ARRGH!!!
Still humms very much. Connecting the 2 boards together helped a little but the humm still persists.
I redid the output wiring shortened and very tightly twisted.
I even tried another PSU board I had kicking around using MUR860 diodes in a CRC arrangement.
Must still reside in the grounding scheme.
Still humms very much. Connecting the 2 boards together helped a little but the humm still persists.
I redid the output wiring shortened and very tightly twisted.
I even tried another PSU board I had kicking around using MUR860 diodes in a CRC arrangement.
Must still reside in the grounding scheme.
Here is my approach to FW amps.
I believe ZM taught me this.
Take all ground wires from Speakers , line signal and and solder all to PS ground at back end, In the middle of the V+ and V- outputs. In that same area, take CL60 and tie that to chassis ground with bolt, and from there, bolt wire to IEC power cord ground.
So a CL60 seperates everything from direct chassis or IEC ground.
From here, shorting the inputs should lead to a quite amp. If not. Twist toriod, or cover toriod and or the Jensen's with MU-Metal (careful not to short).
Is your setup different?
I believe ZM taught me this.
Take all ground wires from Speakers , line signal and and solder all to PS ground at back end, In the middle of the V+ and V- outputs. In that same area, take CL60 and tie that to chassis ground with bolt, and from there, bolt wire to IEC power cord ground.
So a CL60 seperates everything from direct chassis or IEC ground.
From here, shorting the inputs should lead to a quite amp. If not. Twist toriod, or cover toriod and or the Jensen's with MU-Metal (careful not to short).
Is your setup different?
nice
now - do the same , but with just one xformer
Damn. Why didnt you say something before I drew it.
didn't had that idea when we exchanged these few e-mails
it just came to my mind , when I saw your pcb
like always - I need a pen and paper to be able to think , but didn't had time tonight
but , I have some feeling it's possible , using nasty tricksess
(remember that you're having 4 same windings in one xformer ...... and that you are going to drive entire shebang differentially )
( so - I'm semi serious , this time ; am I allowed to use cap , or two ?
edit : if not , then you might see my usual square mile pcb , with plenty of parts ...... while just few are touching the signal
)
it just came to my mind , when I saw your pcb
like always - I need a pen and paper to be able to think , but didn't had time tonight
but , I have some feeling it's possible , using nasty tricksess
(remember that you're having 4 same windings in one xformer ...... and that you are going to drive entire shebang differentially )
( so - I'm semi serious , this time ; am I allowed to use cap , or two ?
edit : if not , then you might see my usual square mile pcb , with plenty of parts ...... while just few are touching the signal
)
Last edited:
I know, and just as I thought had completed something and was going to rest easy. Gollum, Gollum.
I will try mine first to see. I will be thinking on yours. No rest for the weary
Main reason I like the idea is possiblity of no feedback(especially with cascode FE) on local parts with only fancy feedback to clean it up. Dunno, Might be stupid or stupendous. OR should that read stupindous
I will try mine first to see. I will be thinking on yours. No rest for the weary
Main reason I like the idea is possiblity of no feedback(especially with cascode FE) on local parts with only fancy feedback to clean it up. Dunno, Might be stupid or stupendous. OR should that read stupindous
Last edited:
you should try that with two already made pcbs
matter of one wire
or not even that ; careful measurement will only tell is it any gain in common feedback arrangement ( for two halves of balanced ) vs. regular fdbck against real gnd scheme
think of function of McMilan resistors ; in some balanced configurations/parts values combos they're more effective , in some are almost zilch effective
matter of one wire
or not even that ; careful measurement will only tell is it any gain in common feedback arrangement ( for two halves of balanced ) vs. regular fdbck against real gnd scheme
think of function of McMilan resistors ; in some balanced configurations/parts values combos they're more effective , in some are almost zilch effective
Last edited: