It’s all in the translation

October 5, 2021
 by Paul McGowan

For most people, amplification of the musical signal means just that. Taking the original signal as generated by the source and making that signal bigger.

Only, that’s not what happens.

In the same way the energy from your leg pressing a gas pedal is not actually amplified in the movement of an automobile, it would be more correct to think of it as being translated rather than amplified.

Why is this important? Because understanding at a fundamental level that an audio amplifier is a power supply whose output valve is controlled by the input signal shines a bright light on the importance of the valve and the power supply rather than focusing on the input signal.

Going back to our car analogy, we shouldn’t care about the quality of the shoe used to control the gas pedal. Instead, we want to focus on how perfectly the translation of our foot’s instructions is carried out by the car’s drive train.

It’s the translator that we should be focusing on as opposed to harboring the notion we’re somehow preserving tiny signals in their original form.

Thus, we designers must pay strict attention to the power supplies that feed downstream equipment and the responsiveness of the valves used to regulate the flow of their power.

It’s all in the translation.

Subscribe to Paul's Posts

36 comments on “It’s all in the translation”

  1. “It’s all in the translation”…but mostly in the power supply.

    Because of the time differences I’m usually sleeping whilst the rest of you are deep in discussions.
    With regard to yesterday’s topic, I just want to put my 2c worth in, specifically for the guys who find the whole ‘current vs impedance’ thang quite counter intuitive…you all know who you are.

    So; it makes perfect sense that if we are pushing a wheelbarrow full of bricks up a hill (say a 15deg. slope) that gravity will offer an amount of resistance & to move said barrow full of bricks, forward, up said hill, we counter that resistance with energy from our muscles (power/watts) & if the slope of the hill tilts more towards the vertical (say a 30deg. slope) the resistance will increase & we simply increase the effort (power) from our muscles to counter the increase in the gravitational force.

    Now, when it comes to electrical currents moving through a path of resistance you have to keep 2 things in mind:
    (Even though electricity & water are almost always not a good combination, I’m going to use ye olde ‘water flowing through a garden hose’ analogy)

    1/ For the resistance to reduce (get less) in the garden hose, in practical terms, the diameter of said hose would increase allowing more water to flow through it.
    In the same way, as the resistance in a circuit or copper wire reduces it allows more current (water) to flow through it (thicker wire is just one example)
    Therefore, in electrical terms, as the resistance decreases the current *must* increase so 100watts into 8ohms can become 200watts into 4ohms or 400watts into 2ohms or 800watts into 1ohm…less resistance means more current is available, & providing that the voltage remains the same, the power (watts) increases, since Power(watts) = EMF(volts) x Current(amperes).

    2/ As you get closer to 0ohms the workload on the amplifier becomes greater & greater because it is getting closer & closer to a short circuit (zero ohms).

    So, to recap:
    The higher the resistance, the easier it is on the amplifier, because it is required to produce less power…smaller diameter (higher resistance) less water (current) flows through.
    Conversely, as the resistance becomes lower (larger diameter of the hose), the amplifier is required to produce more current (the larger the diameter of the hose, the more water flows through it) & this means that the amplifier is working harder producing that extra current (water).

    In Paul McGowan’s words, “I hope this helps.” 🙂

    1. That seems like a pretty good explanation.
      One thing I meant to mention yesterday. From my days in telecoms I recall that for maximum power transfer impedance matching was required. Now I know hi-fi isn’t telecoms but they are both about getting sound from one place to another. Can anyone explain why we don’t impedance match in hi-fi as discussed yesterday?

      1. As long as the impedance of the loudspeaker is higher than the output
        impedance of the power amp all is well; the reverse, however, is disastrous.
        So I’m assuming that it’s the same for impedances between pre & power amps
        & sources & preamps.

        I have blurred the lines between resistance
        & impedance in my explanation above.
        However, for the sake of trying to impart
        the basic concept (above), said
        ‘line blurring’ is acceptable.

        1. I get the analogy and also understand that a loudspeakers impedance is not fixed but variable, but what I don’t get is why we don’t use output and input impedances that are more closely matched, for maximum power transfer. Is this not desirable in hi-fi or does some other parameter override it.

          Would Paul care to chip in with a simple(ish) explanation or perhaps he doesn’t want to dig the hole any deeper.

        2. OK, So am going to try to address both FR’s comments and Richtea’s comment. FR is right on two counts. His example of the doubling of power for reducing the RESISTANCE by half works for DC circuits. The water hose analogy works well here. The current draw as you go from 100 W to 800 W is 3.5 amps, 7 amps,14 amps and finally a walloping 28 amps.

          So why do telecom companies match impedance ( not just resistance ) in their circuits. It is because IMPEDANCE is not just resistance it includes the REACTANCE of inductive and capacitive elements whose reactance depends on frequency. To truly understand impedance and its dependence on signal frequency requires some degree of mathematics which I know turns most people off ( sadly ). Thus I will just state that when an electrical signal encounters an impedance mismatch some of the signal is transmitted an some of the signal is reflected. What you want is transmission and not reflection. Thus, having a circuit with matched impedance increase transmission and reduces reflection. For audio circuits this is not as big a problem as it is for telecom transmission circuits.

