Is DEQX a game changer?

Bombaywalla,

Excellent discussion of power supplies. Thank you for that!

welcome Roscoeiii

Well said, Bombaywalla. In audio as in many other technically complex fields, "it depends" or "maybe" is often the correct answer. We all hunger for yes and no answers but there is reason why "the devil is in the details" is a well-worn phrase. I know nothing about output ripple (although I'm pretty sure it's bad). What I do know is that several of the designers of products I admire devote a lot of attention to power supply design and execution. In my technically ignorant opinion, it's one of the reasons why the idea of buying an amplifier based on weight, as crazy as it sounds, is not such a bad idea. Good transformers and power supplies are usually pretty heavy, so all other things being equal (not comparing Class D to Class A, and normalizing for the weight associated with "machined form a solid billet of unobtanium"), there is a positive correlation between amp weight and SQ. IMO.

Yes, an excellent post by Bombaywalla, and an excellent response by Michael (Swampwalker). Thanks, gentlemen!

I am in full agreement with both posts, aside from what I believe is an inadvertent and minor misstatement in Bombaywalla's post:

A linear power supply is expensive from a power dissipation perspective - you have to design its max voltage for the max peak voltage of the program material but in normal operation the linear power supply operates mostly at the average voltage of the program material. The difference in the peak & average voltage is dissipated as heat. Of course, you don't know what the max voltage of the program material is so you have to over-design further leading to more heat dissipation.

Shouldn't it be the output of the amplifier that has to operate mostly at the average voltage of the program material, not the output of the power supply? With the difference between the average output voltage and the voltage supplied to that stage (which as you indicated has to provide headroom relative to the maximum anticipated output voltage), multiplied by current, corresponding to the heat dissipated in the output stage, not the power supply? Although the heat dissipated in the power supply will also vary with current demand. And although there are a few amplifier designs in which the output voltage of the power supply is actually varied among a number of discrete levels as a function of signal level, some of Bob Carver's older designs being examples.

Again, though, an excellent and informative post. Thanks!

Best regards,
-- Al

Hi Almarg,
yes, you are correct - misstatement on my part. Sorry guys!

indeed the linear power supply operates at its designed output voltage (which is over-designed to ensure it will accommodate the max program material voltage) & the extra power dissipated will be (power supply voltage to amp output stage - voltage of program material) * current. And, correct again, the power amp output stage mostly operates at the average voltage of the program material.