Need help on RC filter for power supply

[pinkmouse]

moved.

[QSerraTico_Tico]

Nounours,
Don't go higher than 10000uF on the caps directly connected to the rectifier bridge.
More will put too much stress on the transformer and rectifiers. (Horowitz page 330)
If you want lower ripple add extra RC section(s) as in Wintermute's schematic.
I agree on your findings about musicality. Did not dare to post it

[DF96]

The 'snubber' network may reduce resonance problems caused by the small capacitors bypassing the big ones. The proper place for a bypass is not at the PSU cap, but on the amp circuit board. The resistance of the cabling then helps damp any electrical resonance between the capacitance of the bypass and the inductance of the big electrolytic. Uing high ESR caps will also damp the resonance.

BTW putting identical caps in parallel reduces their inductance, not increases it. It also reduces ESR, which may make resonance worse.

[Nounours18200]

Hi friends,

Sorry for my late reply, but I had to stay at the hospital for a serious surgery and I just feel good enough to post again...

Nounours,
Don't go higher than 10000uF on the caps directly connected to the rectifier bridge.
More will put too much stress on the transformer and rectifiers.

My current P.S. (version 1) uses 2x22.000µF, so I keep in mind to use 10.000µF max for the next version... (thanks for the advise)

Horowitz page 330

what does this mean ? does it refer to a dedicated post ?

If you want lower ripple add extra RC section(s) as in Wintermute's schematic.

I see, I will.

Let's back to the diagram suggested :


I understand what you explain but I do not have enough knowledge to properly calculate the resistors value nor the transformer output voltage (provided that the resistors will generate a voltage loss).

Is there a "common" value or a way to calculate them ? : I would avoid to buy parts (particularly the transformer...) and see that the value are not OK... In addition, it seems that you are not totally sure that adding the CRC circuitry would improve the sound ?

Thanks,
N.

[DF96]

Welcome back!

"Horowitz" may be a reference to the standard textbook by Horowitz and Hill.

For a CR section the DC voltage drop is given by Ohm's law V=IR.

The AC filtering is given by b=1/(2 pi f R C).

[wintermute]

Originally Posted by DF96

BTW putting identical caps in parallel reduces their inductance, not increases it. It also reduces ESR, which may make resonance worse.

Woops! for some reason I was thinking that inductors behaved like caps when paralleled

Nounours, do you know the actual current draw of your preamp, if you do (or can measure it) then just plug in to ohms law as given by DF96 to get the voltage drop across your chosen resistor value.

The resistor value I was using was 3.3 ohms. This is fine for my circuit which I don't expect to draw more than about 180mA.

Tony.

[QSerraTico_Tico]

Originally Posted by Nounours18200

Hi friends,

Sorry for my late reply, but I had to stay at the hospital for a serious surgery and I just feel good enough to post again...



My current P.S. (version 1) uses 2x22.000µF, so I keep in mind to use 10.000µF max for the next version... (thanks for the advise)



what does this mean ? does it refer to a dedicated post ?


I see, I will.

Let's back to the diagram suggested :


I understand what you explain but I do not have enough knowledge to properly calculate the resistors value nor the transformer output voltage (provided that the resistors will generate a voltage loss).

Is there a "common" value or a way to calculate them ? : I would avoid to buy parts (particularly the transformer...) and see that the value are not OK... In addition, it seems that you are not totally sure that adding the CRC circuitry would improve the sound ?

Thanks,
N.

Hi N, I wish you good recovery from the surgery.
The trick is I am using only CRC filtering in my power supply; no voltage regulators or shunt regulators. Many forum members will think I am crazy but my ears don't lie.....
As already explained you can calculate the voltage drop across the resistor by Ohm' s law, knowing the current consumption of your circuit.

[Nounours18200]

Welcome back!

Hi N, I wish you good recovery from the surgery.

Thanks and thanks ! I recover slowly but (very) surely !

do you know the actual current draw of your preamp, if you do (or can measure it) then just plug in to ohms law as given by DF96 to get the voltage drop across your chosen resistor value.

Actually I don't -it is a Threshold Fet-10HE preamp- but I could measure it (being cautious because I have so much upgraded it that it is as good as "formula one" preamp and a precious device for me...).

I will spend more time on this subject in a few weeks, because for the time being I have a Conrad-Johnson Premier Ten preamp to upgrade (I do not like tubes, and it is far from my preamp, but I will try to improve it...) and more important, I am learning how to pair Mosfets, using the Nelson article "Practical Mosfet testing for Audio"...
So you will probably soon see another post from me looking for some advises on this approach...

Thanks friends (I will go back to this Power Supply optimization),

N

[hesener]

Quote:
With respect to the 2nd diagram, By paralleling multiple smaller caps the ESR will be reduced, but at the same the inductance will be increased.

wintermute, I think the inductance should be reduced as well (at least the parasitic inductance of the caps), there may be additional inductance due to the wiring and PCB though.

paralleling caps is a good "trick" to do better wtih so-so caps, but you get to a point of diminishing return with really good caps, because of the stray resistance and inductance of the wiring...

just my two cents

[hesener]

sorry wintermute for repeating this, I really should have read the thread to the end....