Over etched

August 25, 2021
 by Paul McGowan

I love this term (though I don’t appreciate its sonic impacts). It’s used to describe an unnatural emphasis on some higher frequencies some of the time.

It is typically associated with solid-state amplification gear.

We rarely ever use the term to describe the performance of a loudspeaker. Here, we would say it’s bright or has a glare to the sound.

Over-etched seems to track along with the music as if it were added as opposed to inherent.

Why would this matter?

Because the causes of over-etching are typically dynamic distortion products generated by specific combinations of frequency or amplitude events. We know this because once identified by competent circuit designers it can be reduced or eliminated through changes in the basic circuitry topology.

For example, it is not uncommon to experience over-etching in high feedback circuits yet extremely rare (or non-existent) in zero or low feedback topologies.

Segregating the differences between bright and glare vs. over-etched can be a real key to the circuit designer of audio equipment.

Subscribe to Paul's Posts

17 comments on “Over etched”

  1. Paul,
    A much better informed & detailed description about this phenomenon than last time you presented this topic.
    Thank you.
    This “unnatural emphasis on some higher frequencies” can induce a degree of harshness coming from the tweeter.
    Having only ever owned SS amplifiers, with varying degrees of negative feedback included in their topology, I have found that over-etching has not been as great a distortion concern for me as some source distortions, like jitter & studio recordings or CD quality.

  2. Talking about experts…
    Every expert has his own opinion about what gives the best (taste !) soundquality.
    A lot of feedback, not a lot, no feedback.
    If there was only 1 right answer we could all agree on, every manufacturer would build the perfect amplifier.
    So, as always, in the end there is only 1 pair of experts I trust….my ears.

    1. Of course you’re right, JB4, but let me add to your thought. One cannot say low feedback or high feedback (as examples) are always right but not because there isn’t a consistent answer but because of the wide variety of applications those circuits are used for.

      For example, the same amplifier topology for a phono preamplifier can be rearranged to become a power amplifier. In each application we might find that feedback levels have to be different to suit the task. Higher feedback’s going to overall sound better for the power amplifier than it is for the phono preamplifier.

      Whenever a complex system is involved there’s no one right answer.

  3. Not being a technical person, I can only use terms I barely understand, but I’m led to believe the etchiness of Class D was largely solved with higher switching frequencies and better power supplies.

    I have heard several speaker designers criticise other manufacturers for designing speakers with a higher top end leading to a bright presentation, which is attractive in the short-term and generates sales, but ultimately is fatiguing.

    I can think of some speakers that I thought were over-etched throughout the frequency spectrum but, as jb4 suggests, home audio is not entirely about accuracy, but also a matter of preference in presentation.

    1. Good morning Steven!
      Had you ever spent any amount of time listening to music threw tube amps?
      The reason why I’m asking, is this.
      Most speakers that have a bright sound to them, sounds a little relaxed when driven by tube amps.
      I haven’t really invested a lot of time in trying to figure this out, but I think for the most part, the output transformers are playing a maigre roll in what’s happening with the sound there.

      1. I tried two Art Audio amplifiers modified or rebuilt by their designer, Tom Willis. One SET zero feedback and the other push pull with feedback. The output transformers are effectively tone controls and have to be high quality. Tom uses transformers made by E A Sowter & Co, for the cheaper amp the transformers alone cost about $700 at trade cost. There was no brightness, wonderful midrange, but even with Harbeth little bass and what there was was bloated.

        The only other brands of tube amplifiers I considered were by Audio Note and EAR Yoshino, because they make their own transformers. AN make all their own components. I should have tried an EAR V12, but I’m past it now. I do have an EAR phono amp and I love it.

        1. Hi Steven.
          Trafomatic make amps and transformers too I don’t think the bass is bloated on my Russell K red 150Se (special edition) and have not felt the need to change from the 45 watt integrated, I do have other amps to hand.
          The sound is great as is.

          1. The push pull was 35w. I tried Hindemith’s sonata for piano and bass tuba and the tuba just did not work. A Quad 909 was a major improvement.

            I once heard a fabulous little Audio Note system in a small room, I think a P2 integrated and a pair of AN-E. It was more fun than an Ongaku/Wilson setup in a large room.

            AN is uniquely old school craftsmanship and in another world I might have a complete AN second system. AN are very low profile and very successful.

            1. Hi Steven.
              This update has taken a while as the Hindemith only came with todays Post the complete sonatas for wind and piano on Brilliant Classics.
              So I’ve played the tracks for bass tuba and piano and still think the bass is not bloated.
              Though I know you have much more experience than I With live and recorded classical music.

        2. McIntosh Labs makes their own transformers in house, or at least they did. I’m not sure if they still do or not, but I think that it’s likely that they do. I have not listened to Mac amps in a long time, so I have no opinion as to their current virtues or lack there of. I still really like the McIntosh visual aesthetic, though.

