The sound of components

July 31, 2019
 by Paul McGowan

Every component in a circuit has a sound to it. Altogether, the hundreds (or thousands) of parts that make our HiFi systems work contribute a mash of sonic qualities that we end of listening through.

It’s rather a dose of hubris to think we can ever get perfection out of all the various flavors and colorations each component adds.

Of course, an easy way to not get flummoxed by all these problems is to ignore them altogether. There’s plenty of engineers who honestly believe one resistor sounds the same as any other. It’s how we wind up with mindless and heartless designs some people are unknowingly subjected to.

I find it somewhat comical to be on both sides of the argument. On the one hand, I sympathize with engineering types fresh out of school or long in the workplace faced with an entirely new set of variables like the sound of components. It’s not listed in textbooks, certainly not taught in schools, and ridiculed by those faced with the concept. That’s what people do when faced with the unknown. Easier to rally against it then be open to the ideas.

I also understand the practical side where engineers wake up to what’s really happening with components. It’s daunting to add the categories of sound for different passive components like caps, resistors, and inductors (let alone semiconductor types) to your knowledge base. But, just because something is daunting it doesn’t mean it’s not true or worth pursuing.

We can’t escape the sound of components.

Fortunately for you, there’s a dedicated handful of companies that not only get it but revel in the choices available. These choices open up new methods to get the sound we’re hoping for, to get closer to the music, and to find the purity buried in recordings.

The sound of components can be the sound of music in knowledgeable hands.

Subscribe to Paul's Posts

18 comments on “The sound of components”

  1. Sound is what our brain consciously perceives when listening to a stereo system in a given listening room. If there is a “sound” of components it should be mandatory that the designer of the recording and mastering equipment also must be the designer of the home stereo system in order to get the most accurate reproduction of the recording! Thus it is impossible that there is a true high fidelity system out there.

  2. I guess the tweaking of the equipments sound by choosing certain components also means that the high claim of „neutrality“ (which makes only little sense anyway imo) can only be held in terms of measurements…every further (usually noticeable) tweaking of sound and tonality means deviation towards own preferences of the designer and his environment. What comes out finally is everything but not a „neutral/accurate“ component. Wrong?

  3. “It’s not listed in textbooks, certainly not taught in schools, and ridiculed by those faced with the concept. That’s what people do when faced with the unknown. ”

    It’s so interesting people act this way. (I’m likely guilty of it too at times.) I can’t tell you how many times I come across comments online where someone claims they know something merely based on logical analysis alone and not through experience.

    Just the other day I asked someone why they thought “overbuilt” power supplies don’t matter for low current devices like preamps. After saying because “the current is so small, it doesn’t matter,” I asked if they derived that through experience or just “what they thought made sense.” They replied the later. I linked them to your video on power supplies and how you found out that size does matter. They were actually receptive of it! Thanks for making the video (and this post.)

  4. If you have the math it’s much easier to understand. The powerful concept of “transfer function” tells you how to understand these characteristics and how to change them. There’s often more than one method to achieve the same goal. Transfer functions compare the difference between what’s happening between any two points in a system. In the case of the absolute sound we look at the transfer function between the sound projected into space by musicians and the sound that reaches your ears in the audience. Both can be described by mathematical functions. The overall transfer function connects them. If there is more than one musician performing we simply compare each one individually to the sound reaching your ears and add the sound reaching your ears together. The process is called superposition. For those mathematically inclined what it says is that; F(a) + F(b) + …… + F(n) = F(a + b + …. + n). The goal of high fidelity sound is to create a different process that has the same overall transfer function at another time and in another place. The means of changing the time is to store it partway through. That’s our recording.

    One problem is that if you are listening to live music and you move from one seat to another, even the one next to you on the left or right, or the same seat in a row in front or back of you the transfer function will change and the sound will be different even if such a small move isn’t noticeable to you. Move a large distance and it will be noticeable. So this brings up the question of whether or not the concept of absolute sound has any meaning. Even if you sat in the same seat hearing the same performance by the same musicians on two separate occasions the sound you hear would be different. Let’s say you went to a matinee performance in the afternoon and then to the same performance in the evening. Besides no two performance being identical, the conditions which affect the transfer function would be different. If the temperature, humidity, and/or barometric pressure changes so will the transfer function. We call those “conditions of state.” If the matinee doesn’t sell out and the place you listen in is half empty during the matinee while it is sold out and filled with people in the evening, other factors will change the transfer function. We call those “boundary values.”

