Originally Posted by MagnumX
1) You shouldn't need that much power driving 13 channels at once. Every doubling of speakers adds 3dB output to the room at the same power level (more in a corner). That is the equivalent of doubling amplifier power. Thus, what you need to drive 2 channels to 105dB is 2x what is needed to drive 4 speakers of the same efficiency to that level in a typical room assuming equal distances. 2)8 channels would be 1/4 of the power to drive two, etc. That assumed an average level, but it tends to work out. 2) Think of it as the same total power either way. If 150 Watts gets you 105dB with two, 150 total (or 75 each) will get you 105dB with four on average. With 13, you'd need right around 23 watts per channel to get about the same output level as 150x2. In reality, channel requirements will vary by how many are playing at once and their level, but the overall point is you won't need anywhere near 150x13 or whatever its max is in two channel mode.
What is significant about 105 dB SPL as the sum total of the outputs from all the speakers?
Originally Posted by MagnumX
2) It's really just an example to illustrate how more speakers mean less power per speaker so one should not be overly concerned if the maximum it can put out is 60-75W with all speakers driven (that's actually pretty impressive). However, I did pick the number because it is the Dolby recommended peak reference level for the primary speakers for a movie (sub should peak at 115dB). That's typically for movie theaters. Most people find it too loud for their home theaters, but it's a number many aim to hit anyway at home so they can state they can play at reference levels.
105 dB SPL is the reference level for each full-range speaker (115 dB sub.), not for the total of all the speakers as you used in your example. This may confuse readers.
This is why reference level is so loud, even with only a few speakers hitting close to 105 dB SPL. It is also why powerful, multi-channel amplifiers, with substantial power supplies are required to play at even close to reference levels, with excellent sound quality, in systems with lower sensitivity loudspeakers and/or large numbers of speakers.
1)(referenced to above)The sound sources for each channel can be considered incoherent since in a home theater installation content played on each full-range speaker is mostly uncorrelated and the speakers are spread throughout the room, which adds phase differences. This means that adding the output from two identical speakers at half the power of one speaker will provide the same sound output as one speaker; not +3dB.
For incoherent sources this calculator and its explanations applies:
Adding multiple speakers, and reducing the input to each speaker, greatly increases the current requirements of the installation without adding output. The whole idea of dividing power between the speakers as described is also not correct and provides incorrect answers, see below.
2) Superposition is appropriate for voltage and current, but not power, which is not linear and is calculated using a power of 2. It is not correct to divide power as you have done.
Amplifiers are voltage sources that attempt to maintain the required voltage by supplying whatever level of current is required. This current has be supplied by a power supply and is the major reason why AVR's can't maintain their two channel rating as more channels are driven.
In reference to your example:
150 watts into 8 ohms requires 4.33 amps RMS or 8.66 amps RMS for two channels
. These current needs have to be fulfilled with proportionate DC outputs from the power supply.
23 watts into 8 ohms requires 1.70 amps RMS for one channel or 22.0 amps RMS for 13 channels
, with a proportionate DC output from the power supply.
A much more substantial power supply is required to supply current to meet the 22.0 amp vs. 8.66 amp requirement, if the amplifier channels are to act as decent voltage sources. Getting something for nothing isn't happening here. Actually adding the speakers gets less than nothing based on the effect it has on power supply loading. Just as an example, almost three times the capacitance will be required to maintain the same ripple level from the power supply at 19.9 amps vs. 7.8 amps. Clearly a bigger power transformer is also required and the output stages will generate more heat so larger heat sinks will also be required. For many AVR’s, the DC current draw from the power supply resulting from the load from the 13 channels, is close to triggering the protection circuitry. Protection, depending on the unit, may involve turning on fans at various speeds, cutting the rail voltage, or ultimately shutting down the unit.
The above example also shows why adding channels, even channels in the ceiling that don't appear to put much stress on the amplifiers in an AVR, actually do add stress. Supplying those first few watts (actually amps as noted, watts will lead one astray) to a speaker places a significant added load on an AVR's power supply, if say four channels are added. Move to speakers with a lower sensitivity and/or a lower impedance and the demands increase again.