SugarMedia
Well-known member
Interesting read. See the attached .pdf file.
Souce: Musical Fidelity
Souce: Musical Fidelity
Dynamic Range and Power Claims
- Thread starter SugarMedia
- Start date
Help Support MartinLogan Audio Owners Forum:
That is an interesting read. What they are saying makes sense to me, but I am no electrical engineer so am not in a position to challenge it. It does fit nicely into some of the discussions we have occasionally about dynamic headroom in an amplifier and the need for higher wattage to obtain true to life dynamics. I have always been a proponent of going with higher wattage to have enough power for dynamic transients.
I know some on this forum have expressed a pretty strong feeling that lower wattage amps sound better, all other things being equal. I wonder if that better sound isn't really the result of a sonically pleasing but nonetheless inaccurate amount of even-order harmonic distortion caused by soft clipping. Or is it really that a simpler circuit just sounds "better" than a more complex circuit, even if it can't reproduce dynamic transients as accurately. I'm sure opinions will vary dramatically on this issue, and ultimately all that matters is that you like the sound of the amp combined with your speakers.
I know some on this forum have expressed a pretty strong feeling that lower wattage amps sound better, all other things being equal. I wonder if that better sound isn't really the result of a sonically pleasing but nonetheless inaccurate amount of even-order harmonic distortion caused by soft clipping. Or is it really that a simpler circuit just sounds "better" than a more complex circuit, even if it can't reproduce dynamic transients as accurately. I'm sure opinions will vary dramatically on this issue, and ultimately all that matters is that you like the sound of the amp combined with your speakers.
Audiophiliac
Well-known member
Depends on how loud you want to listen to your tunes. 
Depends on how loud you want to listen to your tunes.
Yes, I think that is part of it. I personally like to listen pretty loud. I find that (within reason) the more I turn up the volume the better the music sounds on my Logans. Especially if I am trying to recreate the feel of a live event, there is a need for a certain dbl level. Obviously, another factor is the sensitivity of the speaker. Some speakers just need more watts to achieve the same output level.
Audiophiliac
Well-known member
Sensitivity changes with frequency and impedance and who knows what else. And it depends on what live event you are trying to reproduce. A simple acoustic folk set in a small location without sound reinforcement can be easily reproduced as far as SPL and dynamics go.
A Metallica concert on the other hand? I tried it....on a MAXX2/Krell FPB700cx setup. I think I got it loud enough...(I had to literally YELL at the guy standing right next to me for him to even acknowledge that I was saying something)....it was getting difficult to breathe normally, and it did feel like the concert I had attended weeks prior (although it sounded much better)......but alas, I guess it was too much for the speakers.....all the resistors on all the midrange drivers went up in smoke (not literally, but they all were dead afterwards)....although I was told it was a quasi-common issue with new Wilsons being played at those levels....not sure how much truth is in that, or if they underestimated how loud I was actually playing it....
Either way, if everyone followed the advice of that .pdf strictly, we would likely all need at least 1500-3000 watt amplifiers. Not that there is anything wrong with that.
EDIT: Now that I think about it, the lowest level at the Metallica concert was certainly no where near 64dB or even probably 85dB....the loudest? Maybe 110, 120? depending on where you were standing/sitting....so the actual dynamic range of the event may have only been roughly 30-35dB? Maybe thats not unrealistic.....Just a thought.
A Metallica concert on the other hand? I tried it....on a MAXX2/Krell FPB700cx setup. I think I got it loud enough...(I had to literally YELL at the guy standing right next to me for him to even acknowledge that I was saying something)....it was getting difficult to breathe normally, and it did feel like the concert I had attended weeks prior (although it sounded much better)......but alas, I guess it was too much for the speakers.....all the resistors on all the midrange drivers went up in smoke (not literally, but they all were dead afterwards)....although I was told it was a quasi-common issue with new Wilsons being played at those levels....not sure how much truth is in that, or if they underestimated how loud I was actually playing it....
