This modest signal combiner can be an important sound-shaping tool

When you’re configuring your modular synthesizer, it’s easy to leave out the all-important utility modules that will help glue together a patch as well as open up creative options. In this note, we’re going to talk about what I call “utility mixers” – not the final mixer with optional effects sends and the such, but tools to combine signals in the middle of a patch.

The first two sections below (Why Do I Need More Mixers? and Audio versus Control Voltage Mixers) are aimed at those new to modular synthesis who are still gaining familiarity with these building blocks. The third section (Breaking the Rules) may be interesting to even experienced users as it presents a few alternate ideas for processing signals in your system to get new sounds out of your existing oscillators, waveshapers, and filters. At the end I’ve compiled a collection of flexible mixers you might want to check out.

Why Do I Need More Mixers?

Roland 530 dual VCAYou often need to feed multiple signals into the same input – for example, so that more than one oscillator can be fed into the same filter, more than one filter can be fed into an amplifier, or multiple modulation sources can add movement to the same parameter such as pulse width, oscillator pitch, stereo position, or filter cutoff.

If you’re new to Eurorack modules, you’ll be surprised how many of them don’t have mixers built in – sometimes to save space or cost; sometimes to make room for unique functions. Many manufacturers assume you already have utility mixers elsewhere in your system, and don’t want to duplicate the space or expense required to incorporate them into each module. This tends to be less of an issue on the control voltage side, where many modules have multiple inputs, but it’s rare to find multiple audio inputs. (A notable exception is the  Roland System-500 modules, several of which have mixers on the audio and control voltage inputs. That’s one of the reasons I chose them as the core of my “training rack” for my upcoming Introduction to Modular Synthesis course.)

BlackStarSplitter_100hCan’t you just use a multiple or a stacking cable to mix together signals? Don’t do it. Stacking cables are indeed useful for splitting many signals (a subject I’ll cover in a separate note), but I suggest you don’t combine signals with anything other than a mixer: You may end up damaging a module as two outputs attempt to drive each other in addition to the input they’re connected to. (Yes, some modules have protected outputs that allow you to do this…but it’s better to be safe than sorry, and develop good habits just in case.)

So: In addition to a final output mixer, you will quickly find you have need for one or several small, simple utility mixers scattered throughout your system.

Audio versus Control Voltage Mixers

The voltage ranges used by both audio and control signals in a Eurorack modular tend to be similar, so it’s not immediately obvious why there would be a difference between a mixer for audio and a mixer for control. In many cases, they’re indeed interchangeable (and by the end of this note, you will be interchanging them to get some outside-the-box results). But – keeping in mind that in modular synthesizers, rules are designed to be broken – here are the basic differences:

Audio signals tend to alternate between positive and negative voltages. Often you will want them to be balanced evenly around a zero voltage reference rather than offset positive or negative; otherwise, as you mute them or process them through an amplifier with a very fast envelope, you may hear an added click or thump as that offset voltage is suddenly returned to zero and vice versa. Therefore, some audio mixers are “AC coupled” (AC = alternating current) to remove any voltage offset from their alternating nature; a few even have jumpers you can add or remove depending on whether or not you want this behavior.

STG .mix & MA CP3Audio mixers often have simple level controls that can attenuate the signal if desired. This is important because as you combine multiple signals, the result can add up to a higher voltage swing than the next module in line can take resulting in clipping distortion, which may nor may not be desired. Their attenuators tend to have a logarithmic curve, which may be more intuitive when mixing power-based sound levels rather than just voltages. While some audio mixers pride themselves on how clean and accurate they are, some purposely distort the signal– especially at louder levels – to add some interesting color to the sound. The STG Soundlabs .mix and Manhattan Analog CP3 – both based on the old Moog CP3 mixer design – are particularly known for this; STG has a detailed description of this behavior on their web site.

Don’t think of modular synths as hi-fi stereo or lab equipment that always needs to be clinically clean and precise! Philosophically, they’re closer to guitars: sometimes you do want to get a clean, bright sound (steel-string acoustic); sometimes you want a woolly, mellow sound (nylon-string acoustic); sometimes you want to distort them like crazy and knock down some walls (fuzzed solid-body electric).

Control signals can either alternate like an audio signal, or range from zero volts to typically a positive value. They can even be steady, unchanging voltages other than zero. For that reason, you do not want a control voltage mixer removing any offset from the signal. Therefore, control voltage mixers need to be “DC coupled” (DC = Direct Current).

RYO VertMix & Doepfer A-138cControl voltage mixers often have more complex level controls. Not only do you want them to attenuate to the incoming signal, you may want them to invert the signal so their voltages swing in the opposite direction. You will often find these level controls called “attenuverters” indicating they can both attenuate and invert. Some control voltage mixers can also amplify a signal. These level controls tend to have a linear response, which make sense when mixing voltage, but which might not have as natural of a feel when mixing audio. Another common feature is the ability to offset the voltage by a specific amount in one direction or the other. Combined, this is how a control voltage mixer can, for example, convert the 0 to +5 volt output of a particular LFO to, say, a -1 to +1 volt range to create filter sweeps centered around the filter’s set cutoff frequency. Most control voltage mixers pride themselves on their accuracy, making sure not to distort the control signal you are trying to carefully attenuate, invert, and offset.

