Sorry Jerome, I don't understand what you mean.
Your measurement of 12.6V RMS on the -12dB tone means that you need 4X 12.6V for the peaks. Or just over 50 volts. It seems to me that you would need to bring your average level down by 6dB to avoid clipping. No?
To me, an amp that is rated 50WPC @8 ohms means that it can deliver 20V RMS on a sine wave before clipping. That would be 28.3V peak. It won't deliver more voltage into a 4 ohm load, it will be the same, or maybe a little less.
Your measurement of 12.6V RMS on the -12dB tone means that you need 4X 12.6V for the peaks. Or just over 50 volts. It seems to me that you would need to bring your average level down by 6dB to avoid clipping. No?
Yes you are right.
It is my own amp, +-35V voltage. The voltage doesn't change but the power yes, and my amp can supply this power : 100W/4ohms.
My music is not at 0dB level ! I put my favorite music at maximum level i can listen to it and after i measured the test tone. I think it is the explanation of this difference, unless i didn't understand the procedure
.
Ok i reach the limit of my amp, the result should be 7V (6db less) ?
BTW the OB could provide (80 + 20) 100dB SPL
It is my own amp, +-35V voltage. The voltage doesn't change but the power yes, and my amp can supply this power : 100W/4ohms.
My music is not at 0dB level ! I put my favorite music at maximum level i can listen to it and after i measured the test tone. I think it is the explanation of this difference, unless i didn't understand the procedure
. Ok i reach the limit of my amp, the result should be 7V (6db less) ?
BTW the OB could provide (80 + 20) 100dB SPL
I voted as 5-10 V
I normally listen to contemporary music at about 4.5 V but sometimes go 2 or 3 dB louder. I listen to Classical at higher settings, with the highest setting corresponding to 13 V. This is at my limit and rarely done.
DIY Speakers - SEAS ER18RNX (6.5”) 8 Ohm Mid-Woofer [Characteristic Sensitivity 88.5 dB (2.83V, 1m)]
240 Hz test signal was used as the XO to SWs is at 100 Hz and there is EQ boost impacting 120 Hz range.
210 m^3 (7400 ft^3) room with LP at 4 m.
I normally listen to contemporary music at about 4.5 V but sometimes go 2 or 3 dB louder. I listen to Classical at higher settings, with the highest setting corresponding to 13 V. This is at my limit and rarely done.
DIY Speakers - SEAS ER18RNX (6.5”) 8 Ohm Mid-Woofer [Characteristic Sensitivity 88.5 dB (2.83V, 1m)]
240 Hz test signal was used as the XO to SWs is at 100 Hz and there is EQ boost impacting 120 Hz range.
210 m^3 (7400 ft^3) room with LP at 4 m.
Thanks JT. That's a nice big room with average sensitivity speakers, very nice to know. Your contemporary vs classical also seems right in line. Classical recording can be very dynamic. In your case about 9dB more dynamic.
When you measure the -12dB test tone, music peaks will be 12dB (4X) higher when playing at that same volume setting. That's how digital recordings work. -12dB will always be 1/4 the voltage of 0dB. 0dB is is the highest possible level.
Now; you could set your volume to its "normal" setting and measure there, but that would not tell you how much voltage you need to play it loud. As loud as you ever do. We are looking for the most you ever use.
I choose -12dB for several reasons, one being that it isn't likely to blast out your ears or speakers, and not make the neighbors call the cops.
Another reason is because it's easy to do the 4X multiplication in your head to know the absolute peak voltage.
Maybe. Let me see if I can explain again.
When you measure the -12dB test tone, music peaks will be 12dB (4X) higher when playing at that same volume setting. That's how digital recordings work. -12dB will always be 1/4 the voltage of 0dB. 0dB is is the highest possible level.
Now; you could set your volume to its "normal" setting and measure there, but that would not tell you how much voltage you need to play it loud. As loud as you ever do. We are looking for the most you ever use.
I choose -12dB for several reasons, one being that it isn't likely to blast out your ears or speakers, and not make the neighbors call the cops.
Another reason is because it's easy to do the 4X multiplication in your head to know the absolute peak voltage.
I measured ~3.05 volts with receiver @0dB (movie reference setup by Audigy on Denon 3808)
Home built RD75 open baffle w/woofers, no subs @120 Hz
This in a 25' x 40' x 2 story room with lots of furniture. At 15', the spl measures ~ 82 - 85 dB, depending on room modes.
They'll go alot louder, but wife is home...
Home built RD75 open baffle w/woofers, no subs @120 Hz
This in a 25' x 40' x 2 story room with lots of furniture. At 15', the spl measures ~ 82 - 85 dB, depending on room modes.
They'll go alot louder, but wife is home...
Last edited:
yep... back in '05 (scary that phrase... back in '05
) when I built them...They do tend to morph over time, tho...
John L.
Cheap Rat-Shack pocket DMM. On the first measurement, I wasn't thinking and just started the tone up and them dropped the probes across the terminals on the back of the loudspeaker. Amp went into oscillation and I quickly shut things down. The next time I just decided to err on the side of conservancy and after setting the level I disconnected both speakers, hooked up the DMM and *then* ran the test tones.
I was more concerned with the reverse case - if the music I selected was recorded at too low a level and not normalized during the MP3 conversion then even the "peak" music may still be be 12+ dB down, in which case your test tones would return a higher voltage than needed. In essence, if your recorded material is too soft you would have the volume/gain turned up higher than it would have been for a full-range piece of music.
