|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
97.84.205.208
Notice the 560uf cap coming directly off the 6CA4 rect tube...I took both of those out and put in a pair of 35uf poly film caps and left the other stock and already the one with the polys has deeper bass and cleaner mids..I want to implement a 5h or 10H choke but the 22k resistors have to remain as they load the plates on the 12AU7s.
"
Follow Ups:
right down to the component values in that line stage even though 6SN7 subbed in. PS is somewhat different as the divider network only goes two levels and the 47uF caps used as bypasses rather than decouplers.+1 on direct-coupling and I also realized great improvement replacing those 560uF as per your suggestion.
What did your B+2 come to if I may ask?
Edits: 11/15/14
The B+ was within +- 1.5v which is essentially the same.
"
and feeding the first tube of the line stage. Your B+1 on the CF looks a good +150VDC over my SLP-98 and wondering what the B+2 is on your gain stage. Sorry if not clear :(
Coming out of the 47k with the side the polys are on reads 114vdc and the side with the 560uf electrolytics it reads 111.3vdc.Part of that could be slight differences in the 6CA4s.
"
different voltages than yours with about the same B+ on both voltage amp and CF ... 193 and 196VDC respectively. Much lower on the CF and much higher on the voltage amp, yet component values the same.
Michael, I'm sure you were thinking of something else, but the 22K resistors are power supply dropping resistors and are decoupled from the tubes' plates via the 470uF caps. If you can measure the current draw of these tubes you could then substitute both 22K resistors with Hammond's 153-159 series chockes that have the appropriate current rating. These will most likely end up being somewhere in the range of 15 to 30H and in addition they're small and inexpensive units.Because the chokes' DCR values will be a lot less than the resistor values, you would then need to adjust the PSU's voltage by properly sizing the first cap. My guess for this would be anywhere from .22uF to 2uF. If you could measure the transformer's off-load voltage and DCR values at the rec sockets (with no tubes in), you could design a nice power supply having the desired output voltage in PSUDII. If and when you take said measurements, you could post them here and perhaps we could help a bit too. If not, we'll still be very interested in whichever way you decided to go.
Edits: 11/14/14
I think the biggest improvement you could make would be to directly couple the preamp stages.
After that, use all that excess voltage to build a regulated B+ supply.
You might also try not having parallel triodes up front.
The paralleled triodes on the CF and paralleling them raises the transconductance and halves the output impedance.I thought about using one triode in the CF but I hate having one nonfunctional.
"
Paralleling the voltage amp halves doesn't really do anything beneficial.
How about paralleling 3 of the triodes for the cathode follower.
The two triodes in the CF if the line stage are already paralleled..Where are you getting a 3rd CF section? I agree that I haven't heard much sonic benefit to the paralled triode.
"
Edits: 11/14/14
The voltage amp section also has paralleled triodes.
Yes but that isn't a cathode follower tho.
"
Yes, there are four triodes per channel. Two are voltage amps, two are cathode followers in the stock circuit. As a mod, you could take one of the voltage amp triodes and add it as another cathode follower triode.
Wouldn't I lose gain tho?
"
No. If you removed one of the parallel voltage amps, you'd double the cathode resistor and plate load. There would be no loss of gain.
Thanks..When I saw the absence of the plate resistors on the paralleled follower,I somehow thought at glance that the PS resistors in that divider network acted as the load resistors...I like that idea that you've presented.
"
some of the power amps with 1200uF after the rectifier and then they wonder why the rectifiers don't last and arc. The SLA 70 was like this. What is the deal with the huge caps. I put a theristpors in a few amps. Or just changed out the caps.
The Quicksilver Full Preamp was also hard on the dual 6X4's when it used them years ago. Those would arc after a while. Nice preamp though.
iBasso DX100,DX50 DX90. Chord Hugo. HiFiman 901 balanced. RSA Intruder, The Lightning. Fostex TH900 balanced, Hifiman HE-6, 560, JH13 Pro balanced. Lyr2, Audeze. All phones balanced. Whiplash cables.
Photo gallery: www.pbase.com/jamato8
John
Even in my CAry/AES set amps they have a pair of 1200uf series connected which I changed to a pair of 70uf parallel connected motorruns with great improvement.
"
The SLA 70 was a set of two 1200uF in series. What were they thinking?
iBasso DX100,DX50 DX90. Chord Hugo. HiFiman 901 balanced. RSA Intruder, The Lightning. Fostex TH900 balanced, Hifiman HE-6, 560, JH13 Pro balanced. Lyr2, Audeze. All phones balanced. Whiplash cables.
Photo gallery: www.pbase.com/jamato8
I agree with other posters that the original caps are much too large for the rectifier. Nevertheless, how do you explain the phenomenon you're describing, that the bass became more powerful with smaller supply caps?
--------------------------
Buy Chinese. Bury freedom.
Exactly and that's why I took them out and put in 35uf polys.
"
Wow, I was expecting the poly-film cap speech from you Samra where are all the statistics on their efficiencies? We all know you have surrendered yourself to poly filters:)
J
But of course,especially in a preamp.
"
OK, I've got it now. I thought this was a power amp, that's why I asked the other question. This is actually a Class A preamp. No wonder it sounds better with films!
--------------------------
Buy Chinese. Bury freedom.
Yet another fucked up Cary design that BADLY abuses vacuum rectifiers. :> (( Grrr!!!
Eli D.
~!
The Mind has No Firewall~ U.S. Army War College.
Why would they have ever put that large a cap in there?
Probably leftover supply that they wanted to use up. Garbage.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: