Low to high or the opposite?

October 4, 2021
 by Paul McGowan

Impedance can often be a tough to understand subject. Is it better to have high or low? What’s the difference between the two in practical terms?

For those with a clear understanding, this is a good post to skip. For those still unclear about what it means when we speak of high and low impedances, this might be right up your alley.

The simplest way to think about impedance is to consider it an opposer. The higher the impedance the greater the opposition to alternating current. A wire has almost no impedance and thus offers little to no obstruction to the passage of a musical AC signal. On the opposite side of the scale, air has very high impedance (resistance) to passing electrical signals. Basically, not much electrical energy passes through air (its resistivity is approximately 2×1016Ω⋅m).

Let’s now take a simple and practical example of what this means. A speaker cable with big heavy gauge wire has nearly no impedance and thus offers little to no resistance (opposition) to delivering the output power of your amplifier to your speaker. A very thin wire might have relatively high impedance and thus offers much more opposition (resistance) to the passage of power from amp to speaker.

If our goal is to deliver power without hindrance, we want as low an impedance as possible.

Now let’s take a look at another aspect of understanding impedance in the practical world, using musical signal power to make sound.

Imagine for a moment a speaker driver like a woofer. It has a magnetic motor used to move the woofer cone back and forth to generate sound. That motor needs fuel to work. Its fuel is electric current. The more fuel we feed it the harder it works and the louder the sound.

Remembering that impedance restricts (impedes, resists, opposes) the flow of electric fuel, it’s perhaps easy to see how the higher the impedance in the path of powering the speaker equates to less output.

Thus, if my power amplifier has an 8Ω output impedance feeding a 4Ω speaker driver, the higher output impedance of the amplifier (a higher Ohm number means higher impedance) will restrict power flow and the resulting musical output will be cut in half.

If however, my power amplifier has a wicked low output impedance (well below an Ohm) then power will flow without restriction and my speaker will enjoy all the power fed to it without opposition.

If we do not want to lose power or signal level, we always want to have a low output impedance feeding a high input impedance.

Which is why preamps have low output impedance while power amplifiers have high input impedance.

Hope that helps clear the mud.

Subscribe to Paul's Posts

74 comments on “Low to high or the opposite?”

  1. Impedance is the time relationship between voltage and current.
    Just another way to look at it.

    SF Giants winning the NL West Division Title makes Giant fans happy!
    (107-55 all-time franchise record)

    The hell with Cricket, it’s time for the fall classic and cautious optimism.

    Gigantic oil spill off the coast of Southern California makes me sad.
    Have you ever been to Newport or Laguna Beach?

    1. The end of season Rugby League Grand Final was played yesterday in Brisbane, QLD, Australia.
      The Penrith Panthers beat the South Sydney Rabbitohs 14 – 12 & what a nail biting, gritty & hard fought match it was.
      Now that’s *real* football; 2 x 40 minutes of non-stop blood, sweat & effort.
      None of this stop/start, stop/start, stop/start, stop/start, stop/start, stop/start…ad nauseam stuff they call NFL 🙂

      1. One of my issue’s with Paul is that he attempts to make certain things as complicated as possible when with fewer words said, the simpler delivery of a concept.

        If you’ve walked the beaches south of Montecito through Carpenteria you’d understand why the hotel/motels there have complimentary bare foot tar removing agents in the bathrooms.

        Brandon Crawford MVP and Buster Posey our on-field manager this year.

        Ps -> please tell me the ball isn’t juiced this year…

        1. dr. g,
          Imagine how much more complicated ‘Paul’s Posts’
          are going to be in another 5 years time.
          Paul’s in it for the long-haul; he ain’t goin’ anywhere…
          & neither is his dementia 😮

          My “bare foot tar removing agent” is a ‘Black & Decker’ belt sander.

        2. Well, Dr. Goodears, I certainly am not trying to be unclear.

          Perhaps you could entertain us with a clear example of impedance using fewer words? If it works and hundreds of people who have struggled with this understanding suddenly find themselves illuminated, that would be amazing!

          I would welcome it.

