Accustic Arts Power 1 - operates in Class A?

Just wondering if anyone knows if the Accustic Arts Power 1 integrated amp operates at all in Class A at low levels or is it purely Class A/B? I ask because the wattage draw on my PS Audio P3 AC regenerator never changes even as I increase the volume. I need to go to ear splitting levels before I see a change.

Anyway, just curious.
IIUC, by definition all class A/B amplifiers are class A at very low wattage. Some makers will bias their amp so that it is in class A even at moderately loud volume levels. In that way the amp is only operating A/B at very high volumes.
Thank you. I think based on the wattage consumption not moving at until I reach high volumes that the Accustic Arts Power 1 must be biased to Class A at least for the first few watts. Does anyone know for sure?
To the best of my knowledge (as I recall speaking to the people in Germany), their top of the line amps are biased for a few watts in pure class A. Your lack of additional wattage draw registration from your PS Audio is due to the amps' massive storage energy bank, and perhaps the relatively benign load of your speakers. Just my two cents.

Best regards.
On one hand crossover distortions are reduced at slightly higher level when amp stays longer in A class biasing (higher bias) but at the same time gm doubling distortion at higher levels are increased. Measurements made by Douglas Self, Bob Cordel and many others show the lowest distortions at particular bias (optimal bias). Overbiasing is not the trade between quality and power dissipation but it will increase distortion. Mentioned gm doubling is change in gain of amplifier's output stage between class A (two transistors working) and class AB (one transistor working). Perhaps there are ways to minimize effect of gm doubling (trans-conductance doubling) but increasing bias in amplifiers not designed for that might lead to increased distortions.
It's a class A/B amp. I had one a few years ago.
Kijanki, I just recently read a thread on this very subject on DIY. This discussion revolved mainly around BJT outputs and there was an unanswered question regarding mosfets. However, in applying this school of thought, I'm perceiving a sonic difference when backing off the stock 300ma bias setting in my Acoustat TNT200 mosfet amps. Can you please elaborate on this? Thank you.

I guess my question is whether the same principals apply or am I hearing things?
How hot does it run? How big are the heatsinks?
Even a 50 watts per channel at class a is going to be a little oven. If it is not HOT, like you can barely touch the fins when on and no music, then you got class "A".
If it is not a small room heater when at idle, no way is it full class "A"
And yes ALL amps are biased a little into class A before going into class AB.
My Bryston is about 15 watts/channel 8 ohms class A bias.
For normal 300 watts 8 ohms.. 5% is the general percentage of bryston bias to class A (my understanding)
Thank you, Elizabeth. But my question is not whether it's class A. I already know it's A/B. I was questioning whether(optimal bias) as regards BJT outputs applies to mosfets.
Csontos, I don't know. Things I've read apply to bipolars only. In my opinion it should also apply to Mosfets but design might be completely different possibly equalizing gain. Assuming that overbiasing of Mosfets might reduce distortions the question would be "by how much". The main problem I see with class AB is that because of inherent nonlinearities feedback is usually 10x deeper than one in class A amps. This feedback might create odd order harmonics because of Transient Intermodulation. Once you have this feedback set in your amp by design changing bias won't help TIM. Your improvement in distortions might be miniscule and not worth the heat dissipated not to mention possibility of increased distortions if we got it wrong. Increasing bias current and heatsink size seems to be inexpensive proposition and if designer decided against it I wouldn't touch it.
What I've done is reduce it slightly to around 290ma from 300. The thread I'm referring to concludes it's not quiescent current that matters but instead the voltage drop across the emitters is the important number. So optimal bias with bipolars may be a reduction in current after all. But the mosfet amps I have don't have emitters and must be biased at the rails.