When Things Get Loud
What happens as you turn up the volume
There is really a bunch of things coming into play when things get loud, most of them not good. Since there are a lot of them, this will be a long post (I apologize).
There a bunch of non-linear effects that happens in your own brain and auditory system. A very clear indication that the human auditory system is not really designed for very high sound pressure levels is the fact that speech intelligibility drops rapidly with sound pressure level (all other things being equal) and is generally very poor above 100 dB SPL. Here are some of the physiological reasons why this is the case:
Equal Loudness Curve
1. The first one is the "equal loudness curve" (sometime also referred to as Fletcher-Munson curves) that Steve has already mentioned. The main effect is that you have to roll of the low end as level go up (or boost it as the level goes down). It basically compensates that the way humans turn sound pressure (physical property of sound) into loudness (a perception in our brain) is really not the same for all frequencies.
Acoustic Reflex or Middle Ear Reflex
2. The second is the “acoustic reflex” or “middle ear reflex”. It’s a build-in protection in the middle ear that kicks in at sound levels above about 80 dB SPL. Normally sound is picked up by the ear-drum and then transferred through three tiny little bones in the middle ear to the inner ear. These bones that are basically a system of levers that transport vibration in a certain way. If it gets too loud a tiny muscles in the middle ear contracts and changes the coupling between the bones. It basically makes the levers less efficient and that protects the bones themselves and the inner ear from potentially harmful sound level. That fact also gives you some appreciation of what “evolution” regards as excessive sound levels and for what type of sound pressure levels the hearing system is basically designed for. So basically the acoustic reflex is a non-linear reduction in sound pressure level and its also frequency dependent
The Inner Ear
3. In the inner ear, vibrations are converted into never impulses by the so-called “hair cells”. These cells get wiggled by the incoming sound and as a result of that they send a nerve impulse. These impulses are all-or-nothing, i.e. they all have the same strength. More sound level leads to the hair cells firing more frequently, but not to any changes in pulse amplitude. Now the hair cells have a maximum firing rate, above which they just can’t react to more stimulus anymore. The only way to detect more sound pressure is to look at the behavior of a large group of cells which is much less exact as if every single cell would still be operating in its linear range.
4. Another effect is called “masking”. Its due to the way that vibration is transferred from the middle ear to the hair cells (through the so-called basilar membrane). Normally every hair cell is tuned to a specific frequency, but since the tuning is not perfect it will also react to frequencies that are not. The result of this is that the perception of a certain tone can be influenced by the presence or absence of a tone at a different frequency. The effect is highly asymmetric, i.e. low tones can mask high tones but high tones cannot mask low tones. Again this effect is more pronounced at higher levels and a single string bass note can drown out a lot of other sounds.
In summary, above a certain level you will operate your auditory system at a point it wasn’t really designed for and that will result in a lot of “internal distortion” that interferes with clarity, speech intelligibility and overall enjoyment. In other words you are overdriving your input.
There are also some problems with the sound system and the room.
5. Nearly all loudspeakers these days are so-called “dynamic” loudspeakers; i.e. they run current through a moving coil in a static magnetic field. The basic principle has a non-linear component to it, which gets worse as the sound level increases. You can actually operate a loudspeaker WITHOUT a magnet and you will get to hear only the non-linear component.
6. There are a few more mechanisms that can generate distortion at higher levels: these include voice coil heat up, excursion limits, non-ideal magnetic fields, non-linear (i.e. overextended) surrounds and spiders, etc. Off course in a high quality loudspeaker such as ours, these things are all carefully designed, controlled and optimized but they are still all present and may under certain circumstances at very high stress levels product audible effects.
7. In many rooms something (ceiling tiles, curtain rods, stuff in a drawer etc.) can rattle when excited hard enough and that contributes to background noise
8. When you play louder, the audience is likely to talk (or yell) louder (to have a conversation or simply to order another drink) and that again raises the level of the background noise.
9. If there are microphones (or acoustic guitar pickups) involved, these will not only pickup the intended sound but also the sound radiated from the loudspeaker. If you keep cranking that ultimately leads to feedback, but even way before that point these picked up sounds will change the frequency response and time signature of the original signal
So turning up does not only makes it louder, it also has a variety of unwanted side effects. In the end, it is an artistic decision, but maybe being aware of all the things that can or will happen can help making in making this decision.
Hope that helps