          1. Good one Paul! Maybe Dr GE’s wants to conduct an one line course in electronics engineering. I could use a refresher. There’s always a pot stirrer in the group.

            1. Exactly. If he’s clearer than me and helps people understand without the gobbly goop engineers try to foist on people, I would welcome it! In fact, he can have the podium for a guest post.

              I have no axe to grind about this. It’s damned hard to explain complicated electrical interactions to people without any engineering background.

              Surely there are people far better than me.

              Bring it on!

              1. Next, Paul, please remind us why amplifiers have higher input impedance than output impedance.
                And why do SET tube amplifiers typically like speakers with higher impedance?

              2. Paul, maybe you could dumb it down even more for us non-engineers by producing a video in which you (again) shrink yourself down and travel through a cable encountering impedance, like you passed through the electrical outlet and entered the PowerPlant. Or maybe an intestional tract analogy would work. That is, unless Dr. G can explain it better in a few words, in which case the video and intestional analogy would be a waste of time. LOL

          2. Well Paul, i certainly hope i’ve entertained some of your guests today. As you are aware, there are different applications and parameters of impedance some of which indeed are dynamic, complex and difficult to measure.

            I wrote a simple line in an attempt to tie-in my off-topic post to yours which i admit was patronizing that probably should have read the engineering mantra “impedance is everything”.

            From the Oxford Dictionary;

            “the effective resistance of an electric circuit or component to alternating current, arising from the combined effects of ohmic resistance and reactance.”

        3. Let me get this straight. By bringing the time domain into today’s topic you’ve cleared everything up with one short sentence? I’m not sure who’s muddying things up Paul or you. I don’t think it was Paul.

        4. Make it more clear? simple. The two main properties of an a/c signal are voltage and current. Impedance is the time relationship between these two properties.

          Perhaps i should once again mention that Mercury is retrograde and the moon turns dark on Wednesday. (enter smiley face)

          Speaking of common grounds, this guy claims to have invented MTV. Methinks twas The Beatles “A Hard Day’s Night” released in the summer of 1964.

          https://www.youtube.com/watch?v=WnpcTsy10dE

        5. “..the beaches south of Montecito through Carpenteria..”

          ????

          wtf are you babbling about?

          What planet are you on, are all of your “people” as weird as you?

          Oh…and who gives a damn….

        1. Tough loss for the 49ers at home. Kyle Shanahan is the main story here. He faced a Seahawks team that had zero first downs and negative 11 net yards 25 minutes into the game and still lost.

          Which means Shanahan has lost 9 of his past 10 home games. And now, his win-loss record as a head coach is 31-37. Next week, after he loses to the Arizona Cardinals, it will be 31-38.

    2. Had an amusing 5 or 6 days at the St Regis Monarch Beach (now Waldorf Astoria), a short tram ride from Laguna Beach. Laguna was very nice indeed and good restaurants. We were upgraded to the Presidential Suite, so spent most of my time on one of the four terraces, about the size of a tennis court, reading as the blue (now black) waters of the Pacific rolled in. Regrettably cricket was neglected, but a friendly barman from Colombia was football-mad (as in soccer), so we talked football as he tried to educate me in Californian wines.

  2. Oh Lord, wont you buy me a BHK amp,
    My friends all have Pass Labs & I must revamp.
    Worked hard all my lifetime sellin’ ‘Bose’ & ‘Monster Brand’,
    So Lord, wont you buy me a BHK amp.

    Oh Lord, wont you buy me an FR30,
    ‘KEF’ & ‘Dynaudio’ are trying to woo me.
    I wait every day for delivery until three,
    So Lord wont you buy me two FR30’s.

    Oh Lord, wont you buy me an ‘Octave’ CD,
    Exchange rates to Australia are really killing me.
    Prove that you love me, & just buy me three.
    Oh Lord, I don’t need, no colour TV.

    That’s it! 😎

    1. FR, my son. According to the Lord’s Prayer, you should be asking only for your daily bread and forgiveness for your trespasses. Audio equipment is not what the Lord had in mind.

