Class A bias, speaker sensitivity, watts...?

I'll tell you what I think I know and someone can correct me.  Class A watts approximately halve from 8 ohms into 4 ohms, so you have between 5 and 10 watts of class A power. 

Accuphase does the calculation of watts being output at any given time, but they say the calculation is complex.  Here is their explanation:

Indicates the True Power into the Speaker

The power into the speaker is the product of the amplifier’s output voltage (V) and the current (I); that is, W = V×I. While the voltage is easily measured with a voltmeter, the measurement of current is more complex. Typical power meters avoid this complexity by assuming that the load resistance is equivalent to the speaker's nominal impedance. For these meters, then, I = V/R, and therefore, W = V×V/R, and so W = V2 /R. But as Figure 1 clearly shows, the actual impedance of the speaker varies complexly according to the frequency. The load resistance changes considerably as the impedance moves from peaks to dips, causing the power to change as well. Because typical meters measure only the voltage, they are unable to calculate true power at any given time. The A-80 meter, in contrast. uses detection circuitry, installed at the outputs, to dynamically detect both the load voltage and the current. The meter converts these analog readings into digital values that are then used to calculate the power. It can therefore display the true power value even as the power fluctuates along with the impedance

They have a microprocessor in their amps do the calculation. 

Just a guess, but I think that in a small office with 80 db peaks you would stay in class A watts most of the time and probably wouldn't notice if you went into class B watts.

@richardbrand

Yes, I agree with what you’re saying.  I was trying to address a different matter.

I was told by an amp manufacturer whom I called (don’t remember which one) that the class A watts are specified at 8 ohms and would approximately halve into 4 ohms. In my case it was 25 watts in class A into 8 ohms, but that dropped to 15 watts in class A into 4 ohms. I assumed other amps operated the same in that matter. Is that your understanding?

@richardbrand

Yes it would be interesting to hear from some high bias Class A/B amp builders whether their amps switch over to Class B power sooner as speaker impedance drops. Maybe I just misunderstood.

@richardbrand

There is only one number given for Class A watts that I’ve seen, which is why it stuck with me when the guy said they dropped into 4 ohms. It’s nothing to spend your time on, If someone who knows for sure comes across this thread maybe we’ll find out, but until then, forget I mentioned it.

@richardbrand

I found an explanation for, or at least support for, the idea that Class A watts halve as speaker impedance halves. It’s from the December 2023 Stereophile review of the Accuphase A-300 amp. Her’s the link and the relevant quote:

Accuphase A-300 monoblock power amplifier | Stereophile.com

I also wrote to Duffey, who relayed the question to Inokuma and forwarded his response. "Here are our thoughts and technical approach to the question. For a push-pull output stage using a bipolar transistor as the output device, it operates as class-A up to twice the idling current flowing to the output device when the output is zero. If more than this amount of current flows through the speaker, one of the output devices, operating as a +/– pair, will turn off. Of course, current can be supplied to the outputs without any problem, and this is called class-AB amplification.

"In the case of the A-300, the idling current is applied so that the class-A range into 8 ohms is 125W. The amplifier operates as class-A up to 62.5W into 4 ohms and 31.25W into 2 ohms."

This inverse relationship between class-A power and load impedance is easily understood when you consider that current—idle current—determines an amplifier’s class-A range and that power equals current squared times load impedance: Cut the impedance in half and the power is halved as well. "So, the A-300 does in fact provide 125W of class A power into 8 ohms," Duffey wrote. "Into 4 ohms, though, the amplifier’s fixed amount of ’idle current’ can support just 62.5W of class-A power." Inokuma’s response included a table relating impedance, rated power, class-A range, and maximum power at clipping rated at 1% THD. The table showed that the clipping power easily surpasses the rated maximum output power at each load impedance, reaching 1100W into 1 ohm. How much power does a person need?