Slewmaster - CFA vs. VFA "Rumble"

[Bigun]

Originally Posted by ostripper

Having the PS in the middle would be worst of 2 "evils".

Having 2 sensitive Vbe's crossing high current returns

If you need to compromise in the end due to space, one option is to have small high quality caps local to the output so that high frequency currents are kept to short local loops. Then use larger caps in the middle which will handle lower frequencies - and capacitive coupling of small voltage drops produced by currents flowing in low resistance wide traces to any 'sensitive' traces crossing above will be somewhat ameliorated.

[ostripper]

If you need to compromise

(The circuit's) design is enough of a compromise.

Layout is real , I'll make the board 75mm X 200mm... if need be.

OS

[zenelectro]

Originally Posted by ostripper

Thanx ...

I really want advice , have not built a good CFA yet.
The NFB path can not (should not) "snake" around the
sensitive predriver/Vbe area. Shown are the wires (in pink/red - below).
On a double sided PCB one could bring it down along a rail - later date.
Is there a better way ?
OS

Run a wire link on the back of the board for the FB return if you want to keep it
separate.

Z

[apexaudio]

Pcb of my 'slewmaster' 100x80mm

[Bonsai]

In my nx-Amp layout I just put the FB resistor and gain setting resistors all very close together at the diamond summing junction and ran the connection to THD output as a dedicated track to just before the Zobel (advice from D Self). The clean and dirty grounds are separated by 10 Ohms. I terminated the feedback lower resistor on the clean ground side.

I have no issues with either the nx-Amp or the sx-Amp. Both are exceptionally quiet and sound wonderfully clean and open.

[ostripper]

Originally Posted by zenelectro

Run a wire link on the back of the board for the FB return if you want to keep it
separate.

Z

That is what I will do .
If this turns out good (goes to a PCB house) , a real track will be
added.

OS

[ostripper]

Originally Posted by Bonsai

In my nx-Amp layout I just put the FB resistor and gain setting resistors all very close together at the diamond summing junction and ran the connection to THD output as a dedicated track to just before the Zobel (advice from D Self). The clean and dirty grounds are separated by 10 Ohms. I terminated the feedback lower resistor on the clean ground side.

I have no issues with either the nx-Amp or the sx-Amp. Both are exceptionally quiet and sound wonderfully clean and open.

I saw Jim's audio NX PCB.
You have the array of resistors in parallel (the divider) near the OPS and
only run the divided Voltage back to the NFB point.

I have a bit more "acreage". My input stage will be inches from any
high current areas .... Capacitance multipliers will be between drivers/ops
and input stage.

Does anyone think 77 X 200mm for the 3 pair or 77 X 250mm for the 5 pair is
too large of a layout.
Once I do the 3 pair layout - I can extend to any mumber of Op pairs in
sprint ... very easy.

OS

[Krisfr]

Can you put a connector in for a separate front end supply and a jumper to take out the back end...

5 pair is about right and 250mm is great at least for me. Doing it single sided means many can build test and LISTEN.

[ostripper]

Originally Posted by Krisfr

Can you put a connector in for a separate front end supply and a jumper to take out the back end...

5 pair is about right and 250mm is great at least for me. Doing it single sided means many can build test and LISTEN.

Already designed cuttable trace/big pads for boosted supplies.
Ground scheme plotted.

Dirty ground - G1 =
Outputs , main decoupling caps ,driver caps, zoble.

Clean ground - G2 =
onboard cap multipliers , Vas load ground , input ground.
PS - Cordell's suggested 50w to-220 feedback resistor is circled in middle (3W or the 50 watter).
OS

[Bonsai]

"You have the array of resistors in parallel (the divider) near the OPS and
only run the divided Voltage back to the NFB point."

It looks like that but they are physically close to the summing junction. The boards are quite small. You have to watch out for capacitance from the feedback junction to ground or across the feedback resistors otherwise you can get gain peaking.