Either way, if everyone followed the advice of that .pdf strictly, we would likely all need at least 1500-3000 watt amplifiers. Not that there is anything wrong with that.
EDIT: Now that I think about it, the lowest level at the Metallica concert was certainly no where near 64dB or even probably 85dB....the loudest? Maybe 110, 120? depending on where you were standing/sitting....so the actual dynamic range of the event may have only been roughly 30-35dB? Maybe thats not unrealistic.....Just a thought.
Last edited:
Spike
Well-known member
Interesting read, but...
Ok, let's plug these figures into my setting (room & system) to see what comes out, shall we? My ReQuests are rated at 90db sensitivity and my listening position is at 4 meters away (12 feet). This paper used a nominal listening position of 11 feet, close enough if we're talking about 'approximately' 10db of attenuation. Let's see...
BUT, the 10db drop at 4 meters is the correct figure for...point-source type speaker. Logans are line-source due to the large area of the panel and I confirmed with my measurement of only 4db drop at my listening position 12-feet away. The measurement was made by registering 90db on my SPL meter at 1 meter and then backing off to my listening level to see 86db on my SPL. This was done with only 1 speaker playing white noise from my Meridian 565 processor. Now, we need to rework the same steps with this 4db drop instead of 10db.
Spike
Ok, let's plug these figures into my setting (room & system) to see what comes out, shall we? My ReQuests are rated at 90db sensitivity and my listening position is at 4 meters away (12 feet). This paper used a nominal listening position of 11 feet, close enough if we're talking about 'approximately' 10db of attenuation. Let's see...
- ReQuest sensitivity of 90db - 10db = 80db.
Add back 3db for 2 speakers, we have 83db. - Let's pick 105db as the peak level.
- 105db - 83db = 22db, which equates to 156 watts of power needed.
BUT, this 156 watts of power is needed to drive TWO speakers (remember adding 3db to accomodate for 2 speakers in step 1. So, each channel requires 78 watts of output.
BUT, the 10db drop at 4 meters is the correct figure for...point-source type speaker. Logans are line-source due to the large area of the panel and I confirmed with my measurement of only 4db drop at my listening position 12-feet away. The measurement was made by registering 90db on my SPL meter at 1 meter and then backing off to my listening level to see 86db on my SPL. This was done with only 1 speaker playing white noise from my Meridian 565 processor. Now, we need to rework the same steps with this 4db drop instead of 10db.
- 90db - 4db + 3db (for 2 speakers) = 89db.
- Still using 105db as peak level.
- 105db - 89db = 16db, equating to 50 watts total for BOTH channels. So, each channel requires 25 watts.
Spike
Spike, well written analysis.
But I think we also need to consider that amps are rated at 8 ohms, and the impedance curve of our ML’s are all over the place. Specifically, they drop to a couple of Ohms (or lower) towards the higher frequencies.
So your calculations might be accurate at midrange levels or for the bass, but I bet they rise dramatically towards the upper registers.
This is why high-wattage and high-current amps are recommended, as maintaining a similar output level at say 15Khz with a 3 Ohm load will require your amp to deliver more than double the wattage (mostly due to increases in current demand (wattage = voltage x amperage)).
I’ve not had enough coffee to do the math and equations, but my gut says your example would need more on the order of 100 watts or more to do this 20 to 20 on the impeadance curve of a ML.
But you are right to factor the line source effect into the distance SPL drop. It is significantly lower than a point source.
Here is a nifty SPL calculator: http://www.myhometheater.homestead.com/splcalculator.html
But I think we also need to consider that amps are rated at 8 ohms, and the impedance curve of our ML’s are all over the place. Specifically, they drop to a couple of Ohms (or lower) towards the higher frequencies.
So your calculations might be accurate at midrange levels or for the bass, but I bet they rise dramatically towards the upper registers.
This is why high-wattage and high-current amps are recommended, as maintaining a similar output level at say 15Khz with a 3 Ohm load will require your amp to deliver more than double the wattage (mostly due to increases in current demand (wattage = voltage x amperage)).
