MXR M-151 Doubleshot schematic tracing

I had the chance to get my hands on the MXR Doubleshot for a little while, and found out there are no available schematics anywhere. It's nice to do some selfless circuit tracing once in a while.

Starting from what was known already, the Doubleshot is usually described as the dual channel version of the Dime, and we'll see this is accurate. That circuit already has something interesting in its own right, but this one builds upon it with the switching to end up with a true monstrosity.

This wasn't an easy trace and it took quite some time. I can see why some people were scared away by the sheer number of components and the SMT board, but even pulling the board out of the enclosure was a challenge in itself: those familiar with EHX pedals will know what it takes to get out a board when the DC jack is stuck in its hole.

With that done, I could start poking around. Knowing about the similarity with the Dime, the schematic on FSB was a great help in giving me a hint as to what to look for to start drawing the backbone of the circuit.

It was a good thing I had a reason to get the board out, because as soon as I did, I found out the two 10Ω resistors on the supply were both burnt. The (second hand) pedal was somehow still working, but it sounded awful, with half functioning EQ and oscillation at certain settings. Replacing the resistors fixed it completely; it's still not my favorite distortion, but it sounds ok and the EQ is very useful.

So here's the result!

Let's get some of things out of the way:

1. Since this is a traced schematic, I cannot guarantee that it's fully accurate. Still, having the Dime as reference, I'm pretty confident in what's involved in the signal path at least;

2. The schematic might not be complete. Again, I feel good about the signal path, which is the main interest, but the switching scheme was difficult to trace because of traces and joints hidden under the audio jacks. Hopefully, the flip flops and associated logic represent a working circuit, or at least represent something useful both for repairs and for study (I know it does for me).

3. Resistor values are easy to read in SMD, but I wasn't going to desolder each ceramic and film capacitor to get a meaningful measurement. The values in [i]italics[/i] are taken from the Dime schematic, but since the rest (topology, resistor values, up to the arrangement of the op amp sections) is exactly the same, they're pretty good guesses.

4. IC and electrolytic capacitor numbers reflect the silkscreen annotations, the rest doesn't.

5. Since there are many components and it's hard to keep track of all of them, the method I came up with is that all semiconductor components (IC sections, transistors, diodes) are accounted for, and also electrolytic capacitors, switches and pots. The rest follows those.

CMOS, CMOS, on the wall

The first observation about the circuit is that the designer really liked their analog switches. We find 24 CD4016 sections in the whole schematic, plus three JFETs. It's clear that the Dime was the basis for this, and switches were added to swap in a different set of pots, a different gain control, a larger input capacitor, and a different "Focus" mode. Channel 1 (the top one) is identical to the Dime, while channel 2 is more akin to a Dist+ with an overcomplicated EQ.

Speaking of which...

The EQ

The equalizer is another case of obsessive fascination. This is shared with the Dime, three repeated stages of the same topology, which should look more clearly for what it is as I've redrawn it. This is a version of the biquad/SVF using the non-inverting input called KHN biquad. In all cases, the bandpass output is taken, attenuverted, and mixed with the input signal to result in an adjustable band boost/cut. I can't say this is the most efficient way to achieve this result, but they liked it so much they've used it for each of the bands.

The bass control, although being a 100Hz peak in theory, it's so wide with the 3.3k disconnected to practically be a shelving control. In both cases the range seems to be from -3dB cut to 27dB of boost.

The mid control is a band centered around 350Hz, with a wide boost up to 20dB and a much narrower cut up to -30dB ("boost wide, cut narrow"). "Focus 1" lifts this control from virtual ground, leaving you with all the scoop all the time.

The treble control is a 3.3k band with about 16dB of boost and -6dB of cut available.

This is clearly the reason why the pedal requires 18V and provides its own adapter. While it works fine at 9V, the amount of boost available before clipping isn't as large then.

The elephant in the room

That leaves us to talk about the switching. It all revolves around two dual CD4013 flip-flops, with inverted outputs to alternate the analog switches active, a separate delayed output that controls a pair of muting JFETs at the output to prevent any noise while switching channels and logic to have the channel settings have priority over the "Focus". A very proper pair of switches keeps the pedal quiet while bypassed: a 4016 removing any gain from the clipping stage and a JFET muting its output. We even have a nice constant-current arrangement for the channel and focus LEDs, with the side effect that brightness changes based on how many LEDs are active at once.

Considering all that, I'm surprised that the pedal isn't true bypass in the way a Crybaby Wah isn't: the circuit input is connected at all times. In practice, the 913k input load might not make a difference, but I'm still left wondering if there couldn't be another way, even with the DPDT switches being used (Millennium comes to mind).

Conclusions

While this is an interesting circuit to look at, I'm not a big fan of the maximalist approach and while playing it, the complexity of the circuit isn't reflected in what you hear, which is somewhat unimpressive. Other than that, the input buffer is not something I'd like to have in a high-gain circuit, even if it's somewhat justified by the input capacitor switch, unlike in the Dime; the lack of true bypass is probably unnoticeable in practice but still not ideal (but now that you know about it, you can change it!).

That's it for now, I hope you enjoy it, find it useful and tell me your thoughts.