Hi:

John does indeed present an interesting (although not compelling to me at
least) case for the common cathode bias resistor. Don't have time for a lot
of discussion here but note that he didn't get around to actually verifying
his theories in a practical sense.

Overall, I'd agree that if taken to the extreme of an infinite number of
tubes, the single bias resistor may have benefit in eliminating what amounts
to undesirable positive DC bias feedback. It might also be argued that it
converges on fixed bias with it's improved power capabilities in AB
operation. However, four tubes is very different from infinite and I found
from actual measurement that the need for a matched quad (in terms of bias
current) remains (positive bias feedback remains). The fact that Dyna even
recommends matched quads for the ST/SCA35 is evidence that they were aware
of the practical limitations of the scheme. Regarding power output
capabilities, I measured no practical difference between four tubes on a
bias resistor vs two tubes per resistor. This was on an ST35 with
continuous signal, measuring both one channel driven and both channels
driven. Transients like music may provide different results. However, with
two tubes on a resistor I was able to achieve much better DC balance which
probably contributed to a little better low end performance because OPT core
saturation effects were reduced.

If you read the other archived posts I think you'll conclude like I did that
this is a controversial subject with theory on one side and practice on the
other. Fortunately the cathode resistor mod is really easy on the AA151 so
all you need to do is try it and trust your ears. Interestingly I ended up
with the single cathode resistor on my 151 but increased the bypass cap to
220uF AND use a quad of tightly matched EL84s. On my ST35 I ended up with
a cathode resistor for each pair of matched output tubes each bypassed by
220uF. On this one I tried everything inc fixed bias for four tubes, for
each pair and individually for each tube. Still came back to cathode bias
because of sound even though the fixed bias schemes produced a lot more
power. I also added a filter choke to the PS which made a huge difference
in hum and noise when using efficient speakers.

--
Over and out
Steve