    2. Wasn’t there a prequel to your lecture?

      Professor Bull Right (aka Fat Rat) enters the Lecture Hall, puts down his Textbooks, picks up his Ruler for Knuckle Rapping, starts pacing back and forth and quiet’s the class by bellowing…..


      After your lecture(?), all I could think was WTF!

      So this is your analogous explanation?When are you releasing your new Textbook on Amplifier Design? I will be sure not to buy it. You realize that there are many Textbooks that sound just like your explanation don’t you? ◔ ◔. Thanks Mike

      P.S. i increased the size of my biceps about 20 dB (another analogy) just by reading your post. OMG, I’m starting to look like ‘Arnold’

  2. There is no doubt that the supply plays a very important role in the reproduction process. This is why designs should separate independent supplies for digital and analog circuits. The sad part is that in much of the commercially available consumer electronics, this is overlooked in many areas, and the most important section, the analog section, suffers by reason of it.

  3. Thanks for the effort, mr Fat McGowan (or should I say PaulMcRat) 🙂
    But I hope that you get that people who do get all this technical stuff (like you) don’t get WHY people who don’t get all this technical stuff (like me) don’t get this technical stuff.
    The problem is that what for you is “logical” is not necessarily logical to others (like me).
    In highschool mathematics was very logical…to my teacher. Not to me and a lot of my classmates.
    And I don’t think the wheelbarrow example helps : more resistance, more (muscle) power needed.
    We all get that. But then in electricity : more resistance, less (amp) power needed.
    Therefore I used the term counter-intuitive.
    Anyway, my takeaway from these impedance lessons is :
    Less current (ampere), then more pressure (voltage) is needed to get the same power (watts). And vice versa.
    If that’s not correct, I give up.

    1. “Less current (ampere), then more pressure (voltage) is needed to get the same power (watts).”

      As previously stated, in this specific application, the output of a power amplifier is an a/c signal. Therefore, in this application, impedance can be viewed as the time relationship between voltage and current. Now, how can we apply this methodology to the design of the speaker cable interface?

      This concept can also be applied to a/c power delivery.

  4. I’m an uneducated thug on most things including electronics, and my knowledge is limited to what is practically necessary, like wiring a plug.

    When it comes to music, in which I am totally uneducated, I very much enjoy going to masterclasses. I remember Andras Schiff demonstrating to a poor hapless student how to play a certain phrase from Janacek’s “In the mists”. This is sublime, beautiful music, at least when Andras Schiff plays it. This arpeggio lasted a few seconds, and you don’t need any education to know if it’s done properly or not. The beauty of masterclasses is that you get the right and the wrong and you get to know the difference.

    With audio, if you go to enough dealer demonstrations or manufacturer events, or even well set up audio shows, concepts like reactive impedance lose any relevance.

    With regard to today’s post, there is as tool called a proportional divider. There was a plastic version sold as a toy when I were a lad. I remember seeing one from the 18th century at an exhibition about manufacturing money at the British Museum. This divider was so well made and accurate it could be used to convert a full size (A3) portrait down to the head to be engraved on a coin. The head on the coin is not an original carving, but a scaled down version of a much larger original. It is so good you cannot tell the difference if you do not know the size. It’s a sort of amplifier in reverse.

    1. Steven,
      You say, “I am an uneducated thug on most things…”
      To which I would say, “Bullsh!t” 😉

      It looks like ‘The Ashes’ may be postponed 🙁

      1. Nothing beats starting to watch a cricket match in baking Brisbane at 11pm in the UK in the freezing cold. I fear it’s going to be a winter of discontent.

        Plus I plugged in a new TV that’s been sitting here in a box for 4 months, an LG OLED thing, marvellous picture compared to the old non-HD thing we had. Moreover, it’s in the bedroom on a motorised arm that cost more than the TV. Luxury, bloody luxury.

      2. Double jabs needed to save us all. Cricket —so risky.
        In Melbourne now it’s “no Jab—no job” for anyone who deals with anyone. And specifically including dog groomers. Only licensed hermits to be exempt.

  5. Paul, I do not understand what you are trying to say in today’s post. What an ideal power is supposed to do is be is a straight wire with gain. Power amps are supposed to do is increase the amplitude of the input signal and nothing else. Of course in the real world this is not possible. All power amps both distort and add noise to the input signal as they increase the amplitude of the signal. The input signal is not a valve or accelerator pedal. What is important here is exactly what the power amp DOES to the input signal.

    1. Hi Tony. I disagree. An amplifier is a power supply valve. The input signal is only a means to turn that valve up and down.

      Consider a vacuum tube or a FET. The input voltage on the grid of a vacuum tube changes the polarity of the “fly swatter” metal of the grid and current is encouraged to flow from the power supply to the cathode. That now becomes the new signal. It is related to the input signal only by reference or as a translation mirroring what the input signal does.

