Fine details

January 30, 2017
 by Paul McGowan

Champagne has a pronounced taste not found in white wine, yet the difference is only a dose of CO².

A motional feedback corrected woofer is a significant upgrade, yet the difference is only a small bit of circuitry.

A homemade waffle with expertly browned butter-infused maple syrup is an extraordinary experience. A Kellog Eggo waffle…not so much.

It is often the fine details that differentiate good from great.

It's easy to overlook the details and focus on the big stuff, but it's often a mistake.

Subscribe to Paul's Posts

13 comments on “Fine details”

  1. When it comes to Champagne and other wines using the methode Champagenoise there's more too it. After the first fermentation sugar and yeast are added to the liquid and it is put under pressure to keep the CO2 in for the second fermentation. This process alters the flavor considerably. As with all wines, the grapes used, the conditions of storage, and many other factors alter the flavor. These are not minor details.

    Around 1964 I ate the best waffle I ever tasted. It was a Belgian waffle at the NYC World's Fair, dark, crunchy and covered with powdered sugar, whipped cream and strawberries. Fifty years later I still remember that taste. It was not a fine detail, it was a different food like the difference between a Big Mac and a beautifully grilled New York strip steak.

    Having finally figured out how a woofer/enclosure actually works, it was easy to see that the acoustic suspension design when correctly implemented is the best design. The answer comes from Newton's second law of motion applied to forced oscillation which is explained in every first year college physics book. Apply this correctly and you can have a speaker with any frequency response and well under 5% distortion, even 1% distortion without servo feedback. The reduction probably came when AR switched from Alnico magnets to ferrite magnets. There is a lot more to servo feedback than a few parts. There is the transducer that has to be carefully selected and mounted. There is one or two stages of differentiation or integration if the transducer is an accelerometer or velocimeter. And then there is adjusting parameters to prevent the system from becoming unstable, an oscillator. Even with all of this, additional equalization is usually necessary to compensate for room acoustics and each recording. Servo feedback cannot compensate for other factors like cone breakup.

    Bascom King's explanation of how a servo woofer works showed that the open loop gain of the IRS woofer design was less than optimal with a resonant peak at 60 hz. Below this frequency output falls off at a linear 12 db per octave for an acoustic suspension design. AR1 and its derivatives have a system resonance of 42 hz and Teledyne AR9 is at 28 hz. The latter reduces amplifier power requirements by a factor of over 10. The two woofers are also optimally located at the floor wall junction where the most room reinforcement occurs. By comparison, the servo system is far more complex, far more expensive, and offers little or no advantage.

    1. Are you referring to acoustic suspension particularly or to all closed boxes since there is a difference with acoustic suspension being a proper subset of closed box, a not minor detail which seems to be totally forgotten even by reviewers these days? This is almost amazing for a design that in the 60s probably controlled over 2/3 of the audio market. Today closed boxes are very rare. Off the top of my head I think of Magico, YG and ATC, none of which sell great quantities and I suspect they are closed boxes and not acoustic suspension.

      And I suspect the audio community is now so used to reflex sound that that's what they think bass sounds like. A good closed box sounds too dry for today's listeners. Even a well designed reflex box has a resonant affect that listeners find rich and compelling. Even I find it has an alluring richness and I use large closed boxes for bass.

      1. I am talking about acoustic suspension speakers specifically, the concept patented by Edgar Villchur. The design is more ingenious than you'd first think. You can design it for any Frequency Response you want. Any F3, any Q. The approximate solution to the equation f(t) = m*d2x/dt2 + b*dx/dt + k*x is given in every college physics text. The beauty of this design is that unlike all other designs, b and k are not functions of frequency. This is due to the Ideal gas law that says P1*V1 = P2*V2 and Force = P * A where A is area and P is pressure. K is proportional to the difference between the external and internal pressure and what is more it is applied uniformly over the entire surface of the cone. That means that unlike mechanical suspensions there is no difference in the circumferential restoring for at either the inner or outer suspension and there is no difference in force belong any radial line between the inner and outer suspension. This eliminates the shearing and twisting forces that cause woofer cones to double creating harmonic distortion. b is the velocity related loss that controls Q. It is caused by the aerodynamic drag of air being pushed and pulled between the fibers of the stuffing. The cones themselves are usually heavier and therefore stronger than cones for other designs too. This is to get the free air resonance low.

