Gain of line preamplifier

Ralph, thanks for your characteristically informative, knowledgeable, and detailed input. It all makes perfect sense to me.

One small nit, though:

A typical cable might have 15pf per foot. With a 3 foot cable and a 100K volume control set halfway across the scale, this results in a rolloff (-3db point) at 70KHz, meaning that phase shift artifact is going all the way down to 7KHz.

I think you are assuming that the halfway setting would result in a 50K output impedance. But provided that the output impedance of the source component is small in comparison, and assuming that the two ends of the pot are in shunt rather than in series with the signal source, it seems to me that the impedance seen looking back into the output of the pot would be close to 25K at that setting, and the 3 db point would be around 140 kHz.

10-28-14: Czarivey
Almarg, what year/century is it now?? I thought that today we mostly use direct coupled outputs unless it's a tube preamp(many already use direct-coupling as well).

Note that my comment to which you were responding specifically referred to tube preamps. I think it is safe to say that a substantial majority of tube preamps utilize coupling capacitors at their outputs (Ralph's designs being notable exceptions). And as can be seen in the measurements John Atkinson usually provides in conjunction with reviews in Stereophile, it is not uncommon for the output impedance of tube preamps having capacitively coupled outputs to rise from a few hundred ohms at mid and high frequencies to 3K or 4K or even more at 20 Hz.

10-28-14: Czarivey
Unity gain solid-state preamps such as McCormack, Wyred will kick S to all passives ez and swapping A/B is the best way to hear and know. It's as easy and similar as to swapping Chevy Aveo with BMW Z4 roadster.

***In properly configured system*** I think that a better comparison would be between the Bimmer and a Porsche Boxster or Cayman. I have never taken sides in the never-ending active vs. passive controversies because it seems clear to me, based on what I perceive to be an overwhelming preponderance of the anecdotal evidence, that either approach can work well in a well chosen and properly configured system.

Regards,
-- Al

Thanks Al, I can think of a few more nits, for example analog volume controls are not linear, so the halfway point is anything but half of the control! I was assuming a traditional potentiometer, which is a series-shunt configuration. Our output impedance of the source is an unknown however. I did point out that this problem has more to do with higher resistance value controls than those of lower resistance (10K being my example).

I am sure that variables such as this as well as the input impedances and cable capacitances are really pointing to the need for a database for best results.

Atmasphere what if now you throw in a shunt 42 stepped attenuator into the mix. Now the input impedance lowers as the volume is lowered. This will roll off the bass especially using tube phono stages or tube cd players.

Just a raw shunt by itself does not make for a good volume control due to the problem you just described. The source really needs to see a constant load regardless of the volume control setting, otherwise the control will take on tone control aspects.

Atmosphere do you know of a 42 stepped attenuator with only two resistos in line at each click that is not a shunt attenuator?