[
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2008, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]
24.174.113.158
Got together Saturday with a couple of audiophile friends, they had a couple of PP EM7 amps (these little amps sound wonderful), a single tube 12SX7 bread boarded parafeed preamp, and I brought a big bag of Russian caps.
I had helped a bud cob together this 12SX7 bread boarded preamp, with what I had laying around, so we put a 4uf 1000V KBG as a parafeed cap, and a 100uf 1000V paper and oil as the last cap in the parafeed power supply.
I've been thinking for some time that the RC time constant as being important in sound quality, and that we probably had way more capacitance than was really necessary, in both the power supply and the parafeed cap.
Conventional thinking is that the parafeed topology shifts the signal from the last capacitor in the power supply of a single ended design to the parafeed cap. If that's true, I wouldn't expect much change in sound from changing the last cap in the power supply of a parafeed design. When we changed the 1000V 100uf MBGO with a 400V 20uf MBGO, the change was VERY obvious and positive. The preamp was faster, more detailed, better/more bass, and sounded MUCH more dynamic. This one change was way more dramatic than the other experimenting we did.
We also did a coupling cap shootout on the PP EM7 amps. We had some Solen type caps, some Goodman paper in oils (glass sealed), some US made teflon caps made for the oil field, and some Russian SSG silver and micas. Unfortunately we aren't rich enough to buy some teflon VH caps to compare. The Russian silver and micas were the obvious winner here.
We also went down in size on the value of the Parafeed cap. We replaced the 4uf 1000V KBG with a 2uf 400V KBG. The pre sounded faster, tighter, and more dynamic, and did not lose any bass response. Calculations indicated that I could go down as low as .2uf, due to the high 90K load of the nickle output transformer, so I replaced the 2uf KBG with a .22uf 600V Russian FT-3. The pre sounded more detailed and didn't seem to lose any bass, but it just didn't have the flesh and slam as the 2uf KBG. We then went to the 2uf KBG bypassed by the .22uf FT-3, very nice.
What I found odd was how much effect playing with the last cap in the power supply had on this parafeed topology. There must be a pretty strong signal on that last cap in the power supply to have that much of an effect. One day I'll hook up a signal generator and an oscilloscope and find out, unfortunately we didn't have the equipment at that location. Going down in value of both the parafeed cap, and last cap in the power supply, lend some support to the RC time constant theory.
My take on power supplies are somewhat similar to what Jeff and Dennis say, but with a difference. My take: use just enough L to keep the supply from ringing, just enough C to keep the amp from humming, and as little DCR in the inductor and transformer as you can afford. This gives the supply as low of an RC time constant as is practical. Also it helps if you can use a paper and oil type caps in at least the last stage of the power supply.
twystd
Follow Ups:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2008, Audio Asylum, all rights reserved ]