A horn throat distortion question, and eficiency Vs dynamics?

[mixsit]

I'm trying to get a few basics straight, and this seems like a good place to ask.

I have read that at some SP level, the air in a horn throat can distort. Does this happen (even if only on transients on un-compressed signals) at or near the 'loud end' of studio monitoring? (Let's say just to get in the ballpark loud is 100db spl on the average scale and peaks are +8 or 10 above.)

Second-..

Does driver efficiency have an effect on the relative dynamics of the output sound pressure level? This would assume that the systems, a high efficiency and a lower in comparison are both operating within their linear range (not being drive into 'power compression', if that is the applicable term).
I think the answer is no, but would like to be sure, and comes from hearing people say some system might sound 'more dynamic' than others.
And last, it has been said that a cone design may have an edge (potentially?) in the accuracy department over horns.
(It's new studio monitor time.
Thanks in advance.
Wayne Smith

[Robh3606]

For one you need to look in the JBL tech library.

http://www.jblpro.com/pub/technote/tn_v1n21.pdf

Might help.

On the second I think the High Efficiency system has the edge all the time. Doesn't matter if they are both in their linear ranges.

On Three After reading the first reference that may not be true with more modern designs.

Rob

[whizzer]

Efficiency and dynamic range are inextricably intertwined. "Dynamic range" is the term used to describe the difference between the softest audible sound and the loudest. Many types of live music exhibit an actual dynamic range of 20+ decibels, a range of approximately 128:1, while many so-called "audiophile" loudspeaker systems are so inefficient that their rated SPL is in the 85-86 decibel range, measured on axis at meter with a one-watt input of pink noise. A 3 decibel increase in perceived output requires an increase of 10X amplifier power, so raising the inefficient system's output 3 dB to 88 dB requires 10 watts, and raising it to 91 dB requires 100 watts, and to 94, 1000 watts. Unfortunately, although many speakers can "take" that kind of input, they simply convert the electricity to heat. The actual dynamic range of such systems is, if we're feeling generous, 9 dB. In order to reproduce the actual dynamic range of real music, a speaker system must be very efficient indeed, and many horn systems have efficiency ratings approaching 75% as compared to the 3% or so characteristic of most "high end" audiophile systems. Compression distortion is a measure of quality rarely mentioned in "audiophile" equipment reviews, and for good reason, as many highly-touted systems have compresiion distortions of over 50%. Horn systems may present many practical problems, but high distortion is not a necessary part of what horns do.

As for the other part of your question, very narrow throats do not actually "distort" the air itself, which retains its molecular integrity, but the high velocities produced by large compression drivers pushed by excessive input signals may cause turbulence in constricted throats that serves the same purpose--physical distortion of the reproduced signal.

[yggdrasil]

A 3 decibel increase in perceived output requires an increase of 10X amplifier power, so raising the inefficient system's output 3 dB to 88 dB requires 10 watts, and raising it to 91 dB requires 100 watts, and to 94, 1000 watts.

No.

A 3 dB increase requires 2x amplifier power. A 10 dB increase requires 10x amplifier power.

Raising output from 85dB to 95dB when the speaker has a sensitivity of 85dB requires 10 watts.

[johnaec]

A 3 dB increase requires 2x amplifier power. A 10 dB increase requires 10x amplifier power.

Raising output from 85dB to 95dB when the speaker has a sensitivity of 85dB requires 10 watts.

That's my understanding also. And in an open environment, every doubling of distance from the loudspeaker results in -6 dB, correct?

John

[whizzer]