Breaking the Rules

Of course, you do not need to conform to the straitjacket of “this mixer is for audio; that one is for control voltages” (keeping in mind the difference between linear and logarithmic controls mentioned above). You can get a lot of interesting results by mixing and matching their characteristics. For that reason, not only should you consider having more than one utility mixer in your system, you might also think about having more than one model, each with different characteristics. Here are a few ideas of what to do with them:

SSF MIXMODE with Expander•     You can overdrive a DC-coupled “audio” mixer with a useful distortion or saturation response to change the shape of an envelope or LFO (low frequency oscillator) control voltage passing through it. Again, the DC-coupled STG Soundlabs .mix and Manhattan Analog CP3 “audio” mixers both fit this bill, especially as they distort positive and negative voltages differently; the Steady State Fate MIXMODE with Expander also looks interesting as it combines an output saturation circuit with inversion possibilities; its Expander gives more control over the offset and clip points.

Invy_230h•     A mixer that has an inverting input is useful for combining multiple waveforms from the same oscillator, or multiple outputs from the same filter. In the oscillator case, inverting one of the waveforms will subtract rather than add its harmonics, creating a new sound that will have some harmonics (such as the fundamental) cancelled out or reinforced. For those coming from a guitar background, this is akin to flipping the phase of one of your pickups. In the filter case, you can create new filter response shapes by mixing, say, a bandpass output “out of phase” (inverted) with the lowpass output from the same filter. It can also be useful for filters that give you access to different intermediate stages, such as the STG Soundlabs Mankato filter. You can also apply this idea to other modified copies of the original signal, such as inverting the output of a short delay and mixing it with the original to create comb filter effects.

WMD Utility module•     A mixer (or other control voltage processor) that allows you to add an offset voltage opens the possibility of distorting just one half of the waveform as you carefully overdrive the input of a waveshaper, filter, or distorting mixer. This is technically known as asymmetrical distortion, as opposed to symmetrical distortion where you distort both the positive and negative halves of the waveform equally. I surveyed a number of module manufacturers, and many of their filters as well as most of their mixers and waveshapers will react in interesting ways when fed a signal with a DC offset.

Michael Johnsen, a designer at Pittsburgh Modular, explains why this asymmetric distortion can be desirable: “Asymmetrical clipping is the way to generate even harmonics. Depending on the degree of asymmetry, odd-order harmonics are also generated, but their amplitude gets lower as the asymmetry becomes more pronounced (in other words, as you offset more strongly in one direction – Chris). Symmetrical clipping adds odd harmonics only. Strong even harmonic content is characteristic of tube amps and the such; asymmetrical waveforms are also characteristic of many natural sounds, like the voice. Some waveshapers use DC offset with this in mind, and many stomp boxes and solid-state amps use offset to emulate tube behavior. It’s often used in combination with soft clipping, which is something we’ve designed into many of our modules.”

Make Noise MMGWhat does this sound like? Walker Farrell of Make Noise has created a pair of videos using the separate DC input on their very flexible MMG (Multimode Gate) to create asymmetrical clipping for the signal fed into the MMG’s AC (audio) input:

•     MMG Overload and Resonance

•     Make Noise MMG Demo 2 (between 1:30 and 3:12)

Additionally, the AJH Synth Sonic XV mixer includes a bias voltage offset into the input of a built-in wavefolder that creates really nice variations on its sound. I demonstrated this as part of a video I created on their Sonic XV and Ring SM. Try this trick yourself with your own modules and share the results in the Comments below.

Going Shopping

If you’re ready to start incorporating some of these utility mixers into your system, I’ve gathered together a collection of them on ModularGrid:

Inverting Offsetting Mixers

A collection of inverting, offsetting mixers (and more)

•      The top row of this collection contains mixers that can also invert and offset the signal in a pretty straightforward manner.

•      The middle row contains more complex control voltage processors that can achieve the goal of mix, invert, and offset, but they may require a bit more thought or patching than the top row. Although less intuitive or immediate, the upside is that these modules should provide even more creative possibilities in your modular system.

•      The modules in the lower row are each missing one of our desired features, but are still very interesting mixer alternatives in their own right. For example, the RYO Ampmix can provide up to a 2x boost for each signal it mixes, allowing you to overdrive modules that don’t have that option built into their own input controls.

I hope this note has helped you understand why you need utility mixers to help glue together your modular system, plus gave you some alternate ideas on how to use them. A future note will discuss the opposite side of mixer: multiples, including when you need a “buffered mult.”

special thanks to Michael Johnsen at Pittsburgh Modular, Walker Farrell at Make Noise, Tim Ressel at Circuit Abbey, Jason Coates at Manhattan Analog, Jeff at Synthrotek, Eric at STG Soundlabs, Dieter Doepfer at Doepfer Musikelektronik, Paul Schreiber at Synthesis Technology, and Benjamin Ljunggren at RYO/Ljunggren Audio for their help in researching this note – particularly the bit about asymmetrical distortion.