        1. The hardest part is forgiving those who trespass against us. By the way, haven’t you heard, God has no money to buy you a BHK amp. It’s all tied up in staff salaries, real estate and personal jet aircraft.

  3. Am I tripping reading this post and comments? Is Monday weed day where you need it to make sense of it all? Or perhaps I’m not part of the cognoscenti. Maybe senility has settled in stealth.

    Paul was making impedance brilliantly clear until the sentence attempting its application: “if we do not want to lose power or signal level…” I went WTF? The one sentence that needed clarification didn’t get it. I felt like I had fallen in the toilet pit of the movie Slumdog Millionaire. Clear as mud indeed.

    Matching impedances is an important subject and one that I’ve always tried to make sense of. Can you, Paul, or someone explain in a simple way why “if we do not want to lose power…”? Please.

    If it’s really a simple explanation and I’m failing to get it I’m retiring from posting.

    1. I would be happy to help clear the mud. I am not certain what’s not understandable. Are you confused about these two paragraphs?

      “Remembering that impedance restricts (impedes, resists, opposes) the flow of electric fuel, it’s perhaps easy to see how the higher the impedance in the path of powering the speaker equates to less output.

      Thus, if my power amplifier has an 8Ω output impedance feeding a 4Ω speaker driver, the higher output impedance of the amplifier (a higher Ohm number means higher impedance) will restrict power flow and the resulting musical output will be cut in half.”

      Perhaps you are wondering why there’s less power delivered to the speaker when a higher impedance restricts the flow of power?

      I guess I am struggling to understand how that’s not clear. Sorry. I am happy to help.

      1. Thank you, Paul. It’s quite easy to understand how the more impedance (resistance) a speaker has, the more power a speaker needs to achieve the same volume level. That part is not rocket science.

        I take it this also means that any device up the chain with impedance greater than that of the speaker will lower the volume level of the speakers. For example, a thinner wire with higher impedance than the speaker’s input impedance will also reduce volume. So would a preamp, an amp or any other device in the chain possessing higher output impedance that the speaker input impedance.

        Is the takeaway then that we always want the output impedance of a source device in the chain to be equal or lower to the input impedance of its target device, or the input impedance of the speakers for that matter? And, if that is not so, that we will then see a reduction in volume?

        If what I said is true, then I owe it to multiple more careful reads of your post, and I have climbed out of the poop pit in Slumdog Millionaire.

          1. I think some of us are thinking “Why would we want the receiving gear to impede the incoming signal? Why wouldn’t we want both the sending and receiving gear to have low impedance?” In thinking of a garden hose, by pinching or otherwise restricting the downstream end the water pressure can build up in the hose and there will be a strong flow of water molecules at the discharge end. If the upstream end is restricted and the downstream end has little restriction, the water pressure will not build up and the discharge will not be forceful. In an electrical cable, the pressure is the voltage (electromotive force) and the restrictions are the impedances. Am I on the right track?

      2. Your post was clear but it was quite detailed like you were spoon feeding but that’s not a bad thing for a lot of the readers on this post. Maybe the engineers posting here should list their degrees so that your entire audience would understand better who they really are and why they are bringing their intelligentsia to this forum.

    2. Your first post was great.

      I’d like to bring up a quote that I have used often especially regarding the person(s) you’re responding to.

      “The Ego is always at work and needs constant feeding”

  4. Stereophile measured the output impedance of my pizza box as 0.005 ohms at the end of 6 feet of speaker wire, the specified impedance is 0.001 ohms.

    The BHK300 monoblocks were measured by Stereophile as having output impedance of 0.16 ohms.

    Does that mean anything?

      1. I don’t need the encouragement to have a cookie.

        I do wonder if these figures make any practical difference, other than giving the marketing departments something to shout about.

        So may speaker cables are specified as:
        Inductance 0.002uH
        Capacitance 3,100pF
        Resistance 8.4mohms
        Leakage 25Mohms
        Impedance 8ohms

        This is too much for my brain to digest, and I suspect most other people looking to listen to some good tunes from time to time.

