The Rotund Robot (2) – analysing schematics

This was originally posted 21st March 2017 as a page. I’ve rejigged the website, so I’ve copied and pasted it as a blog post (1st Jan 2018)

Initial searches yielded these two schematics.

BOSS OC2 schematic - freeinfosocietyBOSS OC2 schematic - Sabro

The first is from Stellan Lehrberg, and I believe it was one of the first BOSS OC2 schematics to enter the DIY community. The second came from Harald Sabro’s site.

After staring at these two schematics for days on end, sometimes not blinking in a fury of brain farts and attempts to compute the lines and symbols for hours at a time, I started to dissect the circuit, but not before panicking about the redundant component numbers I’d taken in on the original Lehrberg schematic. If you have a look at the second schematic, you’ll see that most of the components from the original have modern day workalikes listed instead. The only one I wasn’t familiar with in the second schematic was the 2SK30 (Q7,Q8). Having read a lot of the notes on, I saw that these can be replaced with 2N5457 JFETs. If a 2N5457 will do, then a J201 will probably be equally effective (something to bear in mind if you don’t have any 2N5457s on hand). Also just to note, the 4013 and 4027 in both schematics often have the prefix CD. They are CMOS chips, and on first glance it looks like they are responsible for all the fun stuff going on.

If we think of schematics as a map, and then we think about the fact that we have both DC powering the circuit, and AC in the form of the signal path, we have a method of looking at the schematic in a simpler way. The first thing I did was was print out both schematics and get a red pen and follow the power on both schematics like this:

BOSS OC2 schematic - Sabro with power

You’ve probably already noticed that the two schematics look different, even though they claim to be the same circuit. This is why it’s important to go through them piece by piece to check for any differences. I’ll primarily refer to the Sabro schematic, though I think it’s important to be familiar with both.

I’ll point out now, having been over both schematics a lot, there are a few parts that aren’t labelled particularly well, and might cause some confusion, have a look at the labelling for the ICs in the Lehrberg schematic for example. The next step was to divide the schematic into sections. Then I took on the task of drawing out each section of the schematic and comparing component values for each part. There are only a couple of small different values between the two schematics. You’ll also probably notice that the input transistor has been removed on the second schematic. The output tranny is also missing a 1u cap and a 1M resistor (from 4V!? which is most likely meant to be from the split supply, though the ratio of the resistors in the voltage divider would suggest more than 4V from a 9V supply).

Hopefully, you’ve already checked out this video from noisedom I mentioned in part 1.

He identifies two parts in the Lehrberg schematic as being the envelope follower (this part helps the octave follow the volume of the dry signal) and a filter section. It’s great that he identifies these parts for us, it gives us somewhere to start from. The only thing is I think he might be wrong. He modded an existing pedal, and my plan here is to get the original schematic onto a breadboard to play around with, hopefully tweak, with the aim of learning, and then doing a limited run of PCBs for a synthy octave effect with whatever mods I come up with. I have two reasons for thinking he’s wrong. Firstly I think the section in red is the envelope follower, whereas he identified the areas in blue as being the envelope follower (see below). Remember he used the Lehrberg schematic instead, so it looks different. But we can easily identify the area he was talking about in the Sabro schematic, because of the JFETs.

BOSS OC2 schematic - Sabro EF

So why do I think he’s wrong about the envelope follower? Firstly IC2 is a LM324, these aren’t particularly good quality op amps. They are very cheap and work well, but generally you don’t use them for anything you want your guitar signal going through, the TL074 would be a much better choice for a quad opamp. There’s even one used in this circuit. The LM324 is perfectly fine however for other things that don’t involve your signal directly, like an envelope follower! Also, if you look at the diodes in the wee red box, that’s another clue. Those diodes are only letting half of the AC signal through, which is exactly what you want in an envelope follower! This bit of the Art of Electronics seems to back me up too.

aoe ef

Also, I’ve heard that the tracking on such CMOS octave effects are truer when you use the neck pick up, and roll a little top end off. It would also then make sense that the filter section would be low pass filters (series resistor, followed by a capacitor to ground), and that these would be before the CMOS ICs.

I’ll be perfectly happy to admit if I’m wrong, I have been before. We’ll push on with this schematic in part 3.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.