Is this how to compare speaker sensitivities?

[Datoyminaytah]

A speaker's sensitivity, or SPL, equals the decibel level of 1 watt measured at 1 meter. Right?

Now...from googling around, I've found (I think) that multiplying watts times about 1.29 should increase the volume by 1 decibel.

That fits pretty well with the general rule I've read that doubling watts increases dB by 3, and that watts x 10 equals double the volume, and that "double the volume" is subjectively measured at between a 6 to 10 dB increase.

So, that should mean that a speaker with a sensitivity of 92 dB, supplied with 100 watts, and a speaker with a sensitivity of 88 dB, supplied with 277 watts, should produce about the same perceived volume. The math I used:

92dB - 88dB = 4 decibel difference

Multiply 100w @ 88dB times 1.29, 4 times

100w x 1.29 x 1.29 x 1.29 x 1.29 = 276.9w

Is that right? If not, please enlighten me.

 

[Svante]

It seems right to me, even though I get the factor 1.26. Also, the more general power gain formula is 10^(dB/10), so 1 dB is 10^(1/10)=1.26, 3 dB is 10^(3/10)=2.00. But you get the same result (if you use 1.26), so your method is OK too.

 

[BillFitzmaurice]

Watts up?

Keep in mind impedance also. 2.83 volts into 8 ohms is one watt, but so is 2.0 volts into 4 ohms, and 4.0 volts into 16 ohms. You cannot compare the relative SPLs of different drivers without taking impedance into consideration.

 

[Datoyminaytah]

Yes, but almost all automotive drivers are 4 ohm, and I wasn't thinking about other impedances.

But, could you explain what you mean? Isn't a watt a watt regardless of the volts and impedance used to get it? In your example of 2.83 volts into 8 ohms vs. 2 volts into 4 ohms, if both drivers have the same sensitivity rating, won't they be equally loud? If not, why?

 

[SY]

A speaker's sensitivity, or SPL, equals the decibel level of 1 watt measured at 1 meter. Right?

Not right. Sensitivity is the SPL at a reference distance (usually 1 m) at a constant reference voltage (usually 2.83 v). That way, under likely drive conditions (most amplifiers are voltage sources), you can know how loud the speaker will be independent of what the impedance is at the reference frequency.

 

[wunhuanglo]

Aren't speaker sensitivities quoted as dB/w/m? At least all the ones I've seen are.

If so, then 2.83V only applies to 8 ohm impedence speakers.

 

[SY]

Aren't speaker sensitivities quoted as dB/w/m?

No, they're not, and specifically because impedance is complex and not constant with frequency. I've seen efficiency spec'ed as dB/W/m, but that's a different kettle of fish. And not nearly as useful in design as sensitivity specs.

 

[BillFitzmaurice]

A speaker's sensitivity, or SPL, equals the decibel level of 1 watt measured at 1 meter. Right?

Yes and no. You can quote SPL with 1 watt input, but it's not the only way to compare relative SPLs. They are often stated with a voltage input, which is to some extent a better indication of how loud a speaker will actually be with a given amplifier. That is because an amp will have different wattage output capabilities with different load impedances but it's maximum voltage output never changes.

For an 8 ohm speaker and a 4 ohm speaker with the same wattage sensitivity and the same voltage input the 4 ohm speaker will be 3dB louder, because with the same voltage input the 4 ohm speaker will be receiving twice the wattage as the 8 ohm speaker. If the two speakers are rated for SPL/2.83 volts the 4 ohm speaker will be 3dB higher than the 8 ohm speaker even though the SPL/watt is the same. Therefore the 2.83 v sensitivity rating gives a more accurate representation of how loud different speakers may be with the same voltage input.