  5. I for once, wish someone would explain this impedance thing to me.
    The question I’m asking here, is about both vintage and modern speakers.
    With speakers made between 1945 and 1975, the impedance stays the same.
    If you have a speaker that puts 16ohms on your amplifier, it will be constantly 16ohms for sure.
    But that’s just witth vintage speakers.
    And it doesn’t really matter what kind of a complex musical passage you’re happening to be playing.
    You can crank the volume all the way up, but no matter what, the amp still sees a load of 16ohms.
    But with today’s modern speakers, the impedance rises and falls.
    Lets justt say, that you are playing a track that has a heavy bass line in it.
    But you have the volume half way up.
    But your amp is driving a pare of modern speakers.
    It will start out putting an 8ohm load on your amp.
    But when things starts getting really bissy, the load dips down to 4ohms.
    My question is, why does that happen with modern speakers, but not with vintage speakers?

    1. To add some muddy here…
      I have a pair of Kef Reference speakers from the min 80s. These present a 4 ohm *resistive* load to the amp, and are 92dB efficiency. These are very kind to the amplifier output, and pretty linear in response.

      So impedance is a reactive load where resistance is not.

      But I digress.

      1. Good morning John!
        I’ve heard of Kef speakers, but I’m not all that formilyor with them.
        The speakers that I was talking about, were made by cumpanies like Electrovoice Scott and bosack.
        Others were made by other cumpanies like Polk Carver and Faze Tech.
        The latter ones, I experienced the impedance dip.
        But the former ones, didn’t do that to any amplifier I happened to be playing with.

  6. For me the mud is not cleared. (and probably never will be, considering my “understanding” of electronics).
    I never understood why a power amp delivers/has to deliver more power driving 4 Ohm speakers then when it’s driving 8 Ohm speakers to get the same level (decibels) out of the speakers.
    Totally counter-intuitive.
    I would expect the opposite. Higher speaker impedance (resistance), thus more power needed from the amp than with lower speaker impedance to get the same output.
    Or do speakers with low impedance draw more watts from the amp ?

    1. Well, there’s a basic misunderstanding people have and you’ve touched upon it nicely.

      4Ω speakers don’t necessarily need more power than 8Ω speakers.

      Speakers are rated in sensitivity terms. That’s expressed as how loud the speaker will play with 1 watt applied. Thus, a 4Ω speaker with a rated sensitivity of 90dB takes as much power as an 8Ω speaker with the same sensitivity. 1 watt.

      The reason amplifiers want to double their output power into 4Ω relative to what they output at 8Ω has to do with this simple formula: amps x volts = watts. To produce a watt of power into a 4Ω speaker requires twice the current (amps) as does an 8Ω speaker.

      Back in the days of tube amps and output transformers this was a big deal. If you matched the output transformer to an 8Ω speaker, then you have a problem if you suddenly switch to a 4Ω speaker. The amp/output transformer are putting out the same voltage and current, but now the demand has doubled, so we get a low output level.

      Hope that helps.

      1. And not to muddy more the waters, JB4, but “I would expect the opposite. Higher speaker impedance (resistance), thus more power needed from the amp than with lower speaker impedance to get the same output.”

        It needs more voltage to get the same wattage as the impedance goes up.

  7. I did once ask Paul to explain how you measure mains impedance to actually see how the incoming power measures against regenerated power. I never got a reply. It seemed an obvious question to me.

    When I had my new mains supply installed a few months ago there was something called loop impedance or earth impedance, which apparently is something to do with your system being able to short out before the house catches fire.

    So Paul, if regenerators produce AC with “lower” impedance, how do you measure the impedance in the first place?

    1. Sorry for missing your question, Steven.

      Measuring output impedance of a regenerator is rather simple in concept ( a bit more difficult in reality).

      We basically add a resistive load to the output of the regenerator. This load demands that the regenerator has low enough output impedance and enough power to maintain that low output impedance so we do not measure a drop in voltage at the output when the load is applied.

      So, let us imagine the output impedance of the regenerator is 1Ω (in reality it is 0.001Ω or something low like that). If we place a 100Ω resistor between the output of the regenerator to ground (or in the case of a regenerator to the neutral leg of the AC power), and measure the output voltage of the regenerator, we should see little to no drop in AC voltage.