    So, with countless ways to make recordings and so many variables and countless different listening rooms how do you engineer a sound recording/reproducing system that gets you into the ballpark if not in the right seat. It starts by understanding this concept of transfer function and how it applies to sound. The system must be engineered to produce sound fields that while not scientifically identical to what was heard live has the same or very similar characteristics. And it must be able to accomplish this adjusting to the way the recordings are made, the variables of the listening room, and the transfer functions (acoustics) appropriate for each type of musical composition. I that asking too much? No, what it is asking for is just right. Anybody got any ideas?

    1. Recreating the soundfield of the original performance is the perfect concept and a most holistic approach – but as every perfect concept only an ideal or an ideal impossible to be reached due to the complexity. Thus a more pragmatic approach would be to start a research finding the relevant acoustic parameters and a reductionistic solution. MP3 was such an approach taking masking effects into the consideration. At least you should build loudspeakers which accurately reproduce the whole spectrum of an instrument without phase shifts finally delivering an identical overtone spectrum. I am not aware of a loudspeaker design fulfilling this requirement – meaning a flat frequency response and a perfect step response! This lack and the deficiencies of actual loudspeaker designs are at least responsible for the endless tweaking of a stereo system whose inherent crosstalk effects even establish additional problems. The phenomenon of optical illusions (the brain fools us by adding visual elements which are not present but seem to be “logical” based on learned pattern recognition) might be a stimulus to find the parameters relevant for acoustic/aural illusion effects!? As far as these basic aspects are not addressed and implemented in technical solutions every audiophile will just tweak his system according his individual preferences concerning a specific music genre. And the biggest problem is that a sound engineer facing the same problems has to tinker a sound which sounds pleasing for a majority of stereo systems resulting always in a most mediocre product! No wonder that compression and artificial reverberation and many other tricks are applied to get the customers satisfied.

      1. “but as every perfect concept only an ideal or an ideal impossible to be reached due to the complexity.”

        It’s not as complicated as you’d think. Not when you take it apart the way I did. It uses the first arriving sound from each source as a reference. That’s called normalization in mathematics. It then breaks down each reflection into four descriptive components; direction of arrival, delay time of arrival after the first arriving sound, loudness compared to the first arriving sound, and change in spectrum compared to the first arriving sound. The reflections can be imagined as passing through a sphere around your head or any other size sphere or any closed surface for that matter. A sphere is convenient because it’s the only shape where each element of the surface has a unique direction for a perpendicular line or normal passing through its center. If you have any calculus you’d know that each element of the sphere dS is infinitesimally small and the arriving field relationship is the triple integral of sound passing through the sphere in frequency df, time of arrival dt , and direction of arrival dS. The dependent variable is amplitude. By this method the first arriving sound and the arriving reflections are completely described. A machine is then constructed to simulate these relationships and they can be adjustable over a very wide range. In a laboratory it can be done with scientific precision using an anechoic recording in and anechoic environment given measured data from a point at one spot and the resulting field at another spot in the real world. The concept can be simplified and modified to work well with commercial recordings at home. That’s what I did.

  5. Individual Components sound different. That fact is not taught, listed, and is ridiculed.

    So why I have no problem accepting that premise, I’m left at the mercy of some audio component savant to dictate which ones sound best.

    But then again, Apparently this fact has been known since the advent of high end equipment. It all starts around which component is actually doing the amplification, and I guess where the voodoo or magic happens is with the supporting cast around them.

    It could also be inferred this knowledge can be used the other way also. Purposely change components to remove enough sonic characteristics to change the selling price without actually having to design new circuits, just stuff different components in.

    It then becomes like cell phone carrier commercials they’re all the fastest most powerful etc…

    Who do you believe? In the case of audio The proof is with your ears….

    1. ” Apparently this fact has been known since the advent of high end equipment.”

      Actually It’s been known since the invention of electronics and even before that such as the difference in sound between cactus needles and steel needles in wind up phonographs.

      “Who do you believe?”

