Power supply upgrades

What value bypass caps should those electrolytics be bypassed with?

I don't think it's particularly critical, as long as the bypass caps are of good quality. Anything from 0.1 to several uf strikes me as reasonable. If you find indications of particular values that worked well for people, in the forums you spoke of, go with those, but quality is probably more important than value.

Rodman -- As I think you indicated towards the end of your post, I'm not sure that the improvements you cited, which were mostly with power amps, would be relevant to a phono stage's power supply, where the current load is much more stable as well as much lower.

Regards,
-- Al

If he could find some Black Gates to directly replace the caps he has, there would definitely be an improvement with regards to dynamics. Some of their old suppliers still have the smaller values in stock, although at a premium. Upgrading to Black Gates, or Sanyo OS-CONs will improve dynamics installed in any circuit that provides gain. I believe he could easily find them for his phono pre. If the pre has rectifiers, replacing them with soft recovery FREDs will yield even greater benefits. These things just hold true, regardless of the power supply's application.

Rodman, my problem in replacing caps with Black Gates is firstly their size, the size of the circuit board is seriously limited. My old Cary monoblocks were all point to point wiring, much easier to swap out ps caps. Secondly, Black Gates are mostly lower voltage, 160v highest for 100uf, 350v for .22uf (I suspect enough here). Is the 160v rating high enough in this application?

I have investigated swapping out rectifiers, the problem is severe space limitations. I have two bridge rectifiers, RS804 and BR310, building my own from FREDs looks impossible to fit.

Caddocks are on my list of possible resistor upgrades.

Al, I'm willing to try premium electrolytics, at this point I'm leaning towards trying the bypass solution to begin with, if that works, fine, if not, try the upgrade electrolytics route. At the very least it would be an interesting experiment. What do you think about the 160v rating on 22uf caps, enough?

I've been doing a bit of investigating on the new Takman 'ultra premium' resistors. Sound very promising considering the few who have tried them. These would go in the signal path. Anyone have experience with these?

What do you think about the 160v rating on 22uf caps, enough?

I don't know what voltage the power supply is outputting to the plate circuits of the 12AX7's, but I suspect you should go with a considerably higher rating on the caps than 160V. Perhaps the 12AX7's are being run somewhat lower than 160, but probably not much if anything below 120 or 130V. And the 12AX7 and 12AX7A have rated plate voltage maximum's of 330V, so in the absence of further information it could very well be that they are even being run a bit above 160V.

Good derating practice in my experience as an electronics design engineer (although it was in defense avionics, not audio!) was considered to be a factor of 2 or more. Meaning use capacitors with voltage ratings of 2 or more times the actual voltages that are likely to be encountered; otherwise long-term reliability is degraded.

And, finally, the 450V rating of the existing electrolytics was probably chosen for a reason!

Regards,
-- Al