tube rectifiers, HexFreds, and the Big Deal

[Bill Layer]

Hi All,

>>But why? I've read several times that tube rectifiers sound better than
>>solid state ones, but never found an explanation for it. Isn't the
>>output impedance of the power supply dominated by the capacitors?
>
>...at high freq, and indeed audio, this is prob true....
>but, at VERY low freq, I suspect the tube has higher internal resistance.
>Which means that if you are using a class B output stage, and you go
>WHANG on the guitar, the B+ voltage rail sags.
>Which gives one of those strange coloration effects so beloved by tube fans.
 
Very true for guitar amps - the signal's abilily to modulate the power
supply plays an important role in the sound of some vintage designs. But
the improvement in sound quality using tube rectifiers is another issue
entirely, as it relates to modern audiophile practice. And we really should
dispell this notion that all tube fans crave distortion, and coloration. 

The sound of tube rectifiers goes to a much deeper level than Zout of the
power supply. The issues at hand are those of AC power factor, and diode
switching artifacts. Power factor is a way of quantifying of distortion of
the 50 or 60 Hz wavform on the AC powerline. Distortion can be caused by
sevral sources, the most common being inductive or electronic loads.
Electronic loads, especially solid state, cause heavy peak currents, with
abrupt switching. Not only does this distort the powerline waveform, but
the diodes themselves radiate noise and RF energy which also effects the
sound. 

In any event, as distortion of the powerline waveform increases, the out of
phase harmonics begin to cancel (attenuate) the current of the fundamental,
and thus the line loses it's abilty to deliver full power. Under the right
conditions, it might be possible to measure nominal line voltages, yet not
be able to draw productive power. There is an excellent article on this
subject by Marty DeWulff in "Bound for Sound". 

Now in contrast to the harsh, abrupt nature of solid-state rectifiers, tube
diodes are soft, gentle and quiet. They generate little or no noise or RF,
and are equally kind the the powerline waveform. There IS a solid state
near-equivalent to the tube rectifer, it is called a HexFred diode. I use
these in all of my audio amps, and the difference is HUGE. If you don't
believe me, call up digi-key and order a few; they're only a buck or two
each. Use them to replace the 1N4004s in your whatever, and then tell
everyone what you think.

The implications of this are of course, massive. Using tube rect's or
HexFreds not only improves the sound of the particular piece, but helps out
the entire system by preserving the integrity of the power waveform, and
reducing noise pollution. Computers are a serious offender here; it's no
wonder that at Atma-Sphere and Mapleshade studios, all digital electronics
are shut down during recording or mastering sessions. If you own computers,
try that one yourself...

Finally, if you're still not convinced, try this one. Warm up you stereo
(the cheaper, the better...) and listen to a short piece that you've heard
many times. Pay special attention to anything that strikes you: parts which
are bright or annoying, parts which come through well, with feeling. Places
where the bass seems the most liquid.

Now, pull the plugs and give the prongs a quick wipe with light oil - like
LPS or even WD-40. The thinner the coat the better. Plug-in and go around
again with the same piece; if you aren't sure what you just heard, clean-up
and do the test again (might be tricky in same outlet...). I'll be
interested to hear your reactions, and theories on what happened.