The Focusrite Liquid Channel ($3,495) is perhaps the most revolutionary piece of audio equipment introduced in the last decade. The box applies Dynamic Convolution technology that can sonically emulate virtually any mic preamp and/or compressor. Dynamic Convolution is the mathematical technique of determining a system output, given an input signal and a system impulse response. This means that if you know what is coming in to your system, and you can control your system's impulse response, the system's output can be defined.

Product Points

Applications: Studio


Key features: 24-bit; 44.1 kHz - 192 kHz sample rates; preamp and compressor controls; 40 preamp presets; 40 compressor presets; 100 memory settings; three-band EQ; Dynamic Convolution technology


Price: $3,495


Contact: Focusrite/Digidesign at 866-362-8774, Web Site.


Through the use of Dynamic Convolution, Focusrite has developed a way to accurately emulate the way that compressors and mic preamps affect sound. While this sounds similar to modeling, it is actually a totally different beast. Modeling looks at the way a device works and then relies on the generation of code to try to emulate the way a device would respond, usually in a certain limited set of situations. Convolution records data about the way a device behaves and then emulates that behavior. The Liquid Channel ships with emulations of 40 classic mic preamps and 40 classic compressors (see for more specific descriptions). These emulations can be combined into 100 user memories. A user memory stores all parameters including mic pre gain, EQ and compressor settings.




The Liquid Channel is a 2RU rackmount device. The 24-bit machine supports 44.1 kHz, 48 kHz, 88.2 kHz, 96 kHz, 176.4 kHz and 192 kHz sampling rates and it makes use of a custom precision-wound FF "Liquid" transformer, designed and built in the UK by Focusrite's R & D department. The transformer has the ability to be either transparent or colored as needed.


The Liquid Channel's ADC signal to noise ratio is 120 dB measured with 20 Hz/22 kHz bandpass A-weighted filter. The frequency response is 0.05 dB from 20 Hz - 20 kHz, the maximum input level is +22 dBu and the THD+N is 0.00035% (-109 dB). The DAC dynamic range is 116 dB measured with 20 Hz/22 kHz bandpass A-weighted filter and the frequency response is 0.05 dB from 20 Hz - 20 kHz, the maximum output level is +22 dBu and the THD+N is 0.0007% (-103 dB). The Liquid Channel's latency is minimal. At worst case scenario (analog in, analog out, all sections in circuit, 44.1 kHz), it is less than 4 milliseconds. At 96 kHz analog in/out it is 1.6 milliseconds.


The LiquidControl software, which allows computer control of the Liquid Channel as well as the ability to load and save programs, is included with the Liquid Channel and is available free of charge from the Focusrite web site. Focusrite supports both PC and Mac versions of the software and since it's free, anyone who routinely works at a studio equipped with a Liquid Channel, even if they don't own one themselves, would find it advantageous to download the software.


The Liquid Channel's rear panel provides all of the input and output connections. Two female XLR connectors provide microphone and line input. A male XLR connector provides analog output. AES digital input and output are provided via a female and a male XLR connector. Word clock in is provided via a 75 ohm BNC connector. A second BNC connector provides word clock out. Two female RCA connectors provide Digital Link Bus in and out.


The line input has a gain range of -10 dB to +10 dB, variable in 1 dB steps with a maximum input level of +22 dBu. The microphone preamp has a gain range from +6 dB to +80 dB, switchable in 1 dB steps with a maximum input level of +16 dBu. The input impedance is variable (determined by the selected preamp). To accommodate the full range of different microphone designs the Liquid Channel allows flexibility in the resistance and capacitance parameters of its transformer. The box physically changes analog circuitry and uses dynamic convolution technology to create mic preamp emulations. To duplicate the sound of transformerless electronic or tube mic preamp, the Liquid Channel's transformer is autoswitched out of the circuit. Focusrite has actually built in the numerous variations required to reproduce the vagaries of a range of electronic mic preamps. The capacitance and resistance are varied in the circuit, and Dynamic Convolution technology is used to emulate the full range of electronic mic preamps. The Dynamic Convolution process also provides accurate tube emulation.


The front panel's dials are all tactile rotary encoders. This means that they can be rotated infinitely and their relative value will be displayed (except for the DATA wheel) by the LEDs surrounding them. When active, the exact numerical parameter values can be seen on the large display in the centre of the front panel. Since the controls are digital, their settings can be easily saved, recalled and edited from the Liquid Control software application.


