Thursday, April 7, 2011

It’s The Power Supply, Stupid!

Mainstream audio engineers as usual may beg to differ, but how much does power supply design and topology affect audiophile grade audio amplifier sound quality?


By: Ringo Bones


Charles G. Whitener, Jr., president of Western Electric Audio Products wrote his views on this particular subject matter which got published in the December 1997 edition of Stereophile magazine which goes: “We have come to this really arrogant conclusion: When tubes are removed from the chassis, the amplifier becomes nothing more than an expensive power supply.” There may be some truth to what Whitener had said – whether vacuum tube based or solid-state designs, it has been experienced by seasoned audiophiles time and time again that that power supply design and topology does play a role in the empirical assessment of sound quality of a typical audio power amplifier marketed to audiophiles. But is there really something more going on here?

The topic of power supply design and topology and its effect on the subjective sound quality of a “well-engineered” audio power amplifier – like the design and topology of the analog output stage of a universal CD / DVD player – has always been and probably will always be a politically charged topic to “mainstream” audio engineers. The issue is often seen by mainstream audio engineers like a mandatory sexual harassment / gender sensitivity training aimed at their machismo driven chauvinist little corner of the universe. But given that audiophile grade audio amplifiers – unlike mobile phones – are aimed at hi-fi / audiophiles who listen to them with their own two ears instead of connecting them to a 22,000 US dollar Fast-Fourier Transform audio analyzer, should mainstream audio engineers pay more attention to the power supply design and topology of audio power amplifiers?

Ever since the latter-half of the 1980s, audio engineers who still care about the sound quality of their designs and final products have come to conclude that the overall impedance of the power supply is a much bigger problem than anyone has previously thought. Electrocompaniet’s Per Abrahamson in a January 1998 edition of Stereophile magazine interview says that the power supplies of 99% of the amplifiers on the market are designed with big capacitors, big transformers, big everything – without taking into account the frequency response of the power supply.

When the impedance of the power supply is all over the place, the problem manifests itself as a problematic unnatural sounding midrange and high-frequencies in the finished audio amplifier during audition. The impedance of the power supply is a mirror into the subjective frequency response – but you won’t find anything wrong with the audio power amplifier if you measure the frequency response by sinewave measurements. No problem even via a 22,000 US dollar FFT test gear. If you do tests by auditioning with well-recorded music and via your own two audiophile-seasoned ears, you can hear that the power supply design and topology does affect the frequency response. Audiophiles, musicians or anyone who gives a rat’s ass can hear it but we can’t measure it with fancy 22,000 US dollar FFT gear.

So if you are fortunate enough to have the opportunity to peek into those audio power amplifier designs famed for their sound quality, I just hope that the ultrafast Schottky rectifiers or those 15-nanosecond HEXFRED diode rectifiers, banks of electrolytic capacitors instead of a single dustbin-sized 100,000 microfarad capacitor won’t pose as an audio engineering culture-shock. All of these hi-fi tweaks serve to make a relatively modern sensibly-priced solid-state audio amplifier aimed at audiophiles to sound like their tube-based Golden Age of Stereo counterparts.