Speakers, All things equal - impedance sensitivity and loudness

Discussion in 'Amps and Cabs' started by waxnsteel, Oct 23, 2005.


  1. waxnsteel

    waxnsteel Member

    Messages:
    3,131
    Joined:
    May 14, 2005
    How does it work together? Is there anything I can read, or can anyone help me put this together?

    They say (or rather, it is true that) 10 db is twice as loud, right?
    As far as sensitivity goes, 3 dB is twice as loud?
    How did wattage relate again?

    As far as impedance goes, provided the amp impedance matches cab impedance, for speakers with the same sensitivity, there should be no difference in perceived volume, right? Or is there a difference when using the same amp but different impedance settings on that amp?

    And mismatching impedance, will a 16 Ohm amp output to 8 Ohm cab louder than 8 ohm to 8 ohm, or 16 to 16?
    As far as changes in impedance, is there any quantitative expective difference?

    And adding speakers? 2 vs 4?
    I'm looking more for the theoretical answers, not so much practical, though all comments are welcome. I'm not posting thinking I know something, I really like learning. Thanks.
     
  2. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    Short answer - it's complicated :).


    Long answer -

    3dB is equivalent to a doubling of power.

    10dB is equivalent to ten times the power and a doubling of volume.

    Speakers of the same sensitivity don't always sound the same volume, because their frequency response isn't even, and that at which the sensitivity is measured may not be totally representative of the overall response. Also, the human ear is more sensitive to mids (especially upper mids) than other frequencies, so speakers with prominent upper-mid peaks (eg Celestion V30) tend to sound louder than ones with less mids (eg G12H-30) even if they're the same sensitivity on paper. 'Complex' sounding speakers with a lot of cone coloration (eg G12M-25) also tend to sound louder than 'cleaner' ones (eg G12H-30) for the same sensitivity as well. Lightly-damped speakers sound louder than heavier-damped ones, since they respond faster to transients where a lot of the perception of volume comes from. All these things are more noticeable in a mix than in isolation too.

    Power rating isn't really related, it's simply a measure of how much power you can put into it without causing damage. Obviously a high-powered speaker can usually create more total volume, since you can put more power into it... but sometimes they don't sound as loud at low power since the cones are often stiffer and heavier.


    With tube amps, the most power is always at matching impedance. You typically get about 30% less (1-2dB) with a factor-of-two mismatch in either direction, but the tone is different in either direction (generally mushier for a low mismatch and flatter for a high one), so they don't necessarily sound the same volume.

    (With solid-state amps, you get less power - exactly how much less varies depending on the amp design - with a higher impedance than the minimum, and although you would in theory get more power with a lower impedance, if you try it you'll overload the output transistors and almost certainly blow them.)


    Adding more speakers (assuming correct impedance matching) does seem to increase volume even though the power is now shared between more speakers so each is driven less hard. Exactly how much depends a lot on the cabinet design and maybe even the room it's in though - and only about 1-2dB from double the number of speakers; although you usually get better dispersion with more speakers, so the amp might sound quite a lot louder from off-axis.
     
  3. GCDEF

    GCDEF Supporting Member

    Messages:
    25,670
    Joined:
    Jun 17, 2005
    Location:
    Florida
    Very good answer John. There's a lot of confusion on this topic, but you nailed it.
     
  4. Soundhound

    Soundhound Member

    Messages:
    2,276
    Joined:
    Feb 24, 2004
    Could you explain ohms as it relates to speakers? I know nothing, nothing I tell you! I've got a couple of amps and am tryng to figure out some replacement speakers and perhaps add an extension cab or two and have no idea of what ohms are, how different speakers can have different ohm ratings, how it all adds up, what it means, how different cabs can have differernt 'loads'. This would be much appreciated!
     
  5. waxnsteel

    waxnsteel Member

    Messages:
    3,131
    Joined:
    May 14, 2005
    Thanks for that, John. In the original post when I asked about wattage, I was reffering to tube amp output wattage. I'm trying to get better at asking questions. Thanks for sharing the knowledge. Much appreciated.
    So you're saying a mismatch in either direction should result in lower overall volume, and a character change? Cool. Good to know. I have work to do, and **** to buy! Thanks again.
     
  6. GCDEF

    GCDEF Supporting Member

    Messages:
    25,670
    Joined:
    Jun 17, 2005
    Location:
    Florida
    That's true with tube amps. With SS amps, too high an impedence will reduce overall volume. Lower impedence will increase power up to the capabilities of the amp or the limits of the speaker. Too low an impedence and you'll hurt your amp, or at least trip its breakers.
     
  7. waxnsteel

    waxnsteel Member

    Messages:
    3,131
    Joined:
    May 14, 2005
    Right, like with PA gear, and don't go <2 Ohms on the K2's. Gotcha. I don't mess with solid state guitar amps much. I blew up a Gorilla once, and I still have this Danelectro Nifty Fifty thing, but I really don't mess witth it much.
     
  8. frank62

    frank62 Member

    Messages:
    5,375
    Joined:
    Aug 16, 2005
    Location:
    Mechanicsburg, Illinois
    does this mean a speaker rated 99db is only half as loud as one at 102db?
     
  9. GCDEF

    GCDEF Supporting Member

    Messages:
    25,670
    Joined:
    Jun 17, 2005
    Location:
    Florida
    No. It means it would take twice the power to output the same volume. Your ears would perceive the difference as fairly small. "Twice as loud" or "half as loud" requires a difference of around 10 dB.
     
  10. frank62

    frank62 Member

    Messages:
    5,375
    Joined:
    Aug 16, 2005
    Location:
    Mechanicsburg, Illinois
    then a speaker that is more efficient than another will not produce any significant increase in real or percieved volume?
     
