Monthly Archives: August 2010

Measuring Speakers. The Map Is Not The Territory.

Here is something that I believe is very relevant to several of the questions being debated about how to measure speakers to achieve good subjective correlation.

Remember, all these types of measurements, Allison’s, Toole’s, Beck’s, Geddes’, mine, anyone’s, are based on models of system behavior. They are not measurements of fundamental physical quantities that we like to imagine they are.  Thus, any of these measurement sets, can at best only approximate the perceptual response certain types of music, under certain conditions.  Take just one example: Music has both sharp transients, and legato, steady-state elements. Even at a fixed point in space, even in an anechoic chamber, the signal envelope of music or speech has modulation time-constants that span a wide range.

Some musical events are much shorter than typical early room reflections, some are much longer than typical listening room decay times.  Why should it be that any one model of room interaction is “best?”  Even if it is statistically most often correlated somehow with listener preference, that DOES NOT MEAN that it is necessarily the measurement model that best quantifies the reproduction of each and every piece of music or recording method. There simply cannot be one perfect measurement technique. If there was, certainly, there would be fewer design approaches on the market.

This hints at the same dimensional transformation problem I was discussing earlier. Ultimately, its about transform reversibility and information entropy. It can be mathematically shown that there exists an infinite set of distinct and different transducers which can satisfy any given criteria of accuracy, if the number of channels is finite. Thus, there cannot be either one perfect measurement, or one perfect loudspeaker response.

(Edited from a post of mine a few years ago at www.classicspeakerpages.net.)

 

A Link To My Page About Speaker Stuffing.

Here’s a link to my page about speaker stuffing.  All kinds of tech and non-tech info buried in there:

Capacitors. (Place holder, will try to finish this year…)

 
There are few topics of on line discussion where the gulf between audio opinion and audio science is wider than the subject of capacitors.  Audio hobbyists, marketers and the audiophile press line up on one side, audio engineering professionals on the other.  Even the most open-minded professional designers I know are fairly well convinced that capacitor behavior is understood, and that a capacitor’s effect on the signal can be predicted or eliminated, as desired, provided that some basic rules about selection and use are followed. Yet, the last few years have seen the subject of capacitor “sound” rise to be front page news, with all sorts of products being offered to the hobbyist and boutique manufacturer that defy basic physics. While speculation about the reasons for this trend, and for such trends in general, will be left to another time, I’d like to address the reality of capacitor behavior here.  I’ll start with some technical factoids, which I hope will answer many of the more basic questions I see posted about capacitor behavior.  
 
– In a simplistic sense, the capacitor can be thought of as a frequency dependent resistor.  It’s a two-terminal electronic component exhibiting a “resistance” to current flow which drops as the frequency goes up.  At DC, an ideal cap looks like an open circuit, and at infinite frequencies it looks like a short circuit.  
 
– The formal description of an “ideal” capacitor is:  i=C(dV/dt).  In other words, the amount of current passing through a capacitor is determined by the size of the capacitor and the rate of change of the voltage across it.  This can be contrasted with the definition of a resistor: i=RV.  Here the current is fixed only by the resistance and the voltage, not the change in voltage. 
 
– Capacitors, in this way, pass AC signals while blocking DC signals.  They also can be used in combination with other components for creating “filters” to enhance or attenuate certain frequency ranges. 

– The primary rating of a capacitor is “Capacitance,” as measured in Farads, (pF, nF, uF).   While this is the only number necessary to define an ideal cap, real world capacitors have limitations that require additional numbers to quantify.  Thus, there are other “specs” which define the ways in which a particular capacitor deviates from the ideal.

– “Operating Voltage” is a common and important specification.  While the ideal cap works at all voltages, real caps have limits.  

– Equivalent Series Resistance, leakage, inductance and signal dependent factors may need to be considered in certain applications.  

– Parallel capacitors add their values.  50uF in parallel with 50uF gives 100uF.  In this case, the effective voltage rating of the pair is set only by the lower-rated unit.  If the two parallel caps have the same rating, simple: the pair also has this rating.  

– Using capacitors in series reduces the capacitance to below that of either individual unit.  C(total)= C1xC2/(C1+C2).   In this case, the working voltage goes up, but in a complex way.  If the two series caps are identical, then their working voltage doubles compared to each individual cap.  If they are not identical, it gets complicated, and depends on the exact values and the working frequencies.  In any event, the working voltage of two series caps is greater than the lower of the pair.  

– Capacitors can be made out of many materials, in constructed in many ways.  Each has advantages, and the selection of caps for a particular circuit application is one of the basic skills an engineer is taught.  

OK, now we can jump into more detail on the technical aspects of capacitor performance. 

Capacitance. 

voltage

esr, etc. 

polarity

non-linearity

modulation

measuring caps 

use in circuits: blocking, filtering, bypass, phase shift, etc.   In the audio path or not?   Used in impedance range or not?

audibility

manufacturing

current thinking

research

“re-capping”

conclusion