In my last thread, someone told me that comparing an Outlaw 950 to a Lexicon processor was absurd. Here is one example of how wrong that is. Some of you may not be old enough to recall 1988 with much clarity. That year, at its annual convention, Stereophile magazine conducted a blind test comparing that then darling of tube lovers, the VTL monoblock, to a "cheap" solid state Adcom 555. Comparing a $4,000 tube amp to a $500 solid state is unfair. Right? The VTLs had to win. Stereophile did A/B comparisons to something like 1,000 listeners (I can't recall the exact number of participants) with levels set precisely the same for the two amps, using B & W 901 loudspeakers, and playing many musical styles from a variety of labels. The result? No one heard any difference. There was one possible exception. On a Telarc recording of the Faure Requiem recording with organ, about 2/3 were able to discern a difference. Likely, they were hearing the "cheap" Adcom's superior damping factor and the resulting tighter bass. That is precisely why you can - and should - compare Outlaw to its more expensive competition. And to the guy who stated that the only pure-sounding pre-amp is a passive volume control, you could not be more wrong. That means setting of the volume control determines in large part the output impedance of the pre-amp. As the output impedance climbs toward the input impedance of the amplifier, the closer you come to creating an interaction called a tone control, only this one cannot be defeated. Ohm's law cannot be broken without consequence.

tbng,

Interesting thoughts. Since no one these days seems interested in A/B comparisons, it is all subjective. I would challenge anyone to A/B the Outlaw Model 200 monoamps against other solid state amps in a blinded, controlled listening room. I would say that my Aragon Stage One is quieter in 2 channel at low level music than the 950, but, I truly wonder what difference I would hear between them at normal listening levels. I sent my 950 back because of a rather loud hiss in the background, and because of ergonomic issues. Other than cables, are there other good A/B comparisons out there? I would like to read them.

The only "underground" magazine I know of that does double-blind testing is The Audio Critic (although I am not current on that genre of magazine; there could be others). TAC editor Peter Azcel long ago concluded that there are no audible differences between most pieces of gear, especially D/A converters and cables and interconnects. Amplifiers, of course, vary in power output and their ability to drive low impedance loads, the latter of which should rarely be an issue in a home system unless you're dumb enough to parallel a half dozen speakers or buy some exotic speaker design. There is no reason that you cannot buy an amplifier based solely on published specifications. Those who claim to hear differences in "air" on subjective listening of basic pre-amps and amplifiers have air between their ears. "Air" in recorded sound comes solely from the recording, the loudspeakers, and the listening room. And Radio Shack's gold interconnects are just fine, thank you (sorry, Outlaw), unless you're a fussbudget who has to change connections once a week. At audio frequencies with the lengths typical of home systems, it is insane to waste money on so-called high-tech cables. Save the money for recordings. You might consult the expertise of engineer Bill Dudleston of Legacy Audio on that subject.

I am "the guy" who wrote about the passive preamp (volume control only). I should have prefaced the statement with the warning that the cable coming from the volume control's wiper to the destiation component's input should be no more than a foot or so in length, and that the cable should have low capactitance per foot.

When such controls are "wide open" the effective circuit impedance is that of the source component (actually a bit lower taking into consideration the parallel loading of the potentiometer), which is almost always less than 100 ohms in a solid state device. As the control is rotated the output impedance increases, but if the cable capacitance is sufficiently low, there will be no high frequency attenuation within the audio band. Resistance as calculated by Ohm's law does not really enter into the picture in this simple circuit - resistance itself has no effect on the ultimate sound, only the interaction of that resistance with cable capacitance. A series resistor with some value of capacitance across it to ground (essentially the circuit we are talking about here, only with that resistance being variable and in the form of a potentiometer and the capacitance coming from the cable) creates a single pole low pass filter with an ultimate slope of 6db per octave. The breakpoint frequency where the attentuaion is 3db is calculated by the familiar F=1 / 2pi RC. If this frequency is 50kHz for example, there will be no audible high frequency attenuation.

The resistance of the potentiometer should be selected to be as low as practical so that it does not load the output of the sending device. Most consumer gear can tolerate loads as low as 2 kilohms without causing an increase in distortion or limiting the maximum output level. A potentiometer of around 10 kilohms is a reasonable compromise provided the cable from the wiper to the destination component is short.

It should be pointed out that in every conventional analog component like power amplifiers which have level controls, there is a potentiometer configured exactly in the above manner, with the only difference being that the wiper-to-input cable length is genrally, but not always, shorter.

