Sony VFET Amplifier Part 2

Jump to Latest
Zen Mod

Zen Mod

Official Court Jester
Joined 2003
zenmod.in.rs/
Paid Member
I believe loading coming from OS itself is sort of variable in nature (node jumping up and down .... so bias circuit need to have impedances as low as possible ; example - take biasing of small tube or transistor vs. biiiiig tube or transistor ) , while value of R20 need to be chosen to tame/neutralize that as much is possible

so , as input impedance of OS and nature of it is result of compromise , value of R20 need to be in same range , to became dominant

that's at least how I understand it
 
Last edited:
Botte

Botte

Member
Joined 2004
Paid Member
front end board size

Tea-Bag said:
I am working at this point to produce a front end board that would be acceptable to eventually go into the diyaudiostore. The backend of it I think will come later once the dust has settled a little. This is of course if papa approves, or has nothing else in the works.
Click to expand...

Hi Mike,

I already have a dual mono chassis I'm planning on using, do you know what size your front end board would be? I have to decide whether to forge ahead point to point or wait for a board.

Thanks,

Chuck
 
Tea-Bag

Tea-Bag

Member
Joined 2006
Paid Member
Botte said:
Hi Mike,

I already have a dual mono chassis I'm planning on using, do you know what size your front end board would be? I have to decide whether to forge ahead point to point or wait for a board.

Thanks,

Chuck
Click to expand...

I do not, but there is no reason to make it large. If possible, an adapter board on current output board is an option I thinkttoo. Others have PM'd interest in using as a line amp without to many mods.

I need to have a talk this weekend with 6L6 about compliance with chassis format if we want to obey that at all in this case. I think the majority will need to adapt any compliant chassis with an L-channel aluminum like in your amps. For a back OS end board, looking to do something that will use TO-3 or TO-247 for IRF's Anyways, use of a bench drill is likely.
 
W

WalterW

Member
Joined 2012
Paid Member
If I want to make a balanced amp like the big Sony VFET can I just connect R24 ( 332Ohm) to the R24 of the other halve? Just like in the BBA3?

Anybody know how to connect the global feedback switch when going balanced?

I have single layer PCB layout ready for a balanced or stereo pair, and will be etching my own boards. It's on half a 'eurocard' size ( 10 x 16 cm).
Is there a 'USAcard' size also :D:D
I made it with a very old freeware program, LAYO1PCB, so I don't know how to share it with you...


Walter
 

Attachments

  • VFET FE.jpg
    VFET FE.jpg
    343.4 KB · Views: 1,044
delecoy

delecoy

Member
Joined 2011
Tea-Bag said:
I do not, but there is no reason to make it large. If possible, an adapter board on current output board is an option I thinkttoo. Others have PM'd interest in using as a line amp without to many mods.

I need to have a talk this weekend with 6L6 about compliance with chassis format if we want to obey that at all in this case. I think the majority will need to adapt any compliant chassis with an L-channel aluminum like in your amps. For a back OS end board, looking to do something that will use TO-3 or TO-247 for IRF's Anyways, use of a bench drill is likely.
Click to expand...

Would there be a benefit to having both channels of j313/k2013 fets in the row ( like the pumpkin ), as opposed to on opposite ends of the board ( like the BA-3FE? I ask only because I already have a galaxy 2u dissipante I could put this in and mounting all the matched fets to the same sink seems convenient, at first glance.
 
generg

generg

Member
Joined 2009
Paid Member
Zen Mod said:
I believe loading coming from OS itself is sort of variable in nature (node jumping up and down .... so bias circuit need to have impedances as low as possible ; example - take biasing of small tube or transistor vs. biiiiig tube or transistor ) , while value of R20 need to be chosen to tame/neutralize that as much is possible

so , as input impedance of OS and nature of it is result of compromise , value of R20 need to be in same range , to became dominant

that's at least how I understand it
Click to expand...

THX Aleksandar!

I reread the article and Nelson does not write that the input impedance of the outputstage is low, but more that the output impedance of the Frontend is low, done by R20 2k2 and the two AC loads (4k7 and 220uF going to ground too).

So the input of the outputstage sees the less 1k of the output of the inputstage.

Sorry for long version.....:)

Nelson's is better....

"You will notice R20 which is a load to ground from the Drains of Q15 and Q16. R5, R6, R29 and R30 provide some additional AC loading to ground, and all told, the output front end sees slightly less than 1K to ground. These help set the gain of the front end along with R24, R25 and R26 and also R43/P3 with R44/P4."

And often asked.....already when noisefree asked about the 30.5 some years ago.....

Is the frontend direct coupled over the 47k to the outputstage or cap coupled over the 220uF. But these are connected over the 4k7 to ground.....:-(

Nobody else reflecting the circuit....?
and showing his ignorance...
:)
 
Last edited:
You must log in or register to reply here.
Top Bottom