
Thanks, I misinterpreted something on the Linkwitz site that seemed to imply that dipoles were a special case.
Could you elaborate? My intuition says that 2 speakers playing different frequency's would sum.
GM
I am making an engineering assumption that high volume music "is" broadband rather than single tone in nature.
Since that would model better with pink noise rather than single tones, it seemed reasonable that the power was additive across drivers in a biamp’ed system. probably 10*log10(10^(SPL1/10)+10^(SPL2/10), + etc.) = 3.01dB for two identical SPLs.
Cal Weldon
This is a music only system for moderate listening levels of Jazz and Rock. From your scale, I never get above 90 db. And I never do anything with chickens
The Score card so far:
105 db 1 meter 1 speaker single tone.
108 db 1 meter 2 speakers single tone.
102 db 2 meters 2 speakers single tone.
105 db 2 meters 2 speakers multiple tones. (Still under discussion)
Greets!
>SPL is a not a measure of energy but presure ... right?
====
Correct, and why it's called sound pressure level.
====
>So SPLs are denoted in 20*LOG .. and power is denoted in 10*LOG .. right? 2 speakers have 3dB SPL (20*LOG ) increased efficientcy..due to the more efficient coupling to the air's impedance...?
====
The multiplier depends on whether the dB SPL is derived from a power or voltage:
power:
dB SPL = 10*log10(2) = 3.01 dB
voltage:
dB SPL = 20*log10(E/2.83V)
====
>2 speakers have 2*more max power dissipation capability, so 2*more power into the air= 6dB more SPL?
====
No, see my comments on summing SPLs.
GM
>SPL is a not a measure of energy but presure ... right?
====
Correct, and why it's called sound pressure level.
====
>So SPLs are denoted in 20*LOG
.. and power is denoted in 10*LOG .. right? 2 speakers have 3dB SPL (20*LOG ) increased efficientcy..due to the more efficient coupling to the air's impedance...?====
The multiplier depends on whether the dB SPL is derived from a power or voltage:
power:
dB SPL = 10*log10(2) = 3.01 dB
voltage:
dB SPL = 20*log10(E/2.83V)
====
>2 speakers have 2*more max power dissipation capability, so 2*more power into the air= 6dB more SPL?
====
No, see my comments on summing SPLs.
GM
Greets!
> Thanks everyone for the responses.
====
You're welcome!
====
>I am making an engineering assumption that high volume music "is" broadband rather than single tone in nature.
====
Volume doesn't matter, music IS broadband from single digits to at least 40kHz and nothing's been said to imply otherwise.
====
>Since that would model better with pink noise rather than single tones, it seemed reasonable that the power was additive across drivers in a biamp’ed system.
====
Acoustic power falls at 3 dB/octave (1/f) just like pink noise, so yes, it's best for level matching a system. Unless there's a screw-up, the XO rolls off the output on both sides of it so that the SPL's of both drivers sum flat. The goal is for both channels to have a ~flat in-room response at least at the listening position in the audible BW and level matched to each other.
====
>This is a music only system for moderate listening levels of Jazz and Rock. From your scale, I never get above 90 db.
====
OK, if you mean this as an average, then based on measuring the various jazz recordings I have, you'll need up to +27dB of dynamic headroom to keep from clipping the system on fast transients. Rock is so compressed I doubt you'll find any recording that has as much dynamic headroom as good jazz recordings, so not a factor.
GM
> Thanks everyone for the responses.
====
You're welcome!
====
>I am making an engineering assumption that high volume music "is" broadband rather than single tone in nature.
====
Volume doesn't matter, music IS broadband from single digits to at least 40kHz and nothing's been said to imply otherwise.
====
>Since that would model better with pink noise rather than single tones, it seemed reasonable that the power was additive across drivers in a biamp’ed system.
====
Acoustic power falls at 3 dB/octave (1/f) just like pink noise, so yes, it's best for level matching a system. Unless there's a screw-up, the XO rolls off the output on both sides of it so that the SPL's of both drivers sum flat. The goal is for both channels to have a ~flat in-room response at least at the listening position in the audible BW and level matched to each other.
====
>This is a music only system for moderate listening levels of Jazz and Rock. From your scale, I never get above 90 db.
====
OK, if you mean this as an average, then based on measuring the various jazz recordings I have, you'll need up to +27dB of dynamic headroom to keep from clipping the system on fast transients. Rock is so compressed I doubt you'll find any recording that has as much dynamic headroom as good jazz recordings, so not a factor.
GM
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.