Power Supply for jfet line Stage

[maxpou]

Hi all,
I want to try a new power supply for my new line stage, but i'm not sure if i have enough filtration in location C5, 6, 7 and 8. My new line stage is in this thread.

no current mirror, current mirror, cascoded current mirror, bjt or jfet VAS?

thank you! Maxpou

[Zen Mod]

enough

[maxpou]

sorry for my bad english, but i mean if i need more filtration or it's ok, because i found a 1000uF ROE EKO and i think it's a good capacitors? Thank you! Maxpou

[Zen Mod]

Originally posted by maxpou
sorry for my bad english, but i mean if i need more filtration or it's ok, because i found a 1000uF ROE EKO and i think it's a good capacitors? Thank you! Maxpou

I meant - it's enough with that amount of capacity you show on schematic ;

of course - more is always better

[AndrewT]

what's the quiescent and peak current draw?

I'm guessing around 20mA and 30mApk.

If as little as this, then 1mF is just fine.

[maxpou]

Hi Guys,
thank you for replys, this is my PCB. comment please? Maxpou

[nigelwright7557]

Originally posted by maxpou
Hi Guys,
thank you for replys, this is my PCB. comment please? Maxpou

More copper !!!
Connect GND to a power plane that fills in all the unused copper.

It decreses the impedance of the tracks and also uses less etching chemicals.

[Mitch Alsup]

A) the two AC inpouts should be farther apart to ensure easy soldering with no possibility of shorts. The AC "wires could be made larger (i.e. beefier).

B) The ground "wire" runs down the center. What would be ideal is for every component to have a wire directly to the common point of all grounds (star patern). However, the easy fix is to change this 'wire' into a solid layer filling all available space in the cernter with a couple of cutouts for the components that intrude on this 'bus'.

C) The V+ and V- "wires" should be beefed up similar to those of the ground system, but section at a time. For example, the wire connecting components around C6 should be approximately a parallelogram.

D) I, personally, would be tempted to gun the grounding point out and under the diodes and capacitors near the AC inputs such that the whole AC part of the system is isolated with a ground strap. Thus, the wire from point PGND is extended to the left and then splits and runs under the Ds and Cs at least to the edge (top/bottom) of the board. Thus isolating the AC section.

E) I would be tempted to rotate R3, R4 90 degrees and move r5, r6 to the right allowing the zerars to move outwards, simplifying the route in the mid section of the supply.

[AndrewT]

Hi,
Mitch's comment B must be adhered to or at least the dirty grounds from the smoothing end MUST have a separate route to the star ground AT THE OUTPUT.
I would recommend that the regulator ground also has a separate route. That makes 5 legs to the star on the PCB side and a further leg for the output side.

I think a ground star will perform better than a ground plane at keeping the dirty charging pulses out of the clean grounds that determine the output ripple.

[jackinnj]

Shouldn't the 1N4148's be reversed?