SPL of two combined speakers

[Miguelón]

Which is the expected SPL of two combined speakers versus one single?

If I have one single speaker at, for example, 80 dB/m for 100 watts and I add another equal it will add power and produces 86 dB combined SPL?

Or power addition is random and will be (100^2 + 100^2)^1/2 = 140 watts => around 102-103 dB?

 

[thewas]

In the bass where the 2 loudspeakers are mainly coherent sources it is theoretically 6 dB, above where they are incoherent theoretically 3 dB, in reality the measured values are between those limits.

Total dB level adding of coherent correlated sound sources combining decibels or SPL sound pressure level audio logarithmic decibel scale loudspeaker add signal noise levels incoherent noncoherentt - sengpielaudio Sengpiel Berlin

Total dB level adding of correlated coherent sound sources combining decibels or SPL sound pressure level audio logarithmic decibel scale loudspeaker add signal noise levels incoherent noncoherent - Eberhard Sengpiel sengpielaudio

www.sengpielaudio.com

Total dB level adding of incoherent or noncoherent uncorrelated sound sources signals combine sound levels two sources resultant level audio logarithmic decibel scale decibels or SPL sound pressure level add signal noise levels noncoherent incoherent

Total dB level adding of uncorrelated incoherent noncoherent sound sources signals combine sound levels two sources resultant level audio logarithmic decibel scale decibels or SPL sound pressure level add signal noise levels incoherent random phase acoustic - Eberhard Sengpiel sengpielaudio

www.sengpielaudio.com

 

[Miguelón]

In the bass where the 2 loudspeakers are mainly coherent sources it is theoretically 6 dB, above where they are incoherent theoretically 3 dB, in reality the measured values are between those limits.

Total dB level adding of coherent correlated sound sources combining decibels or SPL sound pressure level audio logarithmic decibel scale loudspeaker add signal noise levels incoherent noncoherentt - sengpielaudio Sengpiel Berlin

Total dB level adding of correlated coherent sound sources combining decibels or SPL sound pressure level audio logarithmic decibel scale loudspeaker add signal noise levels incoherent noncoherent - Eberhard Sengpiel sengpielaudio

www.sengpielaudio.com

Total dB level adding of incoherent or noncoherent uncorrelated sound sources signals combine sound levels two sources resultant level audio logarithmic decibel scale decibels or SPL sound pressure level add signal noise levels noncoherent incoherent

Total dB level adding of uncorrelated incoherent noncoherent sound sources signals combine sound levels two sources resultant level audio logarithmic decibel scale decibels or SPL sound pressure level add signal noise levels incoherent random phase acoustic - Eberhard Sengpiel sengpielaudio

www.sengpielaudio.com

Nice, that may explain why with one single active monitor I find pleasant the by default settings but when I add the second I find better with the -2 dB low shelving filter, which begins to operate at 1 kHz and reach -2 dB at 100 hZ

Even the -4 dB LS sounds good

 

[thewas]

That can be a good explanation.

 

[Viesturs Balodis]

This is how two speakers work at my experiment place. quite precisely 6 dB, except high FR because directivity etc, mono pink noise more precise mic placement necessary, because comb filters, peak below 30 is car or some noise. light blue is measurement of two speakers working together in mono. two measurements ~6dB below are each speaker separately same signal. daisy chained not switched to mono or stereo via controller.

 

[Miguelón]

View attachment 398033

This is how two speakers work at my experiment place. quite precisely 6 dB, except high FR because directivity etc, mono pink noise more precise mic placement necessary, because comb filters, peak below 30 is car or some noise. light blue is measurement of two speakers working together in mono. two measurements ~6dB below are each speaker separately same signal. daisy chained not switched to mono or stereo via controller.

Nice graphic, very instructive!

So in practice my preference about reducing low frequencies when stereo setup maybe due to some differences in room response because they are slightly closer to side walls, more than other reasons.

To my measurements the single speaker is placed in the center, whereas in my stereo setup each speaker is at 1/3 of the distance to the side walls.

 

[Viesturs Balodis]

Nice graphic, very instructive!

So in practice my preference about reducing low frequencies when stereo setup maybe due to some differences in room response because they are slightly closer to side walls, more than other reasons.

