Re: High voltage capacitors in audio

On Jun 25, 6:52 am, John Popelish <jpopel...@xxxxxxxx> wrote:
MooseFET wrote:
On Jun 24, 8:52 pm, John Popelish <jpopel...@xxxxxxxx> wrote:
ectoplasm wrote:
Thanks for explaining so clearly. That's a very good suggestion to
have the 1000uF caps next to the output stage.
The rectifier diodes might just as well be there, too, then. All
transformer leads will go there directly (15V - 0 - 15V).
No, directly to the +-22 volt supply, the unregulated one.
I think it makes little difference whether the diodes are
close to the output pair or remote.

"like" changed to "little"

I disagree with this idea. There can be a highish charging followed
by a recovery current spike. These have a lot of energy right in the
middle of the audio band. If would be much better to put this stuff
elsewhere.

The main difference
might be capacitive coupling between the secondary AC and
the opamp circuits. Having the diodes more remote keeps all
the traces near the opamps having only DC or slightly
rippling DC on them, lowering the possibilities for hum
injection.

If the bridge and the filter are on the PCB, you need to be careful of
how the ground and supply wires go.

22V
Z ------- Z ! Z ------
AC+ ------+---! !----+------+-+----+---! 7815 !---+----- 15V
! ! bridge! ! 1000! ! ------ !
--- ! ! --- --- --- ! ---
0.01--- 0.01--- --- ---0.1 ! ---100u
! ! ! ! ! !
Z ! ! Z ! -------+------+----- GND
AC0 ------+----------------+------+--------------+------------ GND
--- mirror image for AC----

The ground trace hits the points in the order shown. It is a wide
trace not a plane until you get to the right side of the drawing.

I think you need a few more Zs in there. The main load on
the 1000 uF cap is not the regulator, but the output
transistor.

I thing misunderstood the purpose of the "Z"s. They mark locations
where the impedance will be non-zero.


Where does the headphone ground return make
connection to that schematic, and where is the Z in that
path?

The headphones hook to the GND on the right. There should be no "Z"
in that path.

I would want the regulator to have its ground
reference connection be connected to ground at the point
where the two channels of headphone grounds first connect
together, not at some distant end of their common path back
to the transformer center tap. I like what you show from
transformer to storage capacitor, but not to the right of that.

The only ground in the output section is the return path of the
headphones. What you suggest here is effectively what I've done. The
regulators may be bolted to the chassis forcing the issue somewhat on
exactly how the wires go.



The points I put the "Z"s on are places where ideally, a small lossy
impedance will be in series. The 0.01uFs on the bridge are to keep RF
noise out of the system. They should be right at the legs of the
diodes.

All good.

The 1000uF capacitor does the bulk of the filtering. The 0.1 and 100u
are right on the legs of the LM7815. If the lead from the 22V to the
LM7815 is very long, more capacitance at the input of the LM7815
should be used.

I would keep the regulators close to the storage, if
possible. If this is not possible for thermal or other
reasons, than the 100 nF input cap bypasses quite a bit of
path inductance at the frequencies where it matters. That
is its purpose.

For mechanical reasons, the storage capacitors are likely to be some
distance from the regulators. In a lot of ways, it would be best if
they were nearer the driver stage than the output pair, but
mechanically this doesn't make sense.


I show two ground connections at the output because the layout may
actually be sort of like that. The return current of the output
should not flow through the ground of the preamp stage to get to the
capacitors.

As I spoke about.

Yes, I was just making it clear.




[....]
I.e. the regulators (7815/7915, not to be omitted I think) would be
fed from these two storage caps, too (through separate traces for
their minor load current).
Right. Keep those 100 nF pairs for each regulator right up
against the regulator pins, for stability.

I will add stress to the above. One inch of wire is too much between
the LM7815 and the capacitor.

Okay. I stress that! ;-)


.

Relevant Pages

  • Re: High voltage capacitors in audio
    ... have the 1000uF caps next to the output stage. ... The rectifier diodes might just as well be there, too, then. ... I would want the regulator to have its ground reference connection be connected to ground at the point where the two channels of headphone grounds first connect together, not at some distant end of their common path back to the transformer center tap. ... I like what you show from transformer to storage capacitor, but not to the right of that. ...
    (sci.electronics.design)
  • Re: caps back to back
    ... Mark wrote: ... bias on the capacitor so both caps are biased in the correct direction. ... of diodes to ensure that the caps don't get reverse biased - the diodes ...
    (sci.electronics.design)
  • Re: SP-600 Rebuilding Experiences
    ... Replacing the caps on the L/C elements for the LO may be ... disturbs the connection of the capacitor to ... I changed this cap in both of my receivers. ...
    (rec.radio.amateur.boatanchors)
  • Re: Attempting to build a surge-arrestor for my HF receiver
    ... currently have a neon bulb going from the center conductor of my coax to ... multi-watt resistor in series with these diodes. ... capacitor blown by static, and replaced it with a .1uF 630V capacitor. ... Use robust PIN diodes; ...
    (sci.electronics.design)
  • Re: A7V266-E slow to boot, (display something)?
    ... (Instead of bulging at the top like many bad caps, ... The board is also missing one of the ant-like capacitor next the ICS ... The ICS 94228 is the clock generator chip for the board. ... the harmonics off the 33MHz PCI clock signals. ...
    (alt.comp.periphs.mainboard.asus)