How is this figure supposed to be interpreted in the datasheet for the LM4766 audio amplifier, and determining the needed power requirements?

Question

First off, I really have no experience when it comes to audio electronics, so forgive me if these are naive questions.

In any case, let me explain what I'm doing/trying to do. I have two speakers that can hand 25 watts (RMS) with a 50 watt max. The impedance of each is 8 Ohms.

Ohm's law gives us the simple equation P = V^2/R, so V = 20V when we're at our 50W max. Also, the speakers would draw 2.5A each.

I obviously need to amplify my audio source, and to this, I plan on using the LM4766 from Texas Instruments (Datasheet). According to the datasheet on page three, the supply voltage is defined to be |Vcc| + |Vee|, where Vcc is the positive supply voltage and Vee is the negative supply voltage. The max supply voltage is 60V.

Now here's my question. In figure 18 on page 10, pictured below, Output Power vs Supply Voltage

is plotted. I find this plot somewhat ambiguous. Is it saying, for example, that if Supply Voltage = 25V (Vcc = 12.5V and Vee = -12.5V), then the output power is roughly 27W? Or is it saying that if Vcc = 25V and Vee = -25V, then the output power is roughly 27W?

Any input would be appreciated.

Answer 1

Please note the horizontal scale of the figure "SUPPLY VOLTAGE +/- V". This means that a 25 volt supply consists of +25 volts and -25 volts.

Your application of Ohm's Law is a good start, but you need to learn a little more about AC measurements. For a sine wave (and audio is always represented as such) the peak voltage is sqrt(2) times the RMS (Root Mean Square) voltage, and it is the RMS voltage which counts when computing power. So the most (audio) power a +/- 25 volt supply can provide with an 8-ohm load is (25 / 1.414) x (25 / 1.414) / 8, or 39 watts. Peak current will be 25/8, or 3.12 amps, while RMS current will be 25 / 1.414 x 8, or 2.2 amps. Of course, no real-world amplifier can produce a high current all the way to the supply rails, so the real power will be somewhat less than this. The graph showing 37 watts at 25 volts with a decently-low distortion figure indicates that the folks who designed the amplifier really knew what they were doing.