13 December 2012

sound practices reading club /47

And the beat goes on. Literally, because the article that we read today is a response to last weeks’ article. We are reading Secrets of the Phono Stage by Allen Wright, from sound practices issue 15. In it, Allen presents a nine-point plan to improve phono preamp designs, specifically Diego’s Φ42. After that he introduces his Four Valve Preamp design, discussing the fet cascode input stage and the bootstrapped kathode follower output stage.

teaser quote: ‘Now normally a 1dB shift is not easy to hear, but such an error in one of these bands can make a whole system become light and bright (or dark and mysterious) because it moves at least two whole octaves of music bandwidth around! Although not a cure-all, split time constant EQ is the way to go […]’

my take
The nine points that are raised in the article are of Allen Wright as we knew him. On rereading the article I notice that some of these points go deeper than their title would suggest. #6 (no redundant stages) is overall an encouragement to think in big pictures, even beyond the output connectors of the preamp. #7 (circuits before components) is part of a hierarchy that I detailed before.

Point #8 (no electrolytic caps in the signal path) talks about input and output loops—together with the illustration that is also at the top of this blogpost. If reading that, so many years ago, did not start me off thinking about current loops in circuits, then it surely gave it a big boost.

I have some remarks about some of the points. #3 (accurate RIAA): why does one need channel–to–channel matched caps when all the EQ resistors around them need to be trimmed anyway? Differences in tube plate resistance will give rise to more 0.x dB channel imbalance than the difference in cap value. If you want to be real clever, you assign the lower cap value to the tube with the higher plate resistance, and trim from there—and re-trim regularly…

#5 (star-wire everything) correctly talks about B+ and ground in the same breath, because one simply has to be a mirror image of the other. This means that choosing star grounding—or as Allen did, a multi-level star, call it a tree—topology forces that there is only one choice for the B+ topology: one B+, arranged as a (multi-level) star. Allen illustrates this in his FVP schematic. I personally think that planning the other way around is more logical, flexible and straightforward: fit the B+ topology to the needs of the circuits, and your ground will follow.

pop quiz
What happens when the preamp gets muted by short-circuiting the output of the kathode follower to ground? Its kathode is now AC grounded and the fet on top keeps the plate voltage constant. The only thing that can move to deal with a signal on the grid is the grid–cathode voltage. To do that the tube swings current, quite a bit of it.  The gm tells you how much: 11.5 mA per volt input. Better not turn up the volume on some music, and then mute…

bonus tracks
See also sound practices issue 8 for an excerpt from Allen’s tube preamp cookbook.

Now go and read the article, see you next week.

No comments:

Post a Comment