ZW-44 cap replacements

[g-sus]

Just finished building my ZW-44 using torchys vero layout, works great!

Link to schematic: http://server3.uploadit.org/files/Torchy-Wylde_Schematic.jpg

I had to replace the C1 3.3uF cap with a 4.7uF cap and C6 18nF cap with a 15nF cap because I didn't find ones with right values..
Sounds great with these but would it sound better with original values? Could somebody explain how higher/lower valued capacitors in these C1 and C6 positions change the sound? Thanks in advance, this is a great forum! 

[Mark Hammer]

Those caps are intended to block any DC riding along with the AC signal.  At the same time, they also roll off bass content below some frequency (think of DC as a really, really, REALLY, low frequency).  The rule of thumb is that the rolloff is proportional to the capacitor value. As the capacitor value doubles, the rolloff point goes lower by an octave.  This also means that as the value of the cap is cut in half, the rolloff goes *up* by an octave.

That can be important, OR moot, depending on where you start out.  If, as an example, a cap value sets the frequency where a rolloff begins at 80hz, then substituting a .47uf cap you don't have, with one that is .22uf, will raise that rolloff by a factor of 2.14, or a little more than an octave ( = roughly 171hz), which will have a modest impact on guitar and a deadly influence on bass or 7-string guitar.  If it happened that the starting point was a rolloff of 16hz, then raising it by an octave or even two would do little harm to a guitar signal, and only modest harm to a bass signal.  Similarly, dropping it by an octave (to 8hz) or ven two (to 4hz) would cause no audible change in bass content beyond what you started out with.

Keep in mind that the "rolloff" frequency is defined as the point where the signal is 3db down, which implies that frequencies a little higher up (e.g., 20hz, when the nominal rolloff is 16hz) are also affected, though not so much.  Cascading many such rolloffs that seem to be safe *can* result in a cumulative rolloff that sucks a bit of tone.  So, stick ten 6db/oct rolloffs at 50hz in series, and eventually you're going to lose a couple of db at 80hz.  Happily, most pedals will aim for very wide bandwidth in anticipation of such cumulative rolloff, giving may of us a lot of freedom in subbing parts.

Subbing 4u7 for 3u3 in this circuit will move the rolloff down a bit, not likely influencing the tone at all.  Subbing 15nf for 18nf will raise the rolloff (i.e., chop a bit of bass), but only by a factor of about 20%.  Given that many caps have a 10% tolerance anyways, this change should not be particularly audible.

The take-home message here is that:
1) feel free to make substitutes like you did, the pedal will still work and likely provide the sound it is intended to provide
2) substitutes are fine as long as the original settings/rolloffs were not in any "critical zone" and the subs were not way off
3) you need to keep an eye on the total number of substitutes and rolloffs in a signal path, both within a pedal, and across pedals.

This is part of my own hesitancy about always-on buffers in the input and output of pedals.  One at a time, they rarely present a problem and even improve matters.  Ten at a time, though, and they can create problems.  Similarly, feel free to sub smaller and larger values of caps in the signal path but just keep in mind how many such subs you've made and how much potential tone change they can create in total.

[g-sus]

Greetings Mark and thanks for shedding some light on this.

This was my second pedal project (first was Foxx Tone Machine) and first time I had to make cap substitutes. I also had to make a 1.8M resistor using 3 smaller ones in series 

Maybe I'll get the original parts someday but right now I'm happy the way it sounds.

Cheers from Finland!