Let me clarify this. Insofar as pure electrical power levels are concerned, a 3 dB increase in power is a ratio of 2:1, however, for the hopefully undamaged human ear, a 3 dB increase in applied amplifier power is hardly audible. For a given sound to be perceived as "twice as loud," an acoustic increase of 3 dB (remember, dB are not absolute, but relative measures of power; an acoustic watt is a very large unit as compared to an electrical watt), a bare minimum of 10 times the amplifier power must be applied. The problem with inefficient speakers is that they soak up too much amplifier power just to attain a reasonable average level, leaving little amplifier headroom for cleanly reproducing the transients that give the music its characteristic "flavor." The more efficient the system, the wider its dynamic range. Psycho-acoustics deal with the complex interplay of human perception and the far more quantifiable realm of "physical" physics. The problem in the bass region is even more acute, which is why loudspeaker systems that measure harmonically flat across the range of human hearing produce an output that strikes most listeners as overly "bright," perhaps even "harsh," and that hardly ever bears a favorable comparison to the far more visceral experience of actual live music. Inefficient loudspeakers are simply physically incapable of faithfully reproducing the actual dynamic range of many types of live music, particularly rock and pop, but also including Baroque organ fugues and the like. Measure a system that produces 85-87 dB @ 1 watt @ 1 meter at a realistic listening distance in an auditorium-sized listening environment at its full rated input power and compare the experience of listening to it in that environment to a horn-loaded system rated at over 100 dB @ 1 watt @ 1 meter running along on only a few watts so as to allow plenty of clean amplifier power for transients 15-20 dB above the average level, and the differences will very soon become very apparent.

[johnaec]

For a given sound to be perceived as "twice as loud," an acoustic increase of 3 dB (remember, dB are not absolute, but relative measures of power; an acoustic watt is a very large unit as compared to an electrical watt), a bare minimum of 10 times the amplifier power must be applied..

I was always under the impression that for a sound to be perceived twice as loud it took an increase of 10 dB. I was also under the impression that same 10 dB requires 10 times the power. I fail to see how 10 times the power is required for a 3 dB increase in acoustic output.

John

[Oldmics]

johnaec asks

"That's my understanding also. And in an open environment, every doubling of distance from the loudspeaker results in -6 dB, correct?"

Correct for SPHERICAL radiating enclosures.That would be the inverse square law.

Line array systems only lose 3db for the doubling of the distance generally speaking.

Oldmics

[johnaec]

johnaec asks

"That's my understanding also. And in an open environment, every doubling of distance from the loudspeaker results in -6 dB, correct?"

Correct for SPHERICAL radiating enclosures.That would be the inverse square law.

Line array systems only lose 3db for the doubling of the distance generally speaking.

Oldmics

Thanks for clearing that up - that's good to know.

John

[Tom Loizeaux]

I don't understand why you say speakers are limited in their ability to produce dynamic range. Any speaker/driver has a maximum output level...(before destruction), but they can play very softly before they reach the threshold of "not making any sound". That is dynamic range. I think speakers/drivers have a greater dynamic range then voice and most musical instruments!
Maybe I'm missing something, but I just don't get the limited dynamic rage mentioned in the thread.

Tom

[B&KMan]

power amp rule...

3 db higer require= double real power amp. but subjectively 10 dB is require for you feel double power sound...


so :

if 86 dB at 1 m for 1 watts (ex)

89 @ 2 W
92 @ 4 W
95 @ 8 W
98 @ 16 W
101 @ 32 W
104 @ 64 W
107 @ 128 W
110 @ 256 W
113 @ 512 W
116 @ 1024W

note rms is a average... no peak to peak so add easely 6 db for correct peak to peak. the real formula result is the rms = 0.717 of 1.0 (peak)

unfortunaletly, by the nature of transient, the handle power amplifier run at max 1/3 of power requirement of the rule below for full impulse response.

so add 6 db for peak and multiplie by 3 for full dynamic range evaluation with no compression, no distortion...

my 2 cents commentary...




and in same in others reply;


the power sound the rule is calculate arround concept of intensity, and the power is drop 6 dB by double distance...


so finaly , re-calculate your power requirement in regard of this distance...

[Ian Mackenzie]

I'm trying to get a few basics straight, and this seems like a good place to ask.

I have read that at some SP level, the air in a horn throat can distort. Does this happen (even if only on transients on un-compressed signals) at or near the 'loud end' of studio monitoring? (Let's say just to get in the ballpark loud is 100db spl on the average scale and peaks are +8 or 10 above.)

Second-..

Does driver efficiency have an effect on the relative dynamics of the output sound pressure level? This would assume that the systems, a high efficiency and a lower in comparison are both operating within their linear range (not being drive into 'power compression', if that is the applicable term).
I think the answer is no, but would like to be sure, and comes from hearing people say some system might sound 'more dynamic' than others.
And last, it has been said that a cone design may have an edge (potentially?) in the accuracy department over horns.
(It's new studio monitor time.
Thanks in advance.
Wayne Smith

Wayne,

Part one depends on a lot of variables inc the horn contour/flare rate.