      We do not “listen” to a bigger version of the input signal. In the same way the muscles used in your finger to push an elevator button have nothing to do with the elevator car moving you up a floor, the input signal is only a “button push”.

      More than semantics, this is an important concept to wrap your head around. It’s definitely not your conventional thought process so I understand if its not resonating, but give it some time.

      1. Does this help?

        The input signal is being being reproduced thru each amplification stage to a higher value. The valve part comes from the fact that what it sees as it’s original signal is what turns it on and off. An amp stage doesn’t pass. (An illusion that it does)

        1. Paul and Mike, I get it. The power amp is not like a magnifying glass that enlarges the image as it passes through it. Instead the input signal acts as a valve that modulates the power output of the power amp. Thus the reason why the British call vacuum tubes “valves”. The input signal does not pass through the power amp. This is what I would call an operational description of a power amp.

          I refer to my audio gear as my audio system and my view of it is what is the function of each piece if gear in my system. Thus the cartridge’s function is to convert a mechanical signal into an electrical signal. There are several ways to do this and I understand how each way operates to do the same function. In this functional view the function of the power amp is to produce a bigger version of the input signal that I do listen to. This must happen because the input signal cannot drive my speakers.

          I am not dismissing this to semantics, I think of it as different ways of viewing my system. There are times when I let my mind go inside each piece of gear and see what it is doing in detail. There are times when I let my mind follow the input signal through each part of the system. Fortunately, most of time I just sit my ass down and enjoy the music. 😀

  6. Todays analogy / discussion today makes sense again. The power supply is the key. The signal passage through it’s components utilizes power supplies for amplification. The purpose of the signal is to have all its qualities maintained from the beginning to the end. (Think of it as a seed) The purpose of any power supply is to provide a voltage / current to a component acting upon the signal. The trick / goal is not to have the power supply or component impart any of its characteristics upon the signal. The other trick is having a power supply design (with head room) to easily drive its intended load. In the case of a power amplifier driving speakers, it has to deal with the reactance (ac impedance) along with the resistance provided by the load.

    We may not need to understand all the details, but understanding some (if you are are making the choices yourself without any outside sales help) can be very useful.

    For me realizing speakers are primarily a current driven device has always helped. From a simplistic standpoint, if the load impedance drops in half, and the same voltage is to be maintained, then the current has to double.

    1. Mike,
      I’d have to say that 500watts/ch/8ohms seems to roll right over reactance
      & resistance, as if they didn’t even exist…just like a steamroller 🙂

      ‘Steamroller’ – James Taylor
      (since my links don’t work)

        1. Mike,
          Thanks but I didn’t mean it that way & I appreciate that you’re forcing me to learn something new; you’re a good man.
          The problem is, as ‘JosephLG’ informed me, that my last 2 YT links that I posted on PP came up ‘Video Unavailable’ over there in Yankeeland (one was ‘He’s Gonna Step On You Again’ – John Kongos, a couple of PP’s ago) even though they worked properly over here in the land of Oz…go figure; unless he was just yankin’ my chain??

          1. No issues here FR. I was yanking your chain. 😉

            What I have seen a few times is that some links / videos are not available in the US. Why I’m not 100% sure – that’s a good question for the legal dept. If you check the link you’ve typed in and it works, then you’ve done things correctly. Maybe not the most expeditiously…. 😀

            Enjoy that steamroller, as I know you are…

      1. Good morning Neil and everyone!
        If you would, rap your heads around this.
        Lets just say, you have a recording of someone playing a pipe organ.
        And the producer took care, to record the organ the way it’s heard in real time.
        Not all systems can reproduce what’s on that recording accurately.
        I don’t remember when exactly Paul answered this question for a man, but he made and posted a video about it.
        The man had a 300B single ended tube amp.
        But his amp wasn’t reproducing the notes played on that pipe organ correctly.
        But now that I set here and think about it, it makes perfect sense.
        Paul told him to look at the capacitors in the signal paths of that amp.
        Keeping the voltage the same, he told him to raze the other values of the caps by a margin of one.
        Like for an example, .1MF, should be razed up to about .2 or .3MF.
        And that will get you the rest of what you’re missing out on.
        If you can remember, I told you that I done the same thing to both of my Mcintosh MC-275 vintage tube amps.
        But only, I used V-Caps to get me there.

  7. I like the car analogy. Different cars can have the same rated horsepower, but the behavior of the linkages of the power train affect the response and performance of the car in various road and driver demand scenarios. Driver car selection and pricing is based on performance and quality factors, not just horsepower.

  8. Today was completely above my head but still enjoyed the discussion between fellow members pity FR isnt going to write the book but pleased to know Steven has this amazing thing on his bed for watching the TV but sorry to learn that The Ashes might not happen. So one always learns from Pauls daily titbits

  9. It’s still the whole package. The best power supply in known space with substandard modulating circuitry will sound substandard. Likewise a substandard power supply with superior modulating circuitry will also sound substandard, but probably not in the same way. As you say, all in the translation.

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