FET switching

I have mentioned this type of switching system in other articles, and thought I’d use this space to dig a little deeper into how it works. After all, this is the system that are in most of our pedals. Once again, to refresh the memory, here’s how it works:

The signal enters the pedal through the input jack, and the first thing it sees is a buffer circuit. From there the signal is split between the actual fx circuit and another pcb trace that leads directly to the switching circuit. Unlike the ”real” switch you’ll find in a regular wah pedal, this ”switch” is actually two transistors that either mute the signal or pass it on. A momentary control pulse (from the switch you step on) will cause a pair of other transistors (called the ”flip-flop” pair) to change their status, which in turn opens and closes the audio paths through these two transistors – voila! We have switched from ”bypass” to active. The signal then passes through yet another buffer – to make sure that the output impedance is constant, regardless of switch status – before it is finally sent on its way, via the output jack.

Both the dry and fx lines, up to the mute transistors, are active at all times, which is easy to verify in a Boss or similar delay pedal. Just set the delay time fairly long, play a short chord and activate the pedal. You will now hear the delay repeats, just like it had been active all along. Some delays (the Boss DD-6, for instance) will not do this, and they will also let you hear the effect trail off even after you’ve bypassed the pedal. This can be easily done by simply placing the mute transistor for the fx line in front of the fx circuit, instead of after it.

Anyway, this type of switching arrangement serves several purposes: first off, it is a much cheaper way to build the thing (a few extra components and a momentary switch like the one sitting under your keyboard’s esc key costs a lot less than a DPDT or 3PDT switch that you also might need to hand-solder). Also, the actual switching can be made completely silent – without pops or crackles. Thirdly, since the signal will always pass through at least part of the circuit, it can be permanently transformed into low impedance, meaning that the tone-sucking is forever gone. If only… It is true that most Boss pedals doesn’t steal treble like a wah pedal does – but that doesn’t mean that they will never be tone-suckers. Remember the first rule of tone-sucking: an effect sucks tone if it makes your tone suck. This rule can apply to FET switching pedals as well, both directly and indirectly.

  • Direct tone-sucking: if you plug your guitar into a box, and the dry signal suddenly sounds weaker or just plain different (in a bad way).
  • Indirect tone-sucking: you plug your guitar into a box, and it doesn’t really change the dry tone that much. But as you add more boxes to the chain, the tone changes gradually (getting worse for each box you add).

In the first example there is a single pedal doing the tone-sucking, so the situation is easily identified and rectified; have it modified for true bypass or (when that is impractical for one reason or another) use a true bypass loop box. In the second example, though, the tone-sucking is a little more sneaky – there’s no single box that can be held responsible for the tone loss, and one by one they all seem fine. Still, there’s something missing when you have all of them connected… The solution, however, is the same; have them all true bypassed, or run them through true bypass loops. Here it may be a good idea to keep one of them stock (or outside any true bypass loops), to still provide buffering. The trick is to remove the unneeded buffers (in the pedals you true bypassed), while keeping one around to help your guitar pickups perform. Which brings us to the next chapter…

Enter the buffer…

To fully understand what a buffered pedal does, we need to understand how a buffer works. First off, the word ”buffer” actually means ”devices or pieces of material for reducing shock or damage due to contact”, according to Merriam-Webster’s excellent online dictionary. In short, it’s designed to sit in between (in this case) the guitar pickups and the effects circuit, to cushion the connection between them. It does this job by disconnecting the source (guitar pickups or effects pedal in front of the buffer) from the circuits that follow. Let’s look a little deeper into this, shall we?

Keep in mind that with a buffered pedal, regardless of switching status (”bypass” or active), your sound will have to pass through at least three transistors (one in the input buffer, the switching transistor and one in the output buffer). Now, if you look at the drawing to the right, you’ll see that a transistor is essentially a floodgate, which uses the original incoming signal to control the flow of electrons from its power source to its output. Essentially, it turns the power source flow into a copy of the original signal. If the power source is bigger than the original, the signal is amplified. Or rather – the bigger power source is turned into a larger copy of the original signal. In a buffer application (with equal gain in and out), the output signal is kept at the same level as the original signal. In either case, the original signal is lost forever – the transistor emits a more sturdy, low impedance copy of it, while it also disconnects the original signal’s source from whatever follows the transistor output. If those transistors are substandard (e.g. cheap), the resulting tone emitted might be less than desireable. Then imagine a pedalboard with five or six Boss pedals chained together – you’re looking at fifteen to eighteen transistors, just to get the guitar signal to pass straight through the pedals…

There are of course good buffers too, and in most cases, your garden variety Boss-type pedal will not mess too much with your tone. But you should at least be aware of the stacking factor – just like in the good old days of double-decker cassette players, when you could copy a copy of a copy etc, if you add up too many buffers in a row, the signal at the end of all those transistors will be quite a bit different from what you put into the first one. But no matter how good the buffer is, the low impedance signal it emits will change how it interacts with the next pedal (or the cable to the amp, if it’s the last thing in the chain). Most times, it will be a change for the better – for instance, it will drive circuits with low-ish input impedance much better, as well as long cables to the amp – but not always. If you put a germanium fuzz pedal (like a Fuzz Face or a Fulltone ’69) after a buffered pedal like the Tube Screamer, the fuzz will sound like… well, you don’t want to know. Decent language doesn’t cut it if you want to describe the sound, and *crap* isn’t strong enough. Anyway, a standard Dunlop wah pedal (manufactured 1992 or later) will do the same, and some buffered pedals will disturb fuzz pedals even if they’re placed after them in the chain…