Generalizations as you stated about the interaction of the potentiometer to the rest of the circuit are fine as far as they go, which in this case is not far enough. Like all electronic circuits (or any device for that matter, electronic or mechanical), all the parameters involved must be taken into account and the best compromise arrived at. There are numerous calculations, adjustments and compromises made in designing any electronic device - taking into consideration cable loading is one of those things, and in a good design the various circuit parameters can be juggled so that there is no audible sonic impact, whether we are talking about something as simple as a passive volume control or as complex as Lexicon's latest and greatest preamp/processor.

As for difference between the sound of components, all I can say is that IT DEPENDS! While in the majority of circumstances any differences are below the threshold of audibility, given the right combination of components, the differences can inded be audible. While in the blind test you cite between solid state and vacuum tube power amplifiers and the conclusion that listeners could hear no differences, all that proves is that using that paricular combination of loudspeaker and amplifiers, listeners could tell no difference. In other circumstances such as with speakers presenting either a more or less complex load (impedance, which includes reactance from the drivers and passive crossover network), the conclusions of this listening test could very well be different. The devil is in the details!

[This message has been edited by soundhound (edited June 13, 2004).]

Citing the chart from Legacy Audio's loudspeaker user manuals, the per foot capacitance of a 12 guage interconnect is 24pF per foot. In engineer Bill Dudleston's words, "Approximately 300 feet of 12 guage would be required to establish a corner frequency of 20 kHz with an 8 ohm loudspeaker." If you're wiring St. Patrick's Cathedral for sound, you have a point about interconnect capacitance, but not in any home installation likely to exist. Yes, if the wiper is full out, the resistance of the volume control is zero. You always going to listen that way?

You have completely missed my point! I was not talking about speaker wire, but interconnects. Did you read that part in my post about the capacitance and resistance being optimised in a passive control so that there is no audible impact? It can be done, and very easily. I would suggest you do a bit of reasearch in some engineering texts to get more familiar with the calculations.

No, I did not miss the point. I AM talking about interconnects, and Dudleston's figures are about interconnects.

I don't care about Dudleston's tables or whatever - do the calculation yourself! - Assuming a 10,000 ohm potentiometer and two feet of RG-59U cable (22pf/foot) under worst case circumstances, the 3db down point is well over 300kHz - three hundred thousand Hertz! Can you hear that high? Do you honestly think that would have any impact below 20kHz? I think that makes a pretty darn transparent volume control, and one with a lot more high frequency extension than almost any active preamplifier or power amplifier which typically start to roll off in the vicinity of 60 to 80kHz.

If tracking between channels is a concern or if a conductive plastic potentiometer is not good enough, there is always the option of a stepped attenuator using fixed resistors. These resistors can be of any quality level (and expense) desired, and the tolerance can be controlled to any degree you want.

I still don't see what the loudspeaker and speaker wire reference you cite has to do with low level interconnects and passsive volume controls - you seem to be confusing the two.

[This message has been edited by soundhound (edited June 13, 2004).]

SH and tbng:

I don't want to start a fight here, just want some clarification. I don't understand the EE and physics here. If interconnects matter in terms of transmission of full frequencies and eliminating extraneous noise, how does this statement fit (from a previous post)with the above discussion?

"If changing interconnects does make an audible difference, there is something wrong with the design of some of your components!"

Again, please, some education, not argument.
Thanks!

If components have an excessively high output impedance, then they can suffer high frequency losses that can be calculated by the formula I gave earlier if the interconnect length is excessive relative to the output impedance of the device.

If a passive volume control were advertised as "being compatible with 20 foot output interconnects of your choice", and it had something like a 50k ohm potentiomenter inside, then you could consider it "not properly designed" since connecting a 20 foot interconnect on it's ouput would cause high frequency loss - it would be down 3db in the vicinity of 7kHz! If however a passive volume control had a permanently connected 2' low capactiance interconnect on it's output, and had a 10k ohm potentiometer, then it could be considered "properly designed" since it would be difficult to connect it in such a way as to cause high frequency losses.

The design goal of active electronics such as the 950 is to be as immune as possible to the effects of cable loading. Engineers do this by insuring that the output impedance remains low (something that is easy to do with current devices).

Admittedly things like passive volume controls are not something the average person will ever encounter, but they do have very real advantages because they are an extremely simple and effective solution to the problem of coloration of the sound by active components (preamps etc). When you journey into things like passive volume controls, just like such techniques such as active bi-amplification, there are design and useage constraints that must be taken into consideration - in the case of a passive volume control, you just don't hang a long interconnect on it's output!

You might think of this like adding nitrous to the engine of your car - if you don't pay attention so some design and usage issues that can crop up, things can go very wrong. Pay attention to those details, and the result is improved performance.

[This message has been edited by soundhound (edited June 13, 2004).]