Knowing average listening SPLs is interesting, but it is not at all the point of this test.
This test allows you to know how close to clipping your amp is when you are at your loudest listening levels.
As for more current needed for speakers during peaks, that's a different matter. An amplifier of reasonable quality that can supply 100W RMS can certainly handle occasional peaks of 200W. And if you stay far enough below clipping, you should not run into trouble. The test allows you to know where in your amp's range you are playing.
This test allows you to know how close to clipping your amp is when you are at your loudest listening levels.
As for more current needed for speakers during peaks, that's a different matter. An amplifier of reasonable quality that can supply 100W RMS can certainly handle occasional peaks of 200W. And if you stay far enough below clipping, you should not run into trouble. The test allows you to know where in your amp's range you are playing.
Thanks for posting, Mat.
What does your 104dB figure represent? Is that the SPL you got from the tone after setting your volume to the loudest point you normally use? Or is it something else?
65V is way up there for a -12dBfs tone. What's the stated sensitivity of the speakers?
Pink noise and a low pass filter (such as alias filter) can create higher P-P amplitude than a full scale sine wave does.
Same can definitely be said for active filters.
From what I have heard Class-D amps sound best below 50-70% amplitude, which automatically takes care of that problem.
unlikely to happen in real life. one needs a full-scale square wave or something close in order to reach that scenario. something which is practically never to be found in real music.
Hi Pano,
Yes, forgive me; that's dB SPL with your test tones, with my volume control set at the loudest point I would normally play my music.
According to the manual:
EFFICIENCY: (400 Hz, 1 watt into 8 ohms equivalent at one meter)*............ Spectra 44/4400: 87 dB SPL
* extrapolated from measurements taken at 10 watts into 8 ohms equivalent, measured at a distance of 3.16 meters.
Don't be shy to tell me I'm doing something wrong, if I am.
Last edited:
Yes Pano, Acoustat built the TNT 200 specifically for powering their electrostatic speakers. And as stated previously, this amp and my speaker interfaces have been serviced (totally refurbished) and upgraded by prior Acoustat technician Roy Esposito. Also, at the moment, he is performing the same service on two TNT 200's I provided him specifically to include capability to operate in mono mode (switchable). When I receive these and get them connected, I'll be back to post the values I get with them. Should be interesting.
Last edited:
What I am saying is that volts and SPL are intimately connected.
dB_listen = dB_sensitivity + 20 * log(Volts_listen/Volts_sensitivity)
or:
Volts_listen = Volts_sensitivity * 10^((dB_listen-dB_sensitivity)/20)
Volts_sensitivity is the voltage at which sensitivity is rated.
Volts_listen is the voltage you listen at
dB_listen is the dB you listen at
dB_sensitivity is the rated voltage sensitivity (typically at 2.83 Volts)
Example:
I have 94 dB@2.83V sensitive loudspeakers, I listen at 85 dB RMS.
2.83*10^(-9/20) = 1.00Volts RMS
Since music has a crest factor you will want to add headroom to this, anywhere from 5dB to 20+dB
by similar math as above:
Peak Volts with headroom = 1.414*Volts_RMS*10^(dB_headroom/20)
Lets use 10dB for giggles:
1.414*1*10^(1/2)= 4.47 Volts
Watts=4.47^2/3 = 6.66 Watts.
OK Ron, all well and good - but I don't see the point in this context. What is your math telling us that my test is not? What am I missing? Serious question.
I would say this test is a lot more accurate than SPL as you have precise measurement tool that are already calibrated and you don't have to add headroom as that is already done by setting the volume knob to the loudest you would want to play.
And volt to SPL ratio is not that proportional meaning you have to add a bit for power compression and saturation leaving you with a lot of guesswork if you are to calculate volt from the amplifier by SPL alone.
As a side note, the PA amp for my subs have signal indicators which shows when the level is above 2V RMS. You feel the bass before they light up.
And volt to SPL ratio is not that proportional meaning you have to add a bit for power compression and saturation leaving you with a lot of guesswork if you are to calculate volt from the amplifier by SPL alone.
As a side note, the PA amp for my subs have signal indicators which shows when the level is above 2V RMS. You feel the bass before they light up.
SPL tells how loud you are listening and helps make sense of the voltage reading (when sensitivity is known). As the math shows, input voltage depends on SPL and sensitivity so there are two sources of variability in the voltage results.
I think your poll is fine, but knowing either the sensitivity or the ultimate SPL allows you to use the voltage as a comparison number to make sense of the numbers. Mat's electrostats are a clear degenerate case that proves my point.
Oh, and voltmeters are average or RMS responding so the actual peak voltage is 1.414 times the reading on sine waves. This is built in to my calculation example.
Correct - and it is also built into my calculations, as explained in the first posts.
Knowing your SPL is valuable, but it's a different test than this one.
Mat's electrostats are a special case, as they seem to be driven by a special high voltage amp. His results don't relate to users of more typical systems, but are certainly important for him to know. Using a commercial sound 70V or 100V system would be similar. I've run Quad electrostatic panels that did not require special amps and the voltages needed were surprisingly low.
Knowing your SPL is valuable, but it's a different test than this one.
Mat's electrostats are a special case, as they seem to be driven by a special high voltage amp. His results don't relate to users of more typical systems, but are certainly important for him to know. Using a commercial sound 70V or 100V system would be similar. I've run Quad electrostatic panels that did not require special amps and the voltages needed were surprisingly low.
Yesterday I posted my results with my desktop 87 spl speakers at 1m, 25 watt amp driven by autoformer VCs. I measure 2.14 volts. I am really surprised I am using so little power and only slightly more voltage than my 2v dac.
Pano I hope you will allow me to take this a little OT with this practical issue. The avc's will attenuate a maximum of 41 db and on rare occasions I am at the max. I have my eye on new speakers that are 94 spl so something has to give. I know when I did the test I was attenuating 25db and I wanted to know the amp gain. Since the voltage from the dac and amp are so close the amp gain is about 26bd (calculated 25.8db).
The question is if I were to get a amp with a gain of 20 and 7 db more efficient speakers would I be attenuating at almost the same place? Is this true even though the new amp is only 4 watts max?
Pano I hope you will allow me to take this a little OT with this practical issue. The avc's will attenuate a maximum of 41 db and on rare occasions I am at the max. I have my eye on new speakers that are 94 spl so something has to give. I know when I did the test I was attenuating 25db and I wanted to know the amp gain. Since the voltage from the dac and amp are so close the amp gain is about 26bd (calculated 25.8db).
The question is if I were to get a amp with a gain of 20 and 7 db more efficient speakers would I be attenuating at almost the same place? Is this true even though the new amp is only 4 watts max?
7dB more sensitvity and 7dB less gain in the new arrangement will result in exactly the same volume setting of the vol pot.
But a different frequency response from the new arrangement may result in a slight "adjustment" in overall loudness to get a similar "effect".
7dB less power will be needed by the speakers. 24W less 7dB = 4.79W
Your 4W is a good estimate for equivalence. You would lose only 0.8dB of your overhead for peaks.
But a different frequency response from the new arrangement may result in a slight "adjustment" in overall loudness to get a similar "effect".
7dB less power will be needed by the speakers. 24W less 7dB = 4.79W
Your 4W is a good estimate for equivalence. You would lose only 0.8dB of your overhead for peaks.
Yes indeed! The frequency and power response will probably lead you to listen at a different setting, even tho the overall gain is the same.
But you'll be close.
You got it right.