  4. Etched is the opposite of warm or romantic sounding equipment. Typically heard on poorly designed or misaligned solid state amplifiers, preamplifiers, turntables (usually direct drive), phono cartridges, CD players, (very evident in CD players). If my component has an etched sound I’m done with it. The musical rhythm and flow is not present and the sound is analytical, dry and hard instead of soft and liquid.

  5. To me, an etched sound is one that seems to be missing the full spectrum of sound, most noticeably in the higher frequencies. It is like seeing a divided color wheel that displays a limited number of discreet color slices rather than a smooth gradient including all the transitional hues. In nature one rarely sees colors matching the slices on a color wheel. More typically, one sees colors that are blends of a virtually infinite number of color wheel slices.

    In PhotoShop and other digital photography programs there are apps that convert an ordinary photograph with infinite hues to a simple line drawing with only a few hues, resembling an etching. Or if you photocopy a photo, then make a copy of the copy, and keeping making copies of copies you will end up with something like an etching. That is what bad audio gear or too many components in an audio chain can do with sound. Some people may like the abstraction for certain types of music. My ears and brain prefer a more natural sound.

  6. In the 1950’s acoustic suspension ( AS ) speakers were invented and consumer audio started to change. You could now have big clean bass from a small box. But, there was a catch, you need power to drive that small box. If a tube amp of the that period produced 25 Watts it was considered powerful. People started looking at solid state ( SS ) amps, but it took them a decade or more too catch up and produce the power needed. They did so by using bipolar transistors that did not require transform coupled outputs. amps with 50 W, 75W and even 100 W per channel became common place. But, there was still a problem. Bipolar transistors produce odd order harmonic distortion ( 3rd and 5th order distortion. This is refereed to as atonal distortion which makes the sound seem harsh, edgy or brittle. Add to this too much feedback and you have some pretty bad sounding amps.

    OTOH, many tube amps ( even order harmonic distortion and transformer coupled output ) sounded fat, round and bloated.

    All of this has changed in the last two or so decades. In the late 1990’s John Atkinson at Srtereophile reviewed a conrad -johson ( cj )preamp and phono preamp ( all tubes ) he said that difference between good tube amps and good SS amps was becoming very small ( I later bought those amps and loved them ). SS power amps use FET’s today which sound much more like good tubes amps.

    There is one thing that Steven touched on above about SS versus tube amps that I have also found to be true. Tube amps are basically good at voltage amplification but no so good at supplying current. If you have speakers that are capable of good, tight and powerful bass they will need a good current amp to produce that bass. I still us a cj tube preamp and phono preamp, but I use SS power amp that uses carefully chosen FET’s to produce lots current for great bass.

  7. Paul just published an article about experts – Experts – jest the other day. A whole article. No sooner did the words tumble out of his mind and onto his keyboard, then an Expert Hero’s bellied up to the bar – belly first – to spew forth more expertise. Expertise.

    Expert Hero tonyplachy has proclaimed “…Bipolar transistors produce odd order harmonic distortion (…) 3rd and 5th order distortion…”.

    So, uh, other typologies don’t? Is that what Hero Expert Plaky’s saying? And, duhhhh, zat all those evil, bipolar transistor amplifiers generate in their distortion profiles? Jest smelly odd order (harmonic? IM? TIM? Bob?) distortion? That’s it?

    FET’s, VFET’s – no “odd order distortion” [presumably harmonic, but who knows – could be bubonic, tympanic, microscopic, robotic, lipsiodic, guuhh] with those?

    Class D, G, T? A-OK, eye-wink, finger-snap, good to go? No distortion? Good, “nice” distortion? Ice cream?

    Oh, wait – you think tubes – primitive, thermionic, vacuum tubes – have no “odd order distortion”? Or, maybe jest a tiny bit, Jest a touch. How much? 10%? 1%? 0.1%? 0.01%?

    At what output level? Across all frequencies, or jest one, like an easy peasy 1Khz?

    In the real world, different chube amplifiers produce different distortion profiles, Some bad ones generate a whole bushel full of 2nd harmonic distortion – but that doesn’t mean that they don’t generate the rest of the harmonic series. That means and includes ripe, wet, odd order byproducts.

    In fak, I believe a recently popular Chi-Fi chube amp was recently tested and recently found to have higher 3rd order distortion products than the rest of distortion products in its profile.

    Oh, and those evil, bad, corrupting bipolar transistor amplifier circuits? Mostly, the good ones typically have distortion characteristics that look pretty much like 0.01% THD at any frequency, 20 Hz to 20 KHz at rated power output into a specified impedance load. Zat 0.01% bothering your ears that much?

    But, why should I blather on, like some Hero Expert? I’m an idiot. But, we have an actual, real expert here who can speak with confidence on the topic, He be Paul McGowan. Any body else not trained in physics and/or electronic engineering – that’d be jest another jive turkey – oops – I mean Hero Expert – bloviating.

    Rock on.

Leave a Reply

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

4865 Sterling Dr.
Boulder, CO 80301

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