In my case, I don't think this is a big concern - both recordings I used were well mastered and well known for using all the available dynamic range. But others mileage may vary.
Thanks again,
-bill
more #'s
Ok... I measured at ~12 feet at the location of loudest room mode.and @ ~3 feet
@ 12', volume measured 96dB, voltage was 11.7V. @ 3', the level measured ~101dB
Background levels ~ 45dB all measurements C weighted on Phonic spectrum analyzer set to All frequencies
Room modes are all over the map @this frequency, varies from 87dB at null to the 96dB above
Didn't enter the results in chart, but feel free to if u want.
John L.
Ok... I measured at ~12 feet at the location of loudest room mode.and @ ~3 feet
@ 12', volume measured 96dB, voltage was 11.7V. @ 3', the level measured ~101dB
Background levels ~ 45dB all measurements C weighted on Phonic spectrum analyzer set to All frequencies
Room modes are all over the map @this frequency, varies from 87dB at null to the 96dB above
Didn't enter the results in chart, but feel free to if u want.
John L.
Yes, picking the music to set your "loud level" is important.
With wide dynamic range recordings like classical, you will end up cranking the volume knob more to get it loud enough. Once you do, your peaks will be that much louder. Of course, so will your measurement of the -12dB signal we be correspondingly higher. But you'd want to know that, if you play much wide dynamic range music.
That's why I suggest playing a few different tracks and different genres. That will give you a better idea of were "loud enough" is on your volume dial.
Looks like we've picked up a couple more votes. One is from the open baffle project here:
http://www.diyaudio.com/forums/multi-way/110583-fast-fun-inexpensive-ob-project-46.html#post2752617
About 5V.
http://www.diyaudio.com/forums/multi-way/110583-fast-fun-inexpensive-ob-project-46.html#post2752617
About 5V.
More Power!
The rallying cry of the amp heads. How much do power you need? I promised to talk about power.
This is a voltage test because it's a level playing field (impedance isn't taken into account) and because power is a macho, emotional issue. You don't hear many guys brag about how much voltage their amp puts out.
So how do your test results relate to amplifier power? Here's a little trick for you:
For example:
Of course you may like the sound of amp amp with more power, or even with less power, but that number will allow you to look at your measurement results and know how much amp is "enough." It's simple, easy and informative. Most people have no idea how to size an amp for a system. This gives you a pretty close guide. If you wanted to know the 4 Ohm amp power rating you need, just double your result.
How does the "square the voltage" trick work? I'll let you figure it out, but here are two clues:
Looking at the poll results above, 78% of those who responded can play "loud enough" with an amp of 25 watts or less. 43% could do with 5 watts. That's pretty surprising - at least I didn't expect it.
The rallying cry of the amp heads. How much do power you need? I promised to talk about power.
This is a voltage test because it's a level playing field (impedance isn't taken into account) and because power is a macho, emotional issue. You don't hear many guys brag about how much voltage their amp puts out.
So how do your test results relate to amplifier power? Here's a little trick for you:
- Take your voltage measurement from the test tone
- Square that number
- The result is the power rating (@8 ohms) that you need for your amp not to clip.
For example:
- You play the test tone and measure 3.5V AC
- 3.5x3.5=12.25
- You need an amp rated at 12.25 watts (minimum) to play your "loud enough" level and not clip.
Of course you may like the sound of amp amp with more power, or even with less power, but that number will allow you to look at your measurement results and know how much amp is "enough." It's simple, easy and informative. Most people have no idea how to size an amp for a system. This gives you a pretty close guide. If you wanted to know the 4 Ohm amp power rating you need, just double your result.
How does the "square the voltage" trick work? I'll let you figure it out, but here are two clues:
- The test signal is recorded at -12dBFS (12dB RMS below Full Scale)
- The highest a sine wave can be recorded without clipping is -3dB RMS
Looking at the poll results above, 78% of those who responded can play "loud enough" with an amp of 25 watts or less. 43% could do with 5 watts. That's pretty surprising - at least I didn't expect it.
A long time ago one of my mentors (I forget who) told me to figure out how many watts were needed to get the peak output needed, then get an amp that could output that voltage at 60% output. This has to do with staying in the linier region of the amps output.
Haven't given it much thought lately but it has served me well.
I wonder about how this would apply to different types of amps today.
edit: After looking at the distortion graphs of some of my newer amps, they seem quite a bit better than older pro amps to rated power. I'll keep the overhead rule anyway.
Haven't given it much thought lately but it has served me well.
I wonder about how this would apply to different types of amps today.
edit: After looking at the distortion graphs of some of my newer amps, they seem quite a bit better than older pro amps to rated power. I'll keep the overhead rule anyway.
Last edited:
Your hope is good, you're wrong!
You do not need to divide by the impedance in this case, that's the "trick" of setting the level at -12dB.
Look at the clues, do the math - you'll figure it out. Think about sine waves and an amplifier's power rating.
I'll step thru the answer tomorrow, if no one figures it out before then..
You do not need to divide by the impedance in this case, that's the "trick" of setting the level at -12dB.
Look at the clues, do the math - you'll figure it out. Think about sine waves and an amplifier's power rating.
I'll step thru the answer tomorrow, if no one figures it out before then..