        1. Thanks for the reply. I learned something today. I knew what the difference between acoustic suspension and closed box was but other than a smaller box I didn't realize some of the other advantages of acoustic suspension were and that they were meaningful. It makes me even more interested in speakers like Magico where they describe their bass loading as acoustic suspension. I'm even more interested if they are using acoustic suspension and closed box as the same thing.

            1. Is it acoustic suspension? All closed boxes are not acoustic suspension although all acoustic suspension boxes are closed boxes. Yours may be. Acoustic suspension means the air in the box is the main force restoring the driver to center, rather than the spider. And in an overly small box the air pressure is significant. But your box is so small it requires equalization to achieve it's transfer function and conventional closed boxes do not usually use equalization(electronic) that is. The audio world has forgotten what acoustic suspension is and most audiophiles, including reviewers think closed box and acoustic suspension are the same thing. They are not.

              1. I believe it is... the box is a large 160Lb object with no vents, a tough 13 inch speaker and I think a servo feedback system as well. J L Audio makes some of the best.

              2. Acoustic suspension speakers fall off linearly at 12 db per octave below resonance. Ported speakers fall off at 24 db per octave. This makes acoustic suspension speakers equalizable to extend their bass response while the resonant frequency of a ported speaker is its lower limit. This principle has been used to good effect by Bose and McIntosh. The original Bose 901 speaker and series II were acoustic suspension designs. They used an equalizer to extend bass response. However there were some problems. The resonance frequency was deliberately pushed up over 180 hz because Bose claimed that phase shift associated with resonance below 180 hz was audible, above 180 hz it wasn't. However, the speaker was underdamped resulting in about a 7 db peak at around 250 hz that was clearly audible. The equalizer only provided 6 db per octave of boost. Therefore there was an annoying peak in the midbass and the lower bass was not audible unless it was played at ear shattering levels. The fix is relatively easy with a graphic equalizer eliminating the peak and extending the deep bass further with with more boost. However, the power demands are enormous and the power capacity of the speaker is only 270 wpc. Three or 4 pairs and available amplifier power between 600 wpc and 1100 wpc should do the trick of fixing it. It is very easy to clip my 138 wpc Marantz receiver with just one pair unless the 10 db shelf filter below 40 hz is engaged. It also doesn't take much amplifier gain to cause acoustic feedback in my well suspended turntable without the 10 db cut. Within its power limits, it can produce very deep bass down to about 23 hz according to Julian Hirsch's measurements. It produced about 10% THD which isn't bad at all. AR3a of that era produced about 5% THD. Later variants of the AR 12" woofer produced only 1% THD. This is why a servo system for such a speaker is not necessary. BTW, except for the most recent variants of IRS which is Genesis 1.2 Dragon or something like it, only one speaker per channel has an accelerometer for servo control. The newest and best model has 2 per channel. It is assumed that all of the drivers are sufficiently similar that they will all behave the same way so that only one is necessary per channel.

          1. A large closed box speaker can be similar to an infinite baffle speaker. Large Bozak speakers are an example. The construction of the driver is entirely different. The infinite baffle speaker has a stiff suspension which supplies the restoring force. An acoustic suspension speaker has a very loose compliant suspension. Its restoring force is supplied mostly by the difference in air pressure between the inside and outside of the box. If a true acoustic suspension speaker is operated in an unsealed enclosure or in the open, it may be easily damaged by over-driving it since there is no restoring force to limit it to the end of its travel. In reality the drivers do have some mechanical restoring force. They should not be 100% air tight. If they are then changes in ambient air pressure will make them work like a manometer moving the "neutral" zero voltage point in or out. This is what the relatively low mechanical restoring force overcomes.

            In ported speakers, air travels easily through the port at the frequency it is tuned to and at multiples of that frequency and not easily at all halfway in between. K is a strong function of frequency.This is why they invariably have a high Q.Little back pressure on the cone at the tuned frequencies, much back pressure at midway points between. The turbulence at the point where the port enters and exits the tuning tube causes the chuffing sound characteristic of ported designs. At that point b becomes a function of frequency as well. Flaring the port helps some.

  2. Hmm waffles;-) I have been living on toaster waffles, but I do use real butter and maple syrup.

    Maybe I should start a GoFundMe page. Help feed a poor audiophile?

    I have my ports plugged on my Legacy Focus speakers, I like the bass I get.

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301
1-800-PSAUDIO

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram