Why do some speakers sound better with more power available?

Discussion in 'Audio Science' started by schnesim, Sep 15, 2020.

  1. schnesim

    schnesim New

    Joined:
    Jul 4, 2019
    Likes Received:
    23
    Trophy Points:
    8
    Location:
    north
    I hope this is the right subforum for my question.

    In some reviews I read
    And in others quite the contrary
    So I've been wondering, why is it, that some speakers seem to require more power even if I never exceed say 5Watts with my listening levels?
    Is it maybe the damping factor?

    Thanks
     
  2. SineDave

    SineDave Friend

    Pyrate
    Joined:
    Apr 18, 2016
    Likes Received:
    862
    Trophy Points:
    93
    Location:
    Houston, TX
    Home Page:
    That's a loaded question - but a good one. I'll do my best to answer before the much smarter folks here come in and tear my answer to pieces and correct me :D

    TL/DR: It's a combination of things.

    Speakers have an electrical impedance they present to the amplifier which is often not quite what is advertised (note all speakers are a 4 or 8 ohm load, so the value is often rounded) and can vary quite a bit. On top of this, each speaker has a sensitivity factor that denotes how easy they are to drive. This used to be measured in SPL/w/1m - so sound pressure produced by 1W at 1m. This is now changing to SPL/2.83V/1m - which is a more accurate measurement that good designers are adopting.

    Some designs are highly sensitive and will produce >90dB of sound driven by 2.83V at 1 meter, while others are quite hard to drive, at <85dB sensitivity. Give this a read if you're interested in more detail: https://www.audioholics.com/loudspe...dspeaker-sensitivity-engineering-vs-marketing

    Keep in mind that the dB is logarithmic, a 6dB increase in measured sound pressure level (SPL) reflects a perceived doubling in volume, but a 3dB increase requires a doubling in power[Edit: thanks @spwath for the correction]. More on that here. This means the difference between 91dB sensitivity and 85dB sensitivity is the difference between comfortably driving a speaker with a 5W SET amp and needing a 150W amp. When someone says a low powered class A amplifier drives inefficient speakers with ease, you need to define inefficient (because chances are the speakers in question are not *really* that inefficient), and also question the ears of the reviewer. There are a few speakers that are notoriously hard to drive (Apogee, Thiel) and no matter what some reviewer says, will sound like hot garbage when fed with moderate power.

    I feed my ATC SCM19's which are 85dB sensitive with a 1500WPC amp. I likely never use this kind of power, but when listening at higher levels, transient (momentary) power demand can be much greater than average power demand. That sudden kick drum hit might require 10X the amplifier power for a fraction of a second, and if the amp can't deliver that - it will not reproduce the transient as accurately. This is much more common for low frequency transients, so home theater nerds like me are very concerned about amplifier headroom (excess power for these situations). Here's a good summary on that topic: https://pro.harman.com/insights/har...of-using-power-amplifiers-that-are-too-small/

    Damping factor is typically higher in amplifiers with higher power output, but is not a result of power but output impedance, wire length and the speaker's impedance or load. I think many folks hear tighter bass and assume it's damping factor at play, which is not like always the case.

    Ultimately, tight bass (removing acoustics from the equation) is a function of controlling the moving mass of the transducer. This is the result of the electrical ability to put the brakes on a woofer that is already at peak excursion and bring it back to a resting state. There are a ton of ways amp designers try to do this, but in general you'll find the best engineered amps typically have this consideration in common. Which is why a 20W Schiit Aegir sounds better than a 20W tripath amp from Parts Express.
     
    Last edited: Sep 15, 2020
  3. Hrodulf

    Hrodulf Prohibited from acting as an MOT until year 2050

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    4,276
    Trophy Points:
    113
    Some amps are pretty indifferent to nasty swings in electric phase/impedance, therefore their operational parameters won't go to shit when they meet wild loads. Generally you want an electric device to be completely indifferent to [sane] external conditions.

    As for power requirements, continuous power requirements are orders of magnitude smaller than peak power. For example designing an amp for 500W continuous power is vastly different than making one which can provide such power for millisecond peaks. And there's a good chance that the capacitor banks on the 500W continuous amp can blast out 2000W peaks, depending on available voltage.

    P.S. The good thing about 500W amps is that they can play 5W fine, unless very noisy. Making a 5W amp do 500W is much harder.
     
  4. spwath

    spwath Hijinks master cum laudle

    Pyrate BWC
    Joined:
    Dec 13, 2015
    Likes Received:
    7,894
    Trophy Points:
    113
    Location:
    Madison, WI
    One thing, its a 6dB increase in measured SPL that reflects a perceived doubling in volume.
    A 3dB increase results from a doubling of power.
     
  5. SineDave

    SineDave Friend

    Pyrate
    Joined:
    Apr 18, 2016
    Likes Received:
    862
    Trophy Points:
    93
    Location:
    Houston, TX
    Home Page:
    Thanks for the correction.
     
  6. Dzerh

    Dzerh Friend

    Pyrate
    Joined:
    Apr 1, 2018
    Likes Received:
    425
    Trophy Points:
    63
    Location:
    San Ramon, CA, US
    And I think 6dB is actual doubling, and you need 10dB for perceiving.
     
  7. Priidik

    Priidik MOT: Estelon

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    2,153
    Trophy Points:
    113
    Location:
    Estonia
    Mostly covered by the good posters above.
    I will add that amp that has say 200Wrms capability to 4 ohms will have much beefier power supply than a one with 20Wrms for the same load.