      Once we lower the impedance of the load to match the output impedance of the regenerator (now the load is 1Ω) we will measure half output voltage at the junction of output and load. Thus, we know the output impedance is 1Ω.

      In reality, the regenerator’s output impedance is so low we cannot exactly use the half output voltage point because the regenerator cannot output that much power. Thus, we use a reasonable resistive load that draws perhaps 500 watts or so, and simply use Ohm’s law to calculate the output impedance.

      1. Many thanks Paul. I can understand that with the regenerator you have a closed system you can measure. Sounds like an indirect measurement in that you measure a drop in voltage and calculate the impedance. As I understand it, what really concerns you is the drop in voltage under normal operation of your audio system, so what might be more critical is the impact of transient demand rather than constant demand.

        The question was actually how you measure the impedance of your mains supply. I assume this has to be measured at the power socket, because mains cable increases impedance. It’s a matter of knowing if there is an issue in the first place.

  8. “ Which is why preamps have low output impedance while power amplifiers have high input impedance”

    So why wouldn’t you want the output impedance of the upstream component, e.g. preamp or amp to match the input impedance of the downstream component? Wouldn’t a downstream higher input impedance also restrict current flow? So, given an 8 ohm speaker, if all the ins and outs upstream were also 8 ohms, wouldn’t that be ideal?

    1. Great question especially since it seems like the obvious answer would be to match the two impedances. But no, you would ideally want a factor of 10 between them. i.e. driving an 8Ω speaker you’d want no more the 0.8Ω output impedance on the amplifier.

      Remember that impedance/resistance RESTRICTS flow. It RESISTS flow. The higher the impedance the less that gets through.

      Perhaps it’s helpful to think of each impedance element as a pipe restricting flow. The higher the impedance number the smaller in diameter the pipe power must flow through. If an 8Ω pipe were 1/10th of an inch in diameter, then a 0.8Ω pipe would be 1″ in diameter.

      Imagine how much water can flow through the smaller diameter pipe vs. the larger diameter pipe.

      Maybe that helps. What I don’t like about the water pipe examples is they really don’t express it correctly and thus people start over thinking the example and then we get tangled up.

      1. If I’m correct, then the pipe analogy falls completely flat unless we mention that the real goal is to boost the voltage at the input on the next component down to promote adequate current flow. Otherwise, there’d be no point in resisting the flow by an average of 10 times. Right?

        1. Paul, to put it another way:
          V = IR
          W= IV
          So, given a current say of 2 amps, using the first equation, if R is 10 ohms, we get 20 volts at the input. If R is 100 ohms, we get 200 volts. Plugging those into the next equation, we get 2 * 20 and 2 * 200, which 40 watts vs 400 watts. So the higher input impedance gives more watts. Is this all correct?

          1. Paul, I’m ready for the audio 101 class then. I know basic electricity and electronics, but not that well as applied to audio. I get lost when I try to make sense of some of what John Atkinson publishes. I know there’s plenty of material out there, but if you can recommend some that approaches from the audio perspective, that would be great. The subject of impedance is vexing. I remember it as resistance times reactance, but that’s just rote knowledge.

            1. Yup. Better to think of it in DC terms as simple resistance and voltage.

              Impedance gets more complicated because it is resistance varying with frequency.

              Just try and imagine everything in terms of resistance and voltage.

              Not to be nosey, but any chance you might be Patrick Cullen in Atascadero?

              1. No, I’m Peter Cullen from New Jersey. I’m a long time customer having had several PS Audio regenerators until I landed upon the P20, which has been serving me very well. I also have the Directstream DAC, which is great. I also wonder what magic Ted is creating with the upcoming statement DAC.

  9. Paul, “we always want to have a low output impedance feeding a high input impedance”.

    Then why have speaker impedances in general become lower over the years? At the beginning of the hi-fi hobby 16 ohms was quite common. But that has come down over the years so than many are near 4 ohms now.

    Also, this may be showing my lack of understanding electronics but how can you discuss current without explaining voltage? I thought solid state amps are considered current sourced while tube amps are voltage sourced?

