If you have a Boss SD-1, you may or may not have noticed that it has a “bleed” issue. Or you might have read on various forums that on the SD-1, the distortion can be heard ever so slightly bleeding into the clean bypass signal.
First, let’s be clear – this is indeed a real phenomenon. It may not be a problem for you, or you might not have noticed it, but rest assured – it is there. If you normally run your amp even slightly distorted, chances are you haven’t come across this, as the amp’s distortion will mask the bleeding effect. But with a clean amp, turn the drive/gain knob up and listen to the faint “buzz” underneath the notes you play. The bleed is more prominent on the lower notes, as well as the more drive/gain you apply. Meanwhile, the volume and tone controls don’t seem to change things (which in itself is a clue to what’s going on, as we will find out).
Important note: This article deals with the “original” version of the Boss SD-1 – the Waza SD-1w and the new version Boss SD-1 (read more about that one here) is not covered here. If you’re in doubt, take the back plate off and verify that yours match the image to the right.
Cause and effect
This topic has generated quite a lot of discussion on various internet forums, and over time various theories have been investigated. Early on, it was speculated on whether leaky mute transistors could be the cause (note: see the article on FET switching if you don’t know how the Boss bypass circuit works). This was fairly quickly ruled out – not that it couldn’t happen that something could cause the mute transistor to not close fully, but it isn’t the root cause here. If it were, the volume and tone controls would have an effect on the bleeding. Since they don’t, the bleed must occur somewhere before those controls in the circuit.
Another theory was that the input resistor – which is located right next to the clipping diodes – is picking up a little of the distorted signal. A fix was proposed, simply relocating the input wire and resistor. The idea was to lift the north/top end of the input resistor and soldering the input wire straight to that end of the resistor. That does indeed move the input resistor away from the clipping diodes, and should therefore clear up the bleeding. As you’ve probably already guessed, it didn’t… 😉
So what is the actual problem?
The actual cause is related to the input resistor thing, and that resistor (R1) can absolutely benefit from being moved. But the problem goes deeper than just a single misplaced component. In fact, the entire clipping circuit – not just individual components, but also its circuit board traces – is simply placed too close to the main input circuit. Thus, this issue is caused by a mistake made when the SD-1 was first designed. Let’s take a look, shall we?
In the picture, the input circuit (main input -> input buffer stage) is marked in green. The signal enters the circuit board at wire connection point 3, goes through the input resistor (R1) and from there straight to the input capacitor (C1). From there, it splits off to R2 (which references the signal to ground) and to the transistor in the input buffer stage (Q5).
The clipping circuit, conversely, is marked in red. As you can see, not only does the input resistor live right next to the clipping diodes (D4, D5, D6), the circuit board traces for the clipping circuit also pass right under and around the input circuit components at a couple of points. The two parts of the circuit are so close that you can actually get the pedal to squeal with feedback, by simply touching green and red solder joints with the same finger…
So can it be fixed?
Well, yes and no. To fix this problem at its root cause, you’d need to completely redesign the circuit board, to ensure enough physical distance between the input and clipping sections. I haven’t looked into either the SD-1w (Waza Craft version) or the new Boss SD-1 model (mentioned earlier), but both have radically redesigned circuit boards, so it stands to reason that Boss would have made sure to not repeat this mistake.
That’s not to say that it is a lost cause. There are things that can be done to remedy the situation. First off, if you’re playing with an amp that’s already distorting – for instance if you’re using the SD-1 into a Marshall, Zakk Wylde-style – you need not bother with this at all. Also, if you’re running the pedal with the drive/gain pot set fairly low, you probably won’t have any problems either.
The best fixes I’ve seen include muting the signal going into the distortion circuit – that of course doesn’t solve the underlying circuit layout issue, but it does eliminate the symptom/problem. The DS-1 actually uses this trick – another mute transistor, wired to mute the distortion circuit input at the same time as the one on the distortion circuit output – to cut down on crosstalk. So that can be employed here. Here’s a link to an explanation, if you want to read more about that: http://dantonewac.blogspot.com/2017/01/boss-sd-1-bypass-bleed-noise-fixmod-and.html
I have also seen another variant, where the additional JFET is instead used to turn the gain/drive down in bypass mode. Personally, I prefer muting the input signal to the distortion circuit, as described above. To me, it makes more sense to completely mute the distortion circuit. Also, the JFET/capacitor combination can easily fit in the space where C2 was, for a quite neat installation.