    Apples to apples with no clipping:
    With adequate design of the amp, damping factor would be in the first sight low enough for either amp to control the drivers. However, the damping factor stated is for steady state and often not even the frequency is given. To eat the back emf from a large woofer, this means high damping in dynamic mode, not steady state. Probably the most reliable way to get that is strong psu and lots of parallel output devices (ie low source resistance all the way).
    Good design with barely adequate psu can handle this too with loads of global neg feedback, but somehow the low global neg feedback designs sound better.
    The contradiction is that low power amplifiers in general tend to sound better everywhere but low bass, due to several reasons, but mostly they are more simple.
     
    Last edited: Sep 29, 2020
  8. Hrodulf

    Hrodulf Prohibited from acting as an MOT until year 2050

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    4,276
    Trophy Points:
    113
    Hence flea power for mid/high

    -and-

    FATTY BOMBA 500W Class-D for bass/sub.
     
  9. Azimuth

    Azimuth FKA rtaylor76, Friend

    Pyrate
    Joined:
    Nov 9, 2015
    Likes Received:
    6,838
    Trophy Points:
    113
    Location:
    Nashville, TN
    Home Page:
    I will also add to the Class A post from the OP, that many "Class A" amps have low or no NFB (negative feedback) which results in a very low damping factor amplifier. This is not a good match for a speaker with a high damping or low efficiency. More efficient speakers mate well with speakers with low damping factors. This is why many "flea watt" tube amps sound great with high efficient Klipsch or horn type speakers. And some other speakers with lots of stuffing and a huge crossover network need tremendous power with a high damping factor to push through all of that.

    The other issue is distortion. We all know about Pass's "first watt" theory in that it is only that first watt that is important. Say you have two amps, one rated at 45W and one rated at 100W, and they have the same THD spec, basically that spec is at that actual high wattage. You only need a few watts to get up to decent listening levels, and therefore the higher wattage amp will have lower distortion at that same SPL level as the lower wattage amp (given, like I said before, both having same distortion spec at their rated wattage output). And this is a gross generalization, I know, since there are many amplifier designs, but I am not saying one will sound better, just saying, it will very likely have lower distortion specification. Whether this translates into better actual sound, YMMV. It still needed to be said.

    The last thing I will mention is the difference (which is also somewhat related to the previous paragraph) between RMS and Peak power. I had a friend who always purchased amps 2x the size of the max rating of the speaker. Meaning, if the speaker RMS max rating was 100W, he would get a 200W RMS amp. Why? Because he knows that the speaker manufacturers under rate their speakers and peak power is what matters for transients and for sound not to be dynamically strained. Hardly ever were those speakers done for. AND, speakers are more likely to blow their voice coils due to under-power rather than over-power. Under-power causes more distortion (square waves) and clipping, which results in more heat in the coil by pushing the cone and holding it there, and the amp cannot provide enough movement of the driver to increase airflow and cool down the coil. I know of more than one set of PA on a stick speakers catch on fire because someone tried to hook them up to a home stereo.

    All this, and really the second post by @SineDave is more correct and damping is one of the more mis-understood things in hi-fi.
     
  10. Priidik

    Priidik MOT: Estelon

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    2,153
    Trophy Points:
    113
    Location:
    Estonia
    Then there is the Dartzeel NHB 108.
    Uses [(*1*)] complementary pair of output devices and still clocks about 150 W to 8 ohms (resistive).
    Appears to be in high regard; I can confirm the clone has good amount of SET purity and magic.
     
    Last edited: Jan 30, 2023
  11. OJneg

    OJneg The Most Insufferable

    Pyrate BWC
    Joined:
    Sep 25, 2015
    Likes Received:
    3,923
    Trophy Points:
    113
    Location:
    Grand Rapids, MI
    Others have already made great points. The potential for a higher-rated amp to sound better/different has as many possible answers as the potential for an equivalently rated amp to sound better/different. Momentary power, damping factor, NFB architecture. lower THD @ the applicable ranges, and so on.

    Something that I have noticed is that higher power amps tend to have higher gain. Which makes sense given that the industry operates on standard assumptions of what line-level audio is coming in, and what speaker loads are going to be driven. You need more gain to drive more power into an equivalent load. When someone swaps out their lower power amp for a higher power amp they may get the first impression of booming bass, increased explosiveness, transient sharpness, etc. Even once you make the listening levels equivalent, you have still changed your gain structure significantly. You'll likely be operating a different point on your attenuator, and your preamp will not be swinging as much voltage on its output. I think smart audiophiles need to pay attention to the overall gain structure of their signal chain from start to finish.

    I think there's also something to be said regarding high-gain devices or circuits having a different sound in of themselves. Those who have played around with rolling otherwise compatible vacuum tubes with different mu/transconductance values will understand this intuitively.

    A fun thought experiment.... if you tried to hold all other things equal, you might start with a baseline amplifier design and make only the required modifications so that it could qualify for a higher power rating (say 2-4x the baseline rating). You would probably increase the power supply capacity (more power reserves). You may need to increase the rail voltage itself (potentially more linear operation?). You would probably parallel more output devices (more current output capability, lower open loop Zout) which could change linearity as well. To get more gain without a revised driver stage, you might actually choose to use less NFB. So would the revised amp sound the exact same as the baseline amp, given that the demands of the load are unchanged and both are well within their linear ranges? Well they'd probably be closer than an amp with a totally different architecture, but it's not hard to imagine that the necessary modifications might produce a difference the listener could pick up on. But trying to pin it on any of the changes specifically might not be so easy.
     

Share This Page