I’ve not had enough coffee to do the math and equations, but my gut says your example would need more on the order of 100 watts or more to do this 20 to 20 on the impeadance curve of a ML.
But you are right to factor the line source effect into the distance SPL drop. It is significantly lower than a point source.
Here is a nifty SPL calculator: http://www.myhometheater.homestead.com/splcalculator.html
Sensitivity changes with frequency and impedance and who knows what else. And it depends on what live event you are trying to reproduce. A simple acoustic folk set in a small location without sound reinforcement can be easily reproduced as far as SPL and dynamics go.
A Metallica concert on the other hand? I tried it....on a MAXX2/Krell FPB700cx setup. I think I got it loud enough...(I had to literally YELL at the guy standing right next to me for him to even acknowledge that I was saying something)....it was getting difficult to breathe normally, and it did feel like the concert I had attended weeks prior (although it sounded much better)......but alas, I guess it was too much for the speakers.....all the resistors on all the midrange drivers went up in smoke (not literally, but they all were dead afterwards)....although I was told it was a quasi-common issue with new Wilsons being played at those levels....not sure how much truth is in that, or if they underestimated how loud I was actually playing it....
Either way, if everyone followed the advice of that .pdf strictly, we would likely all need at least 1500-3000 watt amplifiers. Not that there is anything wrong with that.
EDIT: Now that I think about it, the lowest level at the Metallica concert was certainly no where near 64dB or even probably 85dB....the loudest? Maybe 110, 120? depending on where you were standing/sitting....so the actual dynamic range of the event may have only been roughly 30-35dB? Maybe thats not unrealistic.....Just a thought.
Metallica may be loud but no way does it represent w-i-d-e dynamic range, besides when was the last time they played in a Acoustcally correct invrioment ?? As you said generally speaking not much dynamic range to most rock concerts, I suppose another reason why more and more old "Rockers" are teaming up with some decent Orchestral accompaniment at their venues.
Spike
Well-known member
JonFo, I completely agree with you that the impedance curve of the Logans are all over the map. That was the point of one of my earlier posts that one needs to consider the power-supplies of the amplifiers and not focusing strictly on the 1 ohm load @ 20Khz spec. Even for a beefy solid-state power amplifier which "double down" with low impedance, the problems lie with the 30+ ohms load around the 1KHz area where the power output is cut down to 1/4 of its rating! This is where the tube amplifiers or the solid-state McIntosh with the Autoformer have the edge over the solid-state amps. The output transformers take care of the impedance conversion so that the amplifying circuitry "sees" a nominally constant impedance load. The trick with these devices is that you have to consider the quality of the output transformers as well as picking the correct output taps to make the amplifier "happy"!
Theoretically, you correct here, but given the nature of music, the percentage of musical data higher than 15Khz is rather low when compared to the rest of the frequency range.
Spike
SteveInNC
Well-known member
This is something that is easy to get wrong intuitively also. If you were to ask, most people probably think intuitively that "mid-range" tones (vocals, guitars, etc.) are somewhere around 4000 hertz, just because it's near the middle of the typical 20-20Khz frequency-range specs of amplifiers. In reality, 4Khz is a pretty high-sounding note. By comparison, a piano "middle C" is about 260hz.
lugano
Well-known member
- Joined
- Nov 18, 2005
- Messages
- 1,054
- Reaction score
- 0
A related interesting read is to be found here:
http://www.smr-home-theatre.org/Power/Power-How-Much.html
and some very interesting tables here:
http://hyperbol.free.fr/Sommaire/Dossiers techniques/Capacite dynamique.htm
(the article is in french, but the tables are easy to understand).
http://www.smr-home-theatre.org/Power/Power-How-Much.html
and some very interesting tables here:
http://hyperbol.free.fr/Sommaire/Dossiers techniques/Capacite dynamique.htm
(the article is in french, but the tables are easy to understand).