      I believe the professors who taught me electrical engineering and the textbooks I used and others I still refer to. They have nothing to sell except education. Their reputation and jobs depend on being right all the time. When I have questions I can’t find answers to I call other electrical engineers. And sometimes they call me.

  6. When I was training in electronics during the 60’s it never crossed my mind that components would have ‘sound’. It was later with high-end discovering that stepped attenuators sounded better than pots carbon or WW, digital control was even better.
    The differences between components are often not slight, capacitors, resistors, inductors cored or air all show their characteristics massively in a resolving system. I’m afraid deniers having closed minds do not help in any way in advancing our technology

  7. It’s discouraging when in looking for information I read reviewers that say that such and such components or cables just let the sound through and din’t have a sound of their own, as they should. The irony is that if one asks the developers, they are usually quite willing to describe the sound they aimed for, and maybe the components they used in conjunction to help them do so.

  8. I don’t buy the argument. By all means use valves, which are large, inconsistent, expensive, degrade and ultimately blow up. But you get the pleasure of using 100 year old technology. Or use devices using mobile phone technology, that has resulted from billions in research, are effectively 100% reliable, are manufactured to the most exacting tolerances and are cheap. But, hey, high end audio has to be different to just about every electronic design and manufacturing process used the world over.

  9. Perfect timing Paul. I’m about to rebuild another Bose 901 Series 2 EQ Box to serve two purposes. One is configuring the box as a true balanced design by simply using the L & R mirrored circuits as my + and – plus chassis GND for a single channel box for Center Channel duty. The second purpose is exactly what your topic is all about. My “Reference” Dogs Breakfast EQ box is comprised of thruhole Vishay Dale 1W-3W resistors and SoniCraft Capacitors in all Audio signal paths. Power is fed by a big Toroidal Transformer (3.6Ax2) which feeds each of my 901 Boxes and a cheesy pair of balanced/unbalanced adapter boards which feed the “Reference” box and PS Audio Stellar Stack. Low and Super low ESR caps feed power to all the various circuits.

    I am also designing s modular PCB that can be used either as single ended L or R channel Bose Series 2 EQ or as the the + or – channel in true balanced setup.

    I am using the second Bose box to prototype SMT Vishay thin film 1% resistors and SMT KEMET Tantalum Caps in the DC circuits. Main power will be tapped off of the 1st Bose box. Audio paths will be the thruhole Sonicaps.

    Hopefully I hear no difference in Sound Quality. The 2N5088’s are also of the SMT variety (I have thruhole’s on hand – just in case).

    I never matched the transistors in the Reference box and still sounds pretty good (I got lucky I guess).

  10. It is absolutely true that the sound of components is the sound of music. Based on theory alone every resister should sound the same, every capacitor should sound the same, every wire should sound the same. But in real life they do not. The sameness theory is believed by those who have no practical experience. Resisters can make such a big difference in sound that one can be taken aback. Though the differences are small withe the regular run of the mill resisters. A carbon resister sounds different from a metal film resister. Put a Z Foil resister in the audio pathway and be floored. Replace a regular good capacitor with a top of the line Mundorf or better still a Dulund and hear the difference. Wires can make a very big difference too. The way they are cast,the metal used all make a significant difference. Put them all together and you hit a home run. But this can be very very expensive and is contrary to cost containment. That is why corners are cut. Excellent post which may not be to the liking of those who do not want to admit,for whatever reason, that components can make a very big difference. But facts are facts. Regards.

    1. Very true Oliver. One one my McIntosh MC50′, I’d replaced the input Capacitor with one from Arizona Capacitor (red). And WoW !! Imaging was more spacious, highs were sweeter, and mid’s were to die for. Bass on these old amps is deficient but good down to 40Hz (I have a Subwoofer for that…LOL). After replacing all the passives inside the amps ensures another 50 years of service. But recently one of the amps had some input transitor(s) stage let go (that’s the problem with “Vintage”
      – never again).

      4 months with my Stellar Stack and I’m still a Happy camper !

      Components Matter !

    1. All-in-one is a major trend today. It will save thousands of dollars by curing (or at least eliminating) the itch to upgrade components. Just make sure you get the original purchase to your liking.

Leave a Reply

© 2023 PS Audio, Inc.

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