A question I "should already know the answer to", but I don't

[Chris60601]

I have read the Wampler book, Advanced DIY Effects Pedal - in there he states that he likes using this config for power filtering (See image below)

So my question is simply this ... Would this be a standard config to use or would the values of the caps vary?
If they vary, what might some common values be and why


Cheers
Chris

[Muadzin]

I've seen other values being used, 1N5817 for the diode, 100uf instead of 47uf, 47nf instead of 100nf. But the basic schematic seems to remain the same.

I find it odd though that if Wampler likes this config for power filtering, why is it not included on almost all the vero layouts for Wampler pedals? Is it on some separate pcb so that its not included on the traced schematics?

[Chris60601]

As he (Wampler) states in his book, many folks don't include the polarity protection and filtering in schematics as its assumed that its a known inclusion in circuits.

[ξεναγος νεκροπολης]

yeap...
if you look at the main site, almost all wampler veros, have a small resistor in series with 9v (no diode), and an electrolytic from 100uf to 470uf. they don't have the nf cap.

but after all i think that Mark and Miro use their own style of filtering in their layouts. i personalty prefer the one with the small resistor in series, and both uf & nf caps.

[Chris60601]

I get the diode (for polarity protection) but I dont understand the usage of the resistor.

[induction]

Chris60601 wrote

I get the diode (for polarity protection) but I dont understand the usage of the resistor.

It's easy to think of the cap to ground as a 'filter cap' and believe that it shunts any noise on the power line to ground, but it's not quite true. In fact, the cap is the bottom half of a low-pass filter that uses the power supply as the input signal. The corner frequency of this filter is 1/(2*pi*RC).

Without any resistance, R becomes zero, and the corner frequency of the filter becomes infinite, ie. the filter does nothing. In practice, without the resistor, the low-pass filter relies on the resistance of the copper, the output impedance of the power supply, and/or the ESR of the cap to fill in the R in the formula. If all of these are small, then the filter won't work so well because the corner frequency will be too high.

Adding a small value resistor in series between the power supply and the rail used by the circuit gives you a predictable resistance for the power filter to work. Upside: the larger the resistor, the lower the corner frequency for a given cap value. Downside: the larger the resistor, the lower the supply voltage to the circuit. It's a balancing act, choose a resistor large enough that you can get decent filtering with a reasonably small cap, and small enough that you don't drop your supply voltage too much. The supply voltage requirements of all circuits are not the same, neither are the supply filtering needs of different circuits and supplies. In some circuits you can get away with 100R, in others even 10R is too much. This is why there is no single, correct supply filter that works on all circuits. They have to be chosen to fit the circuit in question.

The second small cap is added because many electrolytic caps have a capacitance that is dependent on frequency. So a large electro might be chosen to pass all frequencies above some threshold, but in practice it inhibits higher frequencies more than you'd expect, which means the very high frequencies won't be filtered out so well and the circuit may be susceptible to RF oscillation, which you can't hear directly, but which does often have a very audible effect on the signal. So you use a smaller value cap in parallel to help the filter at the frequencies where the electro is not as effective. This works best if the smaller cap is film or ceramic, because that takes advantage of the different frequency-dependent capacitance of the different materials. If you use a small electro in parallel with a big one, you are just effectively increasing the value of the bigger cap.

[Frank_NH]

Here's a nice  summary of the power supply options, courtesy of the old Beavis Audio website.

The one thing I would add to Induction's excellent summary above is that many older schematics (e.g. fuzz faces) don't have any power filtering at all because they were designed to work with a battery only.  Power filtering is much more of an issue with modern effects which need more current and hence rely on "wall warts" and similar power supplies.

Also note that many power supplies for effects are rated for 9.6V.  This is to accommodate the voltage drop due to an inline resistor or diode.

[Chris60601]

FANTASTIC!

Thank you both Induction and Frank. I should have looked at Beavis to begin with *Head - Desk*
I needed the visual for the tiny LED to go off in my head

[IvIark]

Virtually all the layouts will include a filter cap but usually just the 47u and not the 100n in parallel, although that is included in some where I had the space.  I will always omit the parallel reverse polarity protection because I think it's stupid.  In the event of a reverse polarity being applied the diode will at best burn up, at worst explode.  I've seen big holes in boards cause by protection diodes used in this configuration.  

So if I were to use anything it would be series protection using a 1N5817 for a lower voltage drop.  Under reverse polarity the diode simply won't conduct, so it doesn't have to be a diode of mass destruction.

[Chris60601]

Yes - I read exactly the thing you mentioned. Using something other than the 400x and for exactly the reason you stated (lower voltage drop).

Thanks everyone!
If I did, I didn't mean to pose a question that may be a no-brainier.
I just needed a good understanding.

[Suzukiscottie]

What a brilliant summation of power filtering!