1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.

    Dismiss Notice

what makes some amp better than others?

Discussion in 'Sound Science' started by hughkk, Dec 2, 2011.
First
 
Back
1
  1. proton007
    @Norman,
    The LCD2 is a planar magnetic (also known as orthodynamic) design, so the way the driver works is different from conventional diaphragms, hence the resistance is flat. This has advantages in terms of accuracy, but also costly.
    Regarding the amp - headphone matching, most music can hit peaks of 110-115 dB, for very very short bursts, based on the genre, but its not to be confused with the average level, 70-80 will be pretty loud for most ppl. It is this peak that the amp needs to sustain. So given the sensitivity in dB SPL /mW, the power requirement can be found from the formula I mention in my thread.
    Now consider an amp to be a voltage source. Normally you'll find two combinations of headphone specs:
    High Impedance, Low to Moderate Sensitivity: These are tough to drive, and will need high Vrms to get to the same peak levels. Hence an amp is recommended. Common example is the HD650, HiFiman etc.
    Low Impedance, High Sensitivity: These are meant for portable use, they'll need lower Vrms to get going, but will consume higher current, a characteristic of how batteries behave. Won't work well with amps that have high output impedance. Generally a factor of  headphon impedance being 8-10 times (or more) the amp output impedance is considered sufficient.
     
    Regarding the part about keeping the volume levels at 55 dB, the thing is you won't be able to increase it by much in case of portable devices, because they'll max out in terms of the voltage, and you'll find certain frequencies to be numbed down (not the case with orthodynamic) because of the frequency-impedance variability.
     
  2. Norman314
    @jcx
    I see, so the same equations should work for calculating the power output of any amp at a given resistance, if two values are already given to allow us to solve two simultaneous equations. I was confused because you mentioned the op amp specs in the same post where you first gave the results of the calculations.
     
    @proton
    Thanks for the additional information.
    Of course, it would not be practical to always listen to music at 55dB. But it is practical to do so some of the time, and in this case a big fancy headphone like an LCD2 or HD800 would sound just as good without an amp as with, if I understand correctly.
     
  3. jcx
    right, a Vsource with a Rseries is the simplest generic linear model for an amp
     
    if the amp has nonlinear limits due to implementation details the simple model may fail
     
    it is best used for interpolating loads, powers between the given data points - trying to estimate the power for a load outside of the published data range is more likely to fail
     
  4. proton007
    Quote:
     
    Yep, I suppose.
     
  5. Chris J
    Quote:
     
    It really depends on what headphones you want to drive, but let's say "almost any headphone, of course!"
     
    6)  flat frequency response:  20 Hz to 20 kHz +/- 0.2 dB, basically as flat as possible
     
    7) low IMD, <0.02% (intermodulation Distortion), basically as low as possible
     
    8) Max. output voltage 10 Vrms, you should be able to drive any high impedance 'phone with this figure
     
    9) Max. output current 100-200 mA, you should be able to drive any low impedance 'phone with this figure
     
    10) selectable gain, say between 0 and 20 dB
     
  6. Norman314
    Are there frequency response graphs for amps posted anywhere?
     
    And are you saying that if the output voltage or output current is too high it can lead to trouble? How?
     
  7. khaos974
    Quote:
     
    FR graphs for SS amps are basically useless, they are all flat enough.
    Too much gain (voltage) often equals higher noise.
     
  8. Norman314
    Is there a difference between boosting the gain vs just turning up the volume setting on the amp, in terms of what happens physically?
     
  9. stv014
    Quote:
     
    Yes. The gain switch usually affects a negative feedback loop, while the volume control is a simple voltage divider. In general, amplifiers have lower distortion and noise at lower gain, so you normally want to use low gain if it allows for sufficiently loud sound output. But some poorly designed amplifiers may be less stable at low gain. Also, if the volume control is after the gain stage (like in the O2), setting the gain too high results in clipping regardless of the volume.The volume setting also affects the quality somewhat if it is a simple stereo potentiometer: when it is set very low, there is often an audible channel imbalance, while the 50% level (-6 dB) setting is the most noisy (considering only the noise that comes from the potentiometer itself).
     
  10. Chris J

    No, just trying to say an ideal head amp would have enough voltage and current to drive any 'phone.
    I think those number would do it.
    Apparently, the Bryston head amp outputs more V and I.
     
  11. Chris J

    It's preferable to see a tolerance on the FR, +/- 3 dB would be excessive.
     
  12. Norman314
     
    So, in order to hit its maximum rated wattage output, does an amp have to be set to the highest gain level?
     
    For instance, suppose we calculated, with calculations similar to the ones a few posts up, that a particular amp with a particular headphone will start clipping at 110 dB. The amp has 3 gain settings -- +0, 5, +10. On the +5 gain setting, it is driving the headphone at an average volume of 95 dB, and then along comes a +12 dB dynamic peak. Does it clip at 105dB, or does it go up to 107?
     
  13. stv014
    Quote:
     
    You may or may not need high gain to reach the maximum output level, it depends on the input signal. If it is high enough, then low gain may be sufficient. Reducing the gain generally does not cause clipping at a given output level (assuming that everything before the gain stage stays the same), since the voltage limits in the output stage remain the same. So, it just makes the maximum volume quieter.
     
  14. scompton Contributor
    It's been my experience that most amps have way too much gain, even for moderately hard to drive headphones like vintage orthos.  With most amps I've tried, I'm lucky if I get the volume knob past 8 o'clock, even if there is a low gain setting.  Even portable amps supposedly designed for low impedance, high sensitivity IEMs.   With a typical 1V line level source, you really don't need much gain.
     
First
 
Back
1

Share This Page