Mod Squad: Frap Tools Fumana

Dynamic filtering at its most affordable

One of the most exciting Eurorack module releases in the past few months is the Frap Tools Fumana. Referred to as a Dual 16 Bands Spectral Editor [sic], the Fumana is based on the Buchla 296 and 296e Spectral Processor, which you could think of as a dynamic 16-band graphic EQ, where each frequency band under voltage control. The module’s purpose is to give you the ability to transfer the energy of eight frequency bands to another part of the spectrum, including movement from the even to odd bands, and vice versa.

When I reviewed the Buchla 200e system for EM in 2011 (available at, the 296e was one of my favorite modules to use—not only for vocoder-like duties, but as a wide ranging, real-time spectral processor of audio signals with an immediately recognizable character. Priced at $4,600 as it was at the time, made it even more exotic. (Because two of the modules can be linked together for even greater processing power, Don delivered a pair of the 296e's to me personally for the review, in a shoe box. The contents of that box were worth more than my car at the time!)

Rather than duplicating the exact sound and functionality of the Buchla 296 concept, Frap Tools decided to create a spectral processor that could be used as much to complement external sounds. Consequently, the Fumana was engineered to provide a flat, uncolored sound when the controls are in their nominal settings.

And remarkably, the module is priced just below a grand, coming in at nearly a fifth of the cost of the 296e module. If the Fumana’s price tag seems steep for a Eurorack, check out what it does in the analog realm and you’ll begin to see that its price is justified.


For a Eurorack module, the Frap Tools Fumana gives you unprecedented control over the frequency spectrum.

For a Eurorack module, the Frap Tools Fumana gives you unprecedented control over the frequency spectrum.

The Fumana is comprised of two 16-band, fixed-filter banks, one referred to as the Main (or Synthesis) filter array, the other as the Modulator (or Analysis) bank. Each band of the Main filterbank has its own VCA input, whereas each band of the Modulator array has an envelope follower output. When only one audio input is present, it is nor-malled to the other audio input; similarly, when only one modulation signal is patched in, it is nor-malled to both inputs.

The tops of the faders light when the corresponding band lets signal through as a result of any signal input to the VCA and envelope (or based on the Tilt and Scanner settings, discussed later in the review).

The bands in both filter banks are numbered 1 through 16, and you can control the odd-numbered bands independently of the even-numbered bands. Markings on the front panel make it easy to see which banks are odd and which are even.

A separate audio output for the Odd and Even bands is in the upper left corner of the panel. The All output provides both sets of frequency bands but processed through a lowpass filter set to 18kHz. Both the Odd and Even bands also have their own phase-inversion switch, which is useful for creating effects by combining their outputs with the All signal.

Additionally, each band has a discrete output that taps the signal before the VCA. These stretch across the top of the module and work independently of the Odd, Even and All audio outputs.


And interesting feature of the Fumana is the spectral Tilt function (controlled by the larger knob circled in green), which gives you control over the upper and lower eight bands as a group.

Imagine the frequency spectrum is sitting on a see-saw, with the fulcrum point placed between bands 8 and 9. By turning Tilt clockwise, you emphasize the treble bands 9 through 16, while simultaneously attenuating the low-frequency bands, 1 through 8. Turning it counterclockwise boosts the low end from bands 8 to 1 while gradually cutting the highs from band 9 upward. Tilt has its own CV input and attenuverter.

The other three knobs (circled in yellow) are the scanning parameters and work in tandem to determine which parts of the overall frequency spectrum are emphasized or cut. The Peak/Notch knob (on the left) determines the amount of emphasis or attenuation, whereas the Width knob (on the right) sets the number of bands involved (0 to all 16 bands). The position of the Center control weights the emphasis towards the upper or lower parts of the spectrum. The CV input for each scanning parameter also includes an attenuverter to adjust the external modulator.

For example, when the Peak/Notch knob is fully counterclockwise and the Width knob is fully clockwise, the middle knob, Center, controls which side of the frequency spectrum is heard. You can turn Center clockwise to notch the highs, turn it counterclockwise to attenuate the lows, or patch a CV into it and externally control the notching out of frequencies from a modulation source. With Peak/Notch and Width fully counterclockwise, you will hear a narrowly emphasized set of bands as you turn the Center knob (or control it with an external CV).


That pair of modulation inputs (positioned above the audio inputs) can be used to control the 16 envelope followers at once with one or two external signals. As mentioned earlier, if a signal is plugged into only one of the modulation inputs, it is normalled to the other input to control all 16 bands. Plug two modulations sources in and the left input will control the odd-numbered bands while the right input controls the even-numbered bands. You will need signals with a sharp edge to drive the envelope follower. The All EF (envelope follower) output jack in the upper left corner provides a summation of those signals.

The pair of small knobs in the bottom left corner represent attack and decay settings that can shape the modulation inputs. Interestingly, they have a non-linear response, which adds musicality to the shaping of the envelopes. Working with these in conjunction with the Tilt and Scanning controls allowed me to further focus how incoming modulation signals altered the frequency spectrum of the output.

For example, the Fumana can produce a woody, popping sound reminiscent of a lowpass gate, but with greater control over the frequency components due to the variety of parameters this module offers. The effect can be further enhanced based on the settings of the faders: The higher a fader is set, the more that that frequency band will be heard when an envelope signal is not present.

While you are using the Fumana in this way, you can use the individual envelope follower and VCA inputs to highlight individual frequency bands. Patching the four gate outputs of my Make Noise Pressure points to four VCA inputs allowed me to sustain specific frequency bands against the main rhythm created with the Modulation input. Moreover, I was able to place the sustained sounds in different channels by panning the Odd and Even outputs to opposite sides of the stereo field. The effect could be further enhanced by routing the individual outputs of these select frequency bands through a dedicated echo: Instant filtered dub effects!

I chose these extreme examples to demonstrate how radically the Fumana can alter the spectral content of a rich source signal. In conjunction with the independent CV inputs and attenuverters, these controls also let you dial in very subtle motion, particularly when you’ve panned the odd and even outputs to opposing sides of the stereo field.

A more practical application includes using an envelope to dynamically clear out muddiness in a rhythm part or to automatically enhance a particular frequency component of, say, a kick drum or bass part. In other words, you can use the Fumana as a fancy equalizer that is controlled by the audio material you’re working with, explore it as a creative tool, or use it in some combination of those two approaches.


The Fumana has a slightly different take on the vocoder concept of unvoiced detection. Typically you would patch a noise source into the unvoiced input (which, in this case, has its own level control) to recreate the sibilant sounds found in speech.

On the Fumana, the Unvoiced input works in tandem with bands 14 and 15 (and their independent VCAs). The result is less pronounced than what you would expect from a dedicated vocoder, but it extends the capabilities of this feature to use with other kinds of audio input besides noise.


If dynamic filtering is part of your world, the Fumana is something you need to hear. It’s ability to transform the timbral aspects of audio input is remarkable, and the fidelity in the filter quality makes the module useful in both a standard production role as well as a more adventurously musical one.

The Fumana is by no means cheap, in any sense of the word, but you definitely get your money’s worth.