To my measurements the single speaker is placed in the center, whereas in my stereo setup each speaker is at 1/3 of the distance to the side walls.

hmm.. depends what you want to achieve.. mixing and listening different stuff a bit. But mic does not care much 1,2 or 3 (with condition that response is near same and room does not interfere) 2x will be ~+6dB, 3x ~+9dB etc.. if you have room modal stuff you have different response for each speaker so you will not get good wideband sum (think about xover summing as analogy)... more to that, sub in center is very bad idea in untreated room, because you have two and more equidistant reflections from side walls and all lateral corners etc. for example +6 to resonance peaks and additional -6dB to existing dips…

 

[Palmer]

One of the solutions to the +6db problem. I have the utmost respect for these guys, riggers are the unsung heroes of the live concert universe and nobody ever sees them or knows their names. That's a lot of weight, and it has to stay up. Every time.

 

[Miguelón]

hmm.. depends what you want to achieve.. mixing and listening different stuff a bit. But mic does not care much 1,2 or 3 (with condition that response is near same and room does not interfere) 2x will be ~+6dB, 3x ~+9dB etc.. if you have room modal stuff you have different response for each speaker so you will not get good wideband sum (think about xover summing as analogy)... more to that, sub in center is very bad idea in untreated room, because you have two and more equidistant reflections from side walls and all lateral corners etc. for example +6 to resonance peaks and additional -6dB to existing dips…

I no

hmm.. depends what you want to achieve.. mixing and listening different stuff a bit. But mic does not care much 1,2 or 3 (with condition that response is near same and room does not interfere) 2x will be ~+6dB, 3x ~+9dB etc.. if you have room modal stuff you have different response for each speaker so you will not get good wideband sum (think about xover summing as analogy)... more to that, sub in center is very bad idea in untreated room, because you have two and more equidistant reflections from side walls and all lateral corners etc. for example +6 to resonance peaks and additional -6dB to existing dips…

Thanks, I’ve just thinking about the sum of various speakers, I didn’t realized that different room responses between them measuring at same time can cause a little mess in the graph because of interference.

But for practice, 6 additional dB per doubling the speaker’s number (log sum) is the expected if the waves are coherent. In my actual setup and for detailed and less distortion as possible I have my monitors at 86 dB @ 1 volt and wanted to calculate roughly the maximum SPL for a 2Vrms signal: 94 dB for one single and 100dB expected for both in pair.

In reality I very rarely over 85 dB…

 

[Viesturs Balodis]

One of the solutions to the +6db problem. I have the utmost respect for these guys, riggers are the unsung heroes of the live concert universe and nobody ever sees them or knows their names. That's a lot of weight, and it has to stay up. Every time.

I would guess this is wannabe “cardioid” setup. As much I respect hard work of riggers I think this kind of “endfire” solutions for “cardioid” are flawed. Horn would be adequate “cardioid” and line array, this is bunch of arrays , and most remarkable thing of these multiple array things to get cardio is SPL control on stage and in many places where it is not possible with omni bass… this is another quite serious topic..

 

[Viesturs Balodis]

I no

Thanks, I’ve just thinking about the sum of various speakers, I didn’t realized that different room responses between them measuring at same time can cause a little mess in the graph because of interference.

But for practice, 6 additional dB per doubling the speaker’s number (log sum) is the expected if the waves are coherent. In my actual setup and for detailed and less distortion as possible I have my monitors at 86 dB @ 1 volt and wanted to calculate roughly the maximum SPL for a 2Vrms signal: 94 dB for one single and 100dB expected for both in pair.

In reality I very rarely over 85 dB…

check this site. maybe there is valuable info for you https://sengpielaudio.com/calculator-db-volt.htm

 

[Palmer]

I would guess this is wannabe “cardioid” setup. As much I respect hard work of riggers I think this kind of “endfire” solutions for “cardioid” are flawed. Horn would be adequate “cardioid” and line array, this is bunch of arrays , and most remarkable thing of these multiple array things to get cardio is SPL control on stage and in many places where it is not possible with omni bass… this is another quite serious topic..

Well, obviously a whole 'nother can of worms, I was merely pointing out the rule of +6 db is what it is and how you try to get enough power out there... Plus the riggers just rig, they don't exactly spec the system. I can tell you from my experience with the 2000 concert of the BeeGees, in what we then called the Wayne Huizenga Enrichment Center, the sound was pretty crazy good. Don't know who the man was on the desk, but the whole thing came off as if he'd popped a CD in.
In an Ice Rink.

Amerant Bank Arena (previously known as the National Car Rental Center, Office Depot Center, BankAtlantic Center, BB&T Center, and FLA Live Arena) is the largest indoor arena in Florida and is located next to Sawgrass Mills in Sunrise, Florida, United States.

