Silver Supporting Member
Great explanation. Must be hard getting a line of pedals sounding the same!Exactly, and it might vary unit to unit as much as between manufacturer to manufacturer.
Consider this: a builder orders three hundred Epcos 1nF capacitors with +/-5% tolerance. Most of these caps were probably manufactured by Epcos as part of the same run. We expect them to actually be anywhere between 95-105nF. Measuring them, he finds that maybe half of them come in at about about 98nF. Maybe a quarter are slightly more, say 99.5nF. The other quarter are slightly less, 96nF. A few months later he runs out and places a new order of Epcos caps. These were made in the same factory, but likely came from a later production run. These might hover around 101nF, with some being a bit lower in value and a few being a bit higher. You'll get just as much variance between brands, though you might find that some brands might hover around different actual values.
Now here's where it gets really frustrating- every single component is like this. Resistors tend to be 1% or 5%. Caps are usually 5% or 10%. Pots are 20%. ICs and diodes also have variation in real-world specs.
What can the builder do? He can measure every single component. But guess what? Measurement tools also have tolerance for drift. One meter might be off from another. How do I know that my 100k resistor is actually 100k or 95k when my meter has a 5% tolerance?
Most builders won't measure the components before soldering them in. Not only does it triple the amount of time it takes to populate a PCB, the measurements aren't even guaranteed to be accurate. And what do you do with all those parts that are only slightly out of spec? The builder paid for them, and they're in spec as far as the manufacturer is concerned. Not using them means you'd be buying two or three times as many components as you're using.
Another harsh reality: most builders don't have capacitance meters. So they can't even measure their caps.
What all this means is that we accept a certain amount of variation in character from pedal to pedal.
For Klons specifically, the diodes are also an issue. Most kloners don't even own a Centaur. They just use a PCB layout based on the 2008 soulsonic schematic. They can build the circuit easily enough without ever even being in the same room as the real deal, but they have nothing to reference for getting the sound of the diodes right unless they're willing to invest $1200-$1500 in a Centaur just for that purpose. And the higher quality klones tend to use higher quality diodes, which may be more consistent. Since a good in spec 1N34A doesn't necessarily clip the same way Bill's diodes do, they might be more consistently un-Klonlike in their character.
Then there's the PCB layout. Everything, and I mean EVERYTHING, has electrical properties. Just having component traces near each other on the PCB can result in a little bit of capacitance between those nodes of the circuit. In theory, the capacitance is so small that it should be inconsequential. In practice, it seems to matter sometimes. I've A/B'd some of my old builds using DIY PCB layouts against a much more acurate Centaur layout, and they've sounded pretty much the same even with component tolerance taken into consideration. On the other hand, some DIY Rat layouts I've used seem to come out a bit different from others even with the same components and a lot of measuring on my part to try to keep everything consistent.
I still haven't decided how much the resistance on either gang of the dual ganged pot matters. My Centaur actually seems to be a bit off spec, with one gang measuring as low as almost 80k. But it sounds the same as klons I've built with matched gangs. I've done builds with separated gangs (two pots), and I'm starting to think it matters less than I previously thought after play with both sides of the pot, but at the same time, some seem to have slightly more gain on tap and that might have something to do with having more resistance on the gain gang.
The important thing is that, in the end, a good klone is the Klon circuit without the magic diodes. And always be aware that you're going to hear differences that aren't there because when things look different we expect them to sound different.