    1. Good morning M3 Lover!
      A tube amp is not just a voltage source.
      Because, if it was, then it wouldn’t be able to drive a speaker.
      You have to have some kind of currant in order to get the cone in the speaker to move back and fourth.
      Lets just say, that we have a 5 watt single ended tube amp.
      And we want to use that wattage to drive a small speaker.
      Five watts, equals 3.5 volts.
      And 3.5 volts, equals 1.25 amps.
      And all of this, is loaded in to 8 ohms.
      Remove the currant, you’ll get hardly no sound in less you put your ear real close to the speaker.
      I hope I cleared this up for you.

  10. The pipe analogy works better to illustrate current, rather than voltage. I was confused too until I realized you want voltage at the input to the downstream component. So in that regard a higher input impedence yields more voltage across the terminals, but less current. It is counter intuitive, because why would we want to impede input into our poweramps? But the voltage at the input terminals is higher with higher input impedence compared to the preamp output impedence. And if I have this totally wrong then excuse me.

  11. I really don’t understand other than making sure an amp is capable of driving Ie. Infinity Watkins woofers that dip down way below 4 ohms.

    My “certifiable” electrician pal according to the DSM IV (or is it like 7 now?) loves amps that double the horsepower when the ohms are halved but swears by running multiple speakers in series/parallel to keep them around 16 ohms because it creates an effortless load on the amp and sounds better.

    Other people don’t seem to agree with his theory. Most people want to run a 4 ohm load so they can get more power output from that same amp that doubles down.

    What does Paul have to say about this?

    1. I suppose it depends on the amp. If you’re friend is getting better sonic results at higher impedances that would suggest to me his amp is struggling at those lower impedances and higher power demands. Amps certainly sound different at different output power demand levels.

      A proper amplifier with scads of available horsepower should sound better than a series parallel arrangement. But, again, that depends. For example, a tube power amplifier doesn’t like to deliver current as much as it likes to deliver voltage. The higher the impedance the higher the required output voltage for the same loudness.

  12. I have PSAudio BHK preamp (or sometimes just DS PSAudio DAC) connected to PassLab Class A amps. Of the dozens of highly-rated high-end interconnects I’ve tried over the years including the Jorma Origo, MasterBuilt Reference, Stealth Metacarbon, Kimber Kable KS 1036, Audience Au24 SE and FrontRow, AQ Fire, and a host of other expensive cables, the best sounding in my system is the Stealth Indra, which has the HIGHEST IMPEDANCE of any high-end interconnect on the market. If I were selecting a cable with impedance as a major criteria, I would have passed over the Indra. As it turns out, the Indra pair I bought used on audiogon exceeds all my expectations and ends my search for a better interconnect. It is so clear, resolving, full-bodied and balanced, and full of energy. But strangely enough, it is not the best sounding cable in my other audio system. In that system the AQ Fire wins.

    So, the question is: how can a cable with the highest impedance on the market sound so good in certain systems?

  13. I currently fully understand the relation between voltags and omages.
    When I first started electronics edumacation, I had a difficult time visualizing a finished component circuit from the schematic, although the other way wasn’t as bad. I guess reverse engineering is easier that engineering…
    And Sir Fatrat? Your speakers my big brother are stuck at the Holding Company.
    There (as it should) ends the bad Monday groaners.

    Two tom toms and a cymbal fall down a small flight of stairs,
    Ba-doom -tsch!

    Now I’m done,
    And I’m sorry.
    Like, Canada sorry…

  14. Paul,

    It’s later in the day now so you may not see this. Impedance is frequency related as shown in todays picture. Sourcing with a lower impedance that is at least 10x lower than the load I can take at face value as a good audio engineering practice.

    In my pea brain it makes sense to me that I want the source impedance low so that I have less loss dropped across the source output so that more (most) is available at the load input.

    The questions I have is when the source or output impedance value is given, is it over a frequency range? Or is it at some nominal frequency? Is it the max value or is it the min value?

Leave a Reply

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

4865 Sterling Dr.
Boulder, CO 80301
1-800-PSAUDIO

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