 

[Viesturs Balodis]

Well, obviously a whole 'nother can of worms, I was merely pointing out the rule of +6 db is what it is and how you try to get enough power out there... Plus the riggers just rig, they don't exactly spec the system. I can tell you from my experience with the 2000 concert of the BeeGees, in what we then called the Wayne Huizenga Enrichment Center, the sound was pretty crazy good. Don't know who the man was on the desk, but the whole thing came off as if he'd popped a CD in.
In an Ice Rink.

Amerant Bank Arena (previously known as the National Car Rental Center, Office Depot Center, BankAtlantic Center, BB&T Center, and FLA Live Arena) is the largest indoor arena in Florida and is located next to Sawgrass Mills in Sunrise, Florida, United States.

hats off for riggers for sure and sound guys for hard work they do. Dave Rat understands both, and because of guys like him (but also antena designers) companies like LAcoustics or whatever improve their systems. usually not other way around. and for dBs the more the better, the more headroom the better.

 

[terryforsythe]

View attachment 398033

This is how two speakers work at my experiment place. quite precisely 6 dB, except high FR because directivity etc, mono pink noise more precise mic placement necessary, because comb filters, peak below 30 is car or some noise. light blue is measurement of two speakers working together in mono. two measurements ~6dB below are each speaker separately same signal. daisy chained not switched to mono or stereo via controller.

That is a nice correlation. Much better than what I have.

Below are plots of my in-room frequency response measurements using REW. The orange line is an RMS average of the Left and Right channels. The red line is both channels being measured simultaneously.

The speakers are 1.24m apart measured center-to-center, and the baffles are about 0.64m from the wall. The measurements were performed with the microphone about 2m from each of the speakers.

 

[Palmer]

Since this is a technical forum I might add: The poor chap on the board, British fellow, had a hard time - because everybody and their dog wanted a stereo feed from his mix.
I was with ABC at the time doing a live feed from the gig via Satellite - and this guy ran out of devices to try to isolate his ground from all the other guys ground.
First time for me to see a company use a glass desk for live sound, too. Memories...

Yeah, Dave Rat - knows what he's doing. Big fan of his and all the free knowledge he gives out.

 

[Miguelón]

check this site. maybe there is valuable info for you https://sengpielaudio.com/calculator-db-volt.htm

In this case I think they are correct: 86 dB @ 1 Vrms make 94 dB because 2 volts are roughly 8,5 dBu. Adding with the mentioned 2 speakers should get a maximum of 100-101 dB /m

My question were more on the addition of sources than the transformation of signal units

 

[JohnnyAudio]

Kind of like fireworks... more fireworks, more punch.
Disney uses subwoofers to enhance the fireworks show... saves money of disposal sulfur.

 

[Viesturs Balodis]

That is a nice correlation. Much better than what I have.

Below are plots of my in-room frequency response measurements using REW. The orange line is an RMS average of the Left and Right channels. The red line is both channels being measured simultaneously.

View attachment 398099

I dont think it is that bad as you say. for me looks ok. only bigger problem is at 10k and a bit at 5k, I think you just need to adjust mic position to get this a bit better or speaker, easier with the mic first. everything else can be done with preference EQ. I recommend that you dont average anything, smoothing 1/12oct is workable if room not that good, but helps to p… off yourself when seeing things without averages sometimes, to force yourself to make it better, you need to hate this as a trigger to make it better

 

[Miguelón]

That is a nice correlation. Much better than what I have.

Below are plots of my in-room frequency response measurements using REW. The orange line is an RMS average of the Left and Right channels. The red line is both channels being measured simultaneously.

View attachment 398099

The speakers are 1.24m apart measured center-to-center, and the baffles are about 0.24m from the wall. The measurements were performed with the microphone about 2m from each of the speakers.

Curious to see that in the bass region the difference between individual speaker (average, sorry) is closer to addition

 

[Viesturs Balodis]

In this case I think they are correct: 86 dB @ 1 Vrms make 94 dB because 2 volts are roughly 8,5 dBu. Adding with the mentioned 2 speakers should get a maximum of 100-101 dB /m

My question were more on the addition of sources than the transformation of signal units

this is a good site. search it, many good things there. and dont forget to see difference between rms and peak values.

Calculation voltage and power - gain and loss - sengpielaudio Sengpiel Berlin

Calculation: Voltage and Power - Gain and Loss - Eberhard Sengpiel sengpielaudio

sengpielaudio.com