Confused about speaker "sensitivity" when using multiple speakers

[waspclothes]

Hey everyone,

I'd like to go from using a single 12" speaker to a 4x10" arrangement. The rated sensitivity of the original 12" is 102db, and the new 10" speakers are 96.1db.

What would the effective sensitivity of the 4x10" cabinet be when wired in series-parallel?

[Jazz P Bass]

Maybe this calculator will help: Adding acoustic levels summing sound levels 10 combining addition summation sum decibel levels or SPL of up to ten incoherent sound sources audio logarithmic decibel scale identical summing 1/3 octave spl full octave sum sound pressure level noise so

Got Math?

[The Dude]

Much of this goes out the window when you put the speakers in a cabinet anyway. The specs normally give you "open air" characteristics for comparison. The manufacturer has no idea what cabinet you're going to put their speakers in. Cabinet volume, speaker proximity to each other, ported or not, closed back or open, etc., etc. Not that the question isn't valid from some sort of physics standpoint, but I'm curious if there isn't a "real world" reason for pondering such a thing. It seems like a pointless and impossible venture to try and predict an outcome with so many unknown variables.

[Enzo]

If you are planning to compare the two for loudness, it is only a few decibels, which is easily covered by turning the volume control a notch.

Most older Fender amps have pairs of input jacks, and one is louder than the other. Those two jacks are 6 decibels apart. That is the difference. But as Dude points out, that is only ONE factor among many.

[Gingertube]

As a broad approach:

1 off the "new" 10" speaker is 96.1dB/W/m sensitivity.
If you wire a second in parallel on the output of the same amplifier you effectively double the Sound Pressure Level (+3dB) , so you could say that the combined sensitivity is now 96.1 + 3 = 99.1dB/W/m.
If you then double up the speakers again, that is 4 off the 10" speakers then you get another +3dB, so you could say that the combined sensitivity of 4 off 10" speakers is 99.1 + 3 = 102.1dB/W/m.

As far a Sound Pressure Level goes you will end up with much that same as your old single 12" speaker.

Cheers,
Ian

[Jensen P]

As a broad approach:

1 off the "new" 10" speaker is 96.1dB/W/m sensitivity.
If you wire a second in parallel on the output of the same amplifier you effectively double the Sound Pressure Level (+3dB) , so you could say that the combined sensitivity is now 96.1 + 3 = 99.1dB/W/m.
If you then double up the speakers again, that is 4 off the 10" speakers then you get another +3dB, so you could say that the combined sensitivity of 4 off 10" speakers is 99.1 + 3 = 102.1dB/W/m.

As far a Sound Pressure Level goes you will end up with much that same as your old single 12" speaker.

...at 1 watt and 1 m. -What does that actually mean in a practical application?

Players would typically like to know:

-What's the sound pressure level at max clean volume? -At full (distortion) volume?

and; -What is the sound pressure level required for the speakers to perform near its advertised frequency response curve?

Most guitar speakers will have to be played fairly loud to perform. That point in the output curve would be the real "sensitivity" for any practical purpose. But it appears like the only way to find out is by trial and error...

[Enzo]

The numbers are a standard, it gives you SOMETHING to compare speakers by. Once we get into how loud on stage and stuff, there are so many variables, it would be tough to chart. Is the speaker going to be alone or with others? In an open back or closed back cab? Each speaker would need a book.

[Gingertube]

Yes - These "standards" are very simplistic and in this case means exactly want dB/W/m implies.
It describes the Sound Pressure Level (SPL) in dB, that you would experience standing 1m in front of that speaker, with 1 watt of power applied to the speaker.
Cheers,
Ian

[Leo_Gnardo]

As far a Sound Pressure Level goes you will end up with much that same as your old single 12" speaker.

Cheers,
Ian

If 1 watt is presented to the array of four, then only a quarter-watt goes to each, and the figure is right back down to the starting point. No free lunch in the universe, pointed out in another thread. COULD be a noticeable change in volume/freq response due to comparing one cone to four (more cone area supporting a better LF resonse), open back 12 to closed back 4x10, but all other things being equal the sensitivity of 4 = sensitivity of 1.

[eschertron]

If 1 watt is presented to the array of four, then only a quarter-watt goes to each, and the figure is right back down to the starting point.

Players would typically like to know:-What's the sound pressure level at max clean volume? -At full (distortion) volume?

Leo, you are assuming that the relationship between power and SPL is linear throughout the operation of the speaker. I haven't seen a power vs loudness curve, and that looks to me like what Jensen P. is asking for. Somebody here must have gotten the urge to collect these data at one time or another?

Just from seat-of-the-pants "thought experiment": the loudness law says double the power -> increase SPL by 3dB. so from 1W to 128W (doubled 7 times) our sound has increased by 21dB. Have I reached a limiting point yet, or has my (150W max rated power) speaker gone from 99dB to 120dB? I honestly don't know, but it seems excessive to me...