  11. Scumback Speakers

    Scumback Speakers Gold Supporting Member

    Messages:
    9,507
    Joined:
    Oct 6, 2003
    Location:
    Los Angeles, Ca
    Wrong. More efficient speakers produce more volume because they're more sensitive, or, if you will, have greater sensitivity.

    Example: I put two M class (M75 Scumback/G12M style) speakers in a 2x12 closed back cab. I run the amp I normally use at the local blues jam on 6 with them (30 watt amp). That carries well in a 3000 square foot club (40' wide by 75' deep).

    Next week I put two H class (H75 Scumback/G12H30 style) speakers in the same 2x12 closed back cab. I run the amp at 4 to compensate for the extra sensitivity the H's exhibit over the M's.

    M's are typically 96-97db. H's are typically 99-100 db. Therefore, using the same amp to speaker output ratio, a 3db increase in speaker sensitivity from the M's to the H's results in setting your amp's master volume (or regular volume) down about 1/3.

    So, in real world tests, the difference between M's and H's in volume is about 1/3 with all factors being the same between them, other than the change of speakers.

    So, if you're running an amp and want to get it running hotter so that it hits it's "sweet spot" try running M class speakers so that you have to increase the volume, and run the tubes hotter.

    If you're running a low watt amp and want the most volume, run H class speakers to get an increase in volume.

    M's are more midrange focused, and have some flubby low end if they're not in a closed back cab.

    H's are more full range, and have better low end, tighter, and more defined, and can have slightly sharper highs due to their more sensitive cone/magnet combination.

    Make sense?
     
  12. laurencer83

    laurencer83 Member

    Messages:
    409
    Joined:
    Jun 4, 2005
    Location:
    half way to everywhere
    can i ask a quick question about what impedance i should be running my amp? i'm playing through a dr z RXES head which only has one output per ohm selection. i want to run two cabs, both of which are wired in parrallel to 8 ohms each. my orange cab has two inputs so that if you wanted you could run another chord into another cabinet(daisy chain). and this is what i'm doing.

    so i have my RXES head running into an 8 ohm orange 2x12 cab, running into an 8 ohm harry joyce 2x12 cab.

    i should be running my RXES at 4 ohms right?

    the only reason i'm unsure is because i'm used to amps that have an impedance selector like old marshalls. and i didn't know if if it made a difference since i'm using my orange cab as a daisy chain.
     
  13. GCDEF

    GCDEF Supporting Member

    Messages:
    25,670
    Joined:
    Jun 17, 2005
    Location:
    Florida
    Depends how much more efficient. 1 or 2 db is negligible. 5 or 6 is quite significant.
     
  14. drbob1

    drbob1 Silver Supporting Member

    Messages:
    20,957
    Joined:
    Aug 9, 2004
    Yes.

    John, you said earlier that the output volume of an amp would be highest at the rated impedance, but it's actually a bit more complicated than that, isn't it? If I read my electroncis manual there's a range of output impedances that a tube will tolerate. I forget the formula, but doesn't the highest undistorted output require something like 1/2 the impedance of the highest total (distorted output), so the amp designer choses a compromise somewhere in between. So a mismatch might increase volume slightly at the cost of more distortion (or increase clean overhead at the cost of less volume)?
     
  15. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    I'm not sure... as far as I know the highest clean (as in non-clipped) power and the highest maximum distorted power are both obtained at matching.

    But I do think I remember reading that running into higher impedance loads is often prefered for hi-fi, because it reduces non-linear distortion below the point of clipping, even though it's at the expense of actual headroom. I could be wrong.

    It would figure with what I hear as the sound of a high mismatch though - a flatter, clearer, more clinical sort of tone. A low mismatch gives a softer, mushier, but more 'organic' sort of tone.
     
  16. Tone_Terrific

    Tone_Terrific Member

    Messages:
    24,026
    Joined:
    Dec 26, 2004
    Location:
    Canada-GTA
    FWIW- as an added complication, speaker impedance varies with frequency (non-linearly) AND various cabinet designs affect impedance, hence power output, not to mention acoustic output, multi-speaker interaction, and the way the speaker load ineracts with the amplifer, AND don't forget that a guitar signal encompasses a wide range of frequency at any one time.........totally mind-boggling!!!!

    If it sounds good and doesn't blow up you're OK.
     
  17. scottlaned

    scottlaned Member

    Messages:
    648
    Joined:
    Mar 12, 2007
    Location:
    Richmond VA
    I think you just sold about 20 speakers by writing that.

     
  18. rockon1

    rockon1 Member

    Messages:
    11,456
    Joined:
    Aug 12, 2005
    Location:
    Stamford CT
    FWIW- Twice as loud is harder to quantify as it is very subjective. Most people percieve it to be anywhere from 6 to 10 db louder. Bob
     
  19. Dr. Tweedbucket

    Dr. Tweedbucket Deluxe model available !!!11

    Messages:
    44,169
    Joined:
    Feb 22, 2007
    Location:
    Mt. Kwakkleberry
    An easy way to double check yourself is, after you have your cabs set up the way you like them, before you plug your 1/4 jack into the amp, take an ohm meter and see what your load reads.
     
  20. rockon1

    rockon1 Member

    Messages:
    11,456
    Joined:
    Aug 12, 2005
    Location:
    Stamford CT

    Yep-but remember resistance will be slightly lower than the speakers actual impedance. In other words using a meter an 8 ohm load will read 6-7 ohms ,a 16 ohm load will read 13-14 ohms.Not sure about a 4 ohm load but I imagine around 3 ohms. Bob
     

Share This Page