Hello, greetings to all members here! I'm an electrical and sound engineer from Hungary.
My hobby is designing and building all kinds of analog audio electronics.
As a first topic, I'd like to learn more about the sensitivity measurements of the speakers.
Especially, the measurement method(s) for tweeters.
I've got a Dayton DATS V3 system, and I've tried to use it for woofers or for midrange speakers, with the added mass method. It's easy and working fine for me.
But unfortunately, this isn't suitable for the majority of the tweeters (especially for the more delicate silk or other soft-dome ones).
I've also tried to measure that via a test-box method with a known internal volume.
Unfortunately again, DATS always gives me an error-message, regarding to the volume of the test box (it's either too large, or too small at the other attempt, it's annoying and strange...).
I've also read somwhere here in the forum, that if the TS-parameters of the driver has been measured, then the WINISD for example can calculate the sensitivity somehow... Sadly, I don't know the exact formula, how to get that.
I've also read about the miking method, when you put a mesurement mic in front of the speaker (in axis, 1m away) and put 1W of output power (or 2,83V when the impedance is 8 ohms) to it, and measure the SPL with a calibrated chain... something like that...
So, my question is... what's the best way to do it, with adequate accuracy but in the simplest way...?
My hobby is designing and building all kinds of analog audio electronics.
As a first topic, I'd like to learn more about the sensitivity measurements of the speakers.
Especially, the measurement method(s) for tweeters.
I've got a Dayton DATS V3 system, and I've tried to use it for woofers or for midrange speakers, with the added mass method. It's easy and working fine for me.
But unfortunately, this isn't suitable for the majority of the tweeters (especially for the more delicate silk or other soft-dome ones).
I've also tried to measure that via a test-box method with a known internal volume.
Unfortunately again, DATS always gives me an error-message, regarding to the volume of the test box (it's either too large, or too small at the other attempt, it's annoying and strange...).
I've also read somwhere here in the forum, that if the TS-parameters of the driver has been measured, then the WINISD for example can calculate the sensitivity somehow... Sadly, I don't know the exact formula, how to get that.
I've also read about the miking method, when you put a mesurement mic in front of the speaker (in axis, 1m away) and put 1W of output power (or 2,83V when the impedance is 8 ohms) to it, and measure the SPL with a calibrated chain... something like that...
So, my question is... what's the best way to do it, with adequate accuracy but in the simplest way...?
Several ways of deriving a Vas value from known parameters. They all vary a bit, but typically 'near enough'. Simplest:
Vas (litres) = 0.0014*(Sd^2)*Cms
From that, FWIW (& hopefully excusing the clumsy formulae as I've simply extracted from my long-suffering old Excel sheet & I never bothered refining them):
Reference efficiency η0 (decimal) = (0.00000000096352*(Fs^3)*Vas)/Qes
Nominal SPL in dB @ 1m/w = 112.018+10*(LOG10(η0))
Nominal sensitivity in dB @ 1m/2.83v = Nominal 1m/w SPL + (10*LOG(8/Re))
Vas (litres) = 0.0014*(Sd^2)*Cms
From that, FWIW (& hopefully excusing the clumsy formulae as I've simply extracted from my long-suffering old Excel sheet & I never bothered refining them):
Reference efficiency η0 (decimal) = (0.00000000096352*(Fs^3)*Vas)/Qes
Nominal SPL in dB @ 1m/w = 112.018+10*(LOG10(η0))
Nominal sensitivity in dB @ 1m/2.83v = Nominal 1m/w SPL + (10*LOG(8/Re))
Hi Peter,
Just want to confirm, you are trying to gather T/S parameters for Tweeters? That's normally not needed with sealed back tweets. Only with drivers with open backs where the choice of cabinet changes the low end response. DATS is however very useful in gathering the correct tweeter impedance.
Just want to confirm, you are trying to gather T/S parameters for Tweeters? That's normally not needed with sealed back tweets. Only with drivers with open backs where the choice of cabinet changes the low end response. DATS is however very useful in gathering the correct tweeter impedance.
Agreed; it isn't (normally needed) and it is (useful in establishing the actual impedance). I think he's after a baseline sensitivity figure however, which DATS can generate if you're able to measure the rest of the driver spec. Not as useful obviously as the measured 1m/2.83v value of course, but can serve as a quick & dirty guide if you don't have that or any other details, or a combination of that & manual calculation e.g. the above.
Gentlemen, thanks for your kind replies, okay then, I'll try to calculate those values first. Then perhaps I'd check it with my own mesurements - hopefully they won't be too inaccurate.
Erik, yes, you're right, maybe it makes no sense to gather all the "usual" T/S parameters for a tweeter, I just wanted to see the basics - fs, Qts, Re and such...
And Scott, thanks for the confirmation, I just would like to see a "comaparison" or a side-by-side rough masurement between my woofer and the tweeter in a given pair or set of speakers - whether they differ a lot or not... That way, I can then calculate an L-Pad or any other attenuator into my tweeter's path. As far as I could notice, tweeters are in general the more sensitive members, sometimes 2-4 or even 6dB "louder" - means more sensitive, rather - than the woofer in a given constellation.
Erik, yes, you're right, maybe it makes no sense to gather all the "usual" T/S parameters for a tweeter, I just wanted to see the basics - fs, Qts, Re and such...
And Scott, thanks for the confirmation, I just would like to see a "comaparison" or a side-by-side rough masurement between my woofer and the tweeter in a given pair or set of speakers - whether they differ a lot or not... That way, I can then calculate an L-Pad or any other attenuator into my tweeter's path. As far as I could notice, tweeters are in general the more sensitive members, sometimes 2-4 or even 6dB "louder" - means more sensitive, rather - than the woofer in a given constellation.