[pdf64]

Leo, coupling between coherent sources leads to an increase in SPL of 3dB for the same total sound power level, each time the number of drivers is doubled.
Something to do with power being a scalar quantity (magnitude only), pressure a vector (magnitude and direction).
There was a big thread about it on TGP about 18 months ago, it ended up with Dr dB taking measurements that confirmed the physics.
No free lunches, the gain is measured in front of the drivers, what is gained in front is lost to the sides and the physical separation of the sources leads to wave cancellation and notches in the frequency response.
Those notches mean that for linear systems, the phenomenon can only utilised at lower frequencies.
Pete

EDIT see http://www.sengpielaudio.com/calcula...entsources.htm

[Leo_Gnardo]

Leo, you are assuming that the relationship between power and SPL is linear throughout the operation of the speaker. I haven't seen a power vs loudness curve, and that looks to me like what Jensen P. is asking for. Somebody here must have gotten the urge to collect these data at one time or another?

Just from seat-of-the-pants "thought experiment": the loudness law says double the power -> increase SPL by 3dB. so from 1W to 128W (doubled 7 times) our sound has increased by 21dB. Have I reached a limiting point yet, or has my (150W max rated power) speaker gone from 99dB to 120dB? I honestly don't know, but it seems excessive to me...

<power & SPL linear throughout the operation> nope I din't say that. Just a side by side at 1W. "Assuming" - and you all know what about that "assume" - the SPL vs power is linear, then that's the math all right. There are some factors that cause nonlinearity or "compression" at high power, like the voice coil not being in the magnetic field's "sweet spot" for part of its travel, and VC heating. Hi SPL can also cause other nonlinear events, like the flashing of police car lights and neighbors banging on the doors floors or ceiling but that's another matter.

For our OP, sharing the power across a number of speakers could well mean expecting a more linear operation at power, again "all other things being equal" but we already de-equalized it by going from 12" to 10" and different brands with different VC dimensions and different magnet dimensions and different compliances at the spider & cone edge and can you think of any more. Jeez it is impossible to predict.

Sure would be informative if manufacturers published power vs SPL curves but I don't think most of 'em want us to see 'em with their pants down. Pro outfits like PA companies likely do their own experiments on this and select the best of the lot because they have to buy hundreds of speakers and want to get the most out of them - it can be a multimillion dollar investment for an outfit like Clair Bros for instance. Guitar amp manufacturers? Not so much. Hey it gets ear-splitting loud, what more do you need to know?

Multidriver systems developing "notches" in the frequency response has been noted since the 1920's. Funny the "line array" seems to have made a comeback in the last 15 years. Kind of like democracy, it's a bad system all right but all the others have proven worse, thank you Mr. Churchill. I got one of my guitar-playing customers who is partial to 2x12's to try stacking three of them the long way and got the "OMG you're right it's awesome!" review. Now, how to keep that rig from falling over onstage is the remaining problem.

[The Dude]

Yes - These "standards" are very simplistic and in this case means exactly want dB/W/m implies.
It describes the Sound Pressure Level (SPL) in dB, that you would experience standing 1m in front of that speaker, with 1 watt of power applied to the speaker.
Cheers,
Ian

True, but only useful if the speakers are not in a cabinet, which is not a normal situation. As I said, once you mount the speakers, it's all out the window. Even if two different speakers are installed in identical boxes, they will react differently based on the speaker parameters and their relationship to the cabinet. I'll give you that RAW speaker sensitivity is a fixed spec. HOWEVER, the relevance of that spec means little when boxed.

For example, let's say that speaker A has a better sensitivity spec than speaker B in open air. Then let's say that we put them both in identical boxes wherein the box is better designed for speaker B's parameters. It is quite possible and probable that, IN THIS PARTICULAR ENCLOSURE, speaker B would outperform speaker A in both sensitivity and max SPL.

Which is better- a jar of quarters or the equivalent in dollar bills? It depends. Are you going to the laundromat or the strip club?

[Leo_Gnardo]

True, but only useful if the speakers are not in a cabinet, which is not a normal situation. - - - snip - - - Which is better- a jar of quarters or the equivalent in dollar bills? It depends. Are you going to the laundromat or the strip club?

How can a speaker manufacturer spec cone speakers that aren't at least mounted to a baffle? For instance some approximation of the old "infinite" baffle, one wall of a more or less anechoic chamber so meaningful measurements can be made. Hang a speaker in the air, just a chain and a hook thru one of the frame's bolt holes, and you won't get much volume especially at low frequencies. Can't see making measurements like that would make much sense.

- - - - - - -

Why not combine the operations? You gotta take off your clothes to wash 'em. Careful which slot you push those quarters into!

[J M Fahey]

Speakers *are* measured in a "standard baffle", not alone hanging from a hook.
Here's specs for 8" to 15" speakers.
Not the best baffle in the world, not the worst either.