Part two. In my opinion it does as more sensitive (this is the correct term for expressing driver output versus input voltage) systems tend to suffer less from power compression (when designed correctly) simple because less power is absorbed as heat in the voice coil.

The best way to improve on linear efficiency is horn loading or use of mutiple drivers or both. This is done in pro applications for precisely that reason.

However, correctly designed cone direct radiators can also be relatively sensitive (98- 100 db) with hi power handling and when used in arrays can become even more so.

Research by Passlabs suggests hi sensitivity cone drivers offer improve dynamics, particularly when used in a bi amplified or tri/quad amplifed system.
http://www.passlabs.com/downloads/rushmore_lit.pdf

[whizzer]

This is true. As mentioned before, decibels are measures of relative power levels, so doubling the input power from 1 watt to 2 watts is a 3 dB increase, and so is the difference between 250 and 500 watts. However, doubling the input power does not double the acoustic energy produced by the loudspeaker. Adding a second identical system also receiving the same level of input signal would produce an actual doubling of energy--more in cases of acoustic coupling, as when assymetric folded horn mouths are arranged in mirror-imaged pairs. Simply applying a 3 dB increase in input power will not produce this magnitude of effect.
Nor are decibels absolute measures of sound level. John L. Murphy, of the True Audio Loudspeaker Design Center, states that if a loudspeaker were 100% efficient, an electrical input of 1 watt would produce an output of 1 acoustic watt, which, measured at 1 meter, would be 112.1 dB SPL (radiating into half space). The question ought to be, "112.1 dB above what?" The baseline level cannot be silence, for any multiple of no sound is still no sound. If we use the measurement of one loudspeaker's output that is produced by a 1 watt electrical input and, for instance, call it 85 dB, then how many times can we double the acoustic energy produced by that loudspeaker before it becomes burned out junk? If, on the other hand, if we measure the output of a system the same way, using the same equipment, and find a reading of 95 dB, we may double that acoustic energy perhaps the same number of times before destroying the transducer, but the ultimate output will be much greater. Now, obviously, either system will play softly enough to produce sounds that are barely audible, but the more efficient one will play far more loudly, thus its dynamic range is wider. Finally, listening to music at the same average level, the more efficient system with the wider dynamic range will, by allowing the amplifier greater reserves for reproducing transient peaks, simply sound "more dynamic."

[Mr. Widget]

Finally, listening to music at the same average level, the more efficient system with the wider dynamic range will, by allowing the amplifier greater reserves for reproducing transient peaks, simply sound "more dynamic."

I won't dispute this statement, but I am wondering if what we determine as "sounding more dynamic" isn't more correctly a freedom from power compression? Perhaps also in conjunction with a better transient response.

Typically high efficiency systems have low moving masses and have better transient response and since they aren't being driven as hard in a domestic environment, they typically don't heat up very much and therefore are inherently less prone to dynamic compression. Lately drivers being designed by JBL and others have been tackling dynamic compression head on and modern drivers have much less dynamic compression than previous models. The fact that an older design that was originally intended for theater use is barely being driven in a domestic situation makes it's susceptibility to dynamic compression much less apparent and therefore it's performance is seemingly more effortless.


Widget

[whizzer]

[QUOTE=Mr. Widget] I am wondering if what we determine as "sounding more dynamic" isn't more correctly a freedom from power compression?


It certainly is. A mechanical and/or electrically-induced restriction of dynamic range is exactly what "power compression" is. While efficiency of a high order will allow a wider dynamic range by allowing for sufficient amplifier headroom, it by no means ensures it. The driver's motor assembly must be capable of controlling diaphragm movement in such a way as to cleanly and accurately respond to the signal and reproduce those transient responses. One knock against very efficient speakers has been that they don't sound so good at low levels, principally because they draw so little power, for it is at low output levels that transistorized amplifiers are at their worst. Old-style theater speakers truly sound better with low-power tube amps for this very reason. The effort made by JBL to preserve the old high-efficiency approach while using modern materials and design techniques for greater compatibility with modern amplifier technology has, indeed, resulted in lower levels of power compression than seen in many other modern drivers--many examples from Eminence spring readily to mind--but at the cost of some small decrease in ultimate dynamic range compared to earlier designs.