I noticed awhile ago that my modular synthesizer needed more ways of adding together control voltages. It also had a shortage of adjustable voltage sources, which can be handy for varying several parameters at once. After some thought I came up with a design for a new module, which I call the CV Processor for lack of a better name.
The CV Processor has four reversing attenuators labeled A through D, each with its own input and output jacks and level knob. The knobs are center-zero; turn the knob clockwise and the gain of the attenuator goes from zero to +1X, turn it counter-clockwise and the gain goes for zero to -1X. Each input jack has a switching contact connected to +5V, so if you don’t plug anything into the jack the attenuator becomes a variable voltage source with a range of -5 to +5V. (The idea for adding a fixed voltage to the reversing attenuator came from a conversation with Robert Rich in which he mentioned that he gets a lot of use from a custom-built module consisting of three such circuits.)
There are also two adders in the module. If you don’t plug anything into output A’s jack, output B consists of the sum of the signals from the A and B attenuators. Outputs C and D behave correspondingly. Hence the module can be used to add together pairs of signals and/or to add a fixed offset voltage to a signal.
I also tossed in one instance of the Min/Max circuit from the Yusynth site. This unusual module has two inputs and two outputs. One of the outputs is the smaller of the two inputs; the other output is the larger of the two. Its obvious application is creating interesting control voltages from two varying inputs but it can also be used to process audio signals. I added it to the CV Processor by connecting the outputs of the A and B attenuators to its inputs.
Finally, borrowing an idea from Scott Juskiw, I squeezed the whole thing behind a 1U-wide panel by using small knobs arranged in an offset pattern. Here’s the completed module, sporting its panel from Front Panel Express:
I laid out a PC board, using mostly SMT components, and had it fabricated by BatchPCB:
In retrospect I would have done a couple of things differently. I would have used slightly larger holes for the wiring pads. The holes I used were for 0.1″ header pins and they were just small enough to be obstructive when I tried to thread stranded wire through them. I ended up using solid wire for most of the connections. That worked fine, of course, but solid wire is kind of pain for wiring panels because it’s less flexible. Also, I would have put more time into the silkscreen markings and added labels for all of the wiring pads. However, there were no mistakes on the board and it all worked the first time.
Here’s a view of the wiring:
I connected the wires to the back of the panel because the standoffs I had on hand were a little long and the PCB ended up a bit far to one side of the panel. (Yes, I could have fixed this by changing the brackets that hold the panel, but I didn’t feel like doing the drilling and cutting all over again.) I also flipped the power connector to the other side of the PCB since it protruded beyond the width of the panel. Try to avoid making this mistake; unsoldering a MTA-156 connector is no fun.
If anyone’s interested in the schematics, post a comment and I’ll try to overcome my laziness and clean them up. It’s all pretty much textbook op-amp applications, however.