10/25/20 A Boss DF-2 pedal has been brought to me to do some troubleshooting on an LED indicator that remains on. I have not done any repairs in a long time, and sometimes it’s good to get under the hood and see if you get any design inspiration. I thought this would be a simple fix, after playing the pedal, I realized this was not a simple on-off switch. This was an on-off, but if the button was long-pressed it would sustain the most recent note played and oscillate. So I had to open it up.
- I always do a visual inspection, in this case everything checked out.
- The switch itself was working as well.
- With the voltage off I decided to trace the path from the LED using a voltmeter in continuity mode, but this was a little too tedious, so I looked up a schematic. Found here https://www.synthxl.com/boss-df-2/
- Again starting with the LED and working my way through the circuit on the schematic…. I became mesmerized….
In the midst of dissecting this design I found myself drifting into full nerd mode. I have never seen a pedal design like this, this pedals has everything enclosed in the chassis… LFO, JFET switches, BJT switches, buffers, gain stage, tone control, frequency detector, and logic gates. But the most impressive thing for me was the switching circuit. The switching circuit was a full mix of analog transistor switches, with digital logic gates, fed back and twisted while all worked in harmony. Just check it out!
The easy way of solving this issue would be to probe pins and see which ones were switching high/low starting from the LED and working backwards. But I wanted to work my way through the switch logic maze the hard way. This took around 6 hours of review and state tracing. To be honest, I stopped a little short of full end to end understanding of the switching due to the feedback pins, but was happy to call it complete with where I landed knowing I could finish with another 6 hours or so.
11/1/20 After watching a video of the pedal this confirmed my initial gut feeling on the failing transmission gate . The give away was the press and hold feature that made the LED oscillate which was failing in this pedal as well as the on/off. No probing was needed with this evidence. I know I needed to replace the HD14066BP, which i used equivalent NTE4066B.
11/11/20 Take this IC out, whoops not that one. I accidentally removed the HD14046BP, and this super sucks, anyways the replacement IC for this was the CD4046BE. I was able to salvage the IC, like an animal and still can’t believe this worked(see photo). After the salvage I replaced the HD14066BP transmission gate IC with the NTE4066B, which to my disbelief didn’t fix the LED issue. This left me flabbergasted as fuck. The issue wasn’t the transmission gate? Now I really have to probe the circuit, which I initially avoided due to my confidence. I thought this fix was complete 10 days ago. Well time to probe and really need to learn the circuits logic. Here are the states the pedal is seeing. Foot switch [press, release, hold]. With quick probing I found a hack for this that I know will work, but I still want to know why this is isn’t working.
11/17/20 Looking at the circuit and stepping back for almost a week… maybe this is just the switching transistor. I probed and in fact it was not switching. Simple replacement with a generic BJT and just like that it worked. Everything end to end works. So I buttoned everything up and then the feedbacker was no longer working. I went through and re-tested everything that I replaced and updated my animal hack on the 4066, with no luck. I started probing some more around Flip flop leading to the PLL. And kind of reach a dead end of frustration. I decided to buy an operational version of this pedal and do some side by side probing.
12/15/20 Now that I have opened up both of these pedals, the new pedal stopped working after I opened it up. Now that I had two broken pedals I decided to build the feedbacker myself. This went quite well, but over the couple months of putting it together I never even tested it, its still on a bread board… I kind of lost motivation. This took around 2-3 months until the fire on the candle went dim.
9/5/21 Along with messing with the bread board here and there, I would start working on this, then get distracted and/or too busy working on projects that were less painful. With a clear head I came back, and once I figured out that it was just a grounding issue, plus mixed up input/output, along with broken PCB traces…(3 hours), the issue was clear to me. The weird thing about the grounding issue was that the first pedal did not need to have the pots grounded and the new one that I bought did. Once I got the new DF-2 operational, I was able to do a side by side. From there, I focused on the Low Frequency Oscillator. I noticed that the LFO on the old pedal wasn’t working and assumed it was an op-amp failure. I piggy backed an Op-amp and isolated the pins from the old op-amp, resulting in getting one of the other channels working. But this didn’t fix it. On the channel that was working, I couldn’t find any issue there, is just passive components that all seemed to be working. So I gave up again. Only this time I decided to build my own LFO to replace it.