The power switch applies power to the Liquid Channel. A single, vertical, peak hold LED bar graph displays the level of the signal being fed from the mic preamp to the A/D converter. The scale shows from -20 dB FS up to the maximum 0 dB FS (digital clipping level) with an additional LED to represent overload (O/L). The Digital Clip LED is a single red LED that signifies when a digital clip is occurring. The input Select switch steps through the mic, line and digital inputs. An LED illuminates to signify which input is active and a second LED illuminates when the transformer has automatically been switched in circuit. The dial below the input select switch adjusts the level fed to the A/D. The levels for mic and line are shown by the outside (mic) and inside (line) arc around the dial. The exact numeric value is displayed next to the pre-amp replica name on the front panel LCD. Pressing the +48V switch provides 48V phantom power to the rear panel XLR microphone connector. Pressing switch inverts the phase of the selected input. The HP switch activates an analog, pre-A/D high-pass filter into the signal path. It is switchable between 75 Hz and 120 Hz. The rolloff is 12 dB per octave, 6 dB down at 75 Hz or 120 Hz.


The "Session Saver" switch activates the Session Saver function. When activated (indicated by the corresponding LED), this circuit applies gain reduction if the signal starts to clip, providing protection from continual overload if the level is consistently too hot. The signal isn't compressed, just turned down to avoid overloading. The feature monitors the signal at the input and output. If the input is overloading, the mic pre gain is reduced. If the output is clipped but the input is clean, the level of the makeup gain is lowered. When the section has reduced the gain, the Activated LED illuminates. Once the level is decreased, it remains activated until it is reset.


The Clock Select switch steps through the sample rate options indicated by the corresponding LED. Options are 44.1 kHz, 48 kHz, 88.2 kHz, 96 kHz, 176.4 kHz or 192 kHz. The Ext switch allows The Liquid Channel to synchronize to an external source, either the word clock or AES input on the rear panel. When a lock is achieved, the corresponding LED illuminates.


The main LCD shows a variety of information including preamp gain, program name and program number. To select a compressor, simply press the Compressor Select switch and rotate the DATA wheel until the desired compressor replica is displayed and then press the DATA wheel to load. The same process is followed to load a mic preamp replica.


The harmonics dial allows harmonic distortion to be applied to the signal. The exact amount of 2nd, 3rd and 5th-order harmonics depends on the type of preamp chosen and the amount of mic pre gain. This control allows the subtle variations between various analog devices of the same model to be emulated and allows the user to create an overdriven sound without having to overdrive the preamp. The amount of harmonic distortion is indicated by the LEDs surrounding the dial and a value from 0-15 shown above it. '+ODD' appears next to the value when third and fifth order harmonics are present.


The compressor is activated by pressing the Comp In switch. The Threshold control sets the level at which the compression begins. The Ratio control determines the amount of compression applied to the signal with increasing input. The Attack control determines how quickly compression is applied once the level of the source signal has risen above the threshold. The Release control determines how quickly the compression is removed once the level of the source signal has fallen below the threshold. The exact attack and release values are shown in milliseconds (mS) in the display window. The Makeup control allows for compensation of the compressors reduction in level. The range of makeup values varies depending on the chosen compressor. The exact value is shown in dB in the display window. The vertical LED Gain Reduction meter indicates the amount of gain reduction in expanding increments down to -15 dB.


The box's digital three-band EQ is based on the curves of the ISA110 (this circuit does not use convolution technology). The EQ In switch activates the EQ section. The Pre Comp switch allows the EQ to be moved to a position prior to the compressor in the signal path. The Show Value switch causes the exact values of each EQ dial to be shown in the display window allowing for more accurate parameter modification. All three EQ switches illuminate when active. The EQ has a high shelf, a parametric mid band and a low shelf. The high shelf has a frequency range from 200 Hz to 20 kHz and a variable gain of 18 dB. The parametric mid band has a frequency range of 100 Hz to 10 kHz and a variable gain of 18 dB. The Q switch toggles the Q between 0.8 and 2.5. The low shelf has a frequency range or 10 Hz to 1 kHz and a variable gain of 18 dB. The Sidechain EQ switch routes the parametric mid-band EQ section to the side chain of the compressor for frequency-conscious compression. The Sidechain Listen switch allows the side chain to be monitored independently. Both side chain switches illuminate when active.


The bypass switch allows the user to hard bypass the harmonics, compressor and EQ sections. The Compare switch allows the user to temporarily revert back to the saved program memory settings, allowing an A/B comparison between the saved and adjusted session setups. The Revert switch returns the box to the original saved settings.


Rotating the DATA dial allows the pre-amp and compressor replicas to be chosen and allows various parameters of the replicas to be modified. Pressing the Save switch allows the current session setup (all front panel controls) to be saved to an allocated program memory slot. The Recall switch allows a program memory to be loaded. Using the Name switch in conjunction with the DATA wheel allows a unique name to be assigned to each program.


The Setup menu allows various Liquid Channel settings to be adjusted. These settings include the LF filter's frequency selection or the toggling between 'As Original Model' and 'FREE' modes. In the 'As Original Model' mode the compressor's parameter ranges are the same as on the original unit being emulated. In the 'FREE' mode, the threshold range is adjustable from -40 dB to 20 dB switched in 1 dB steps, the ratio is adjustable from 1:1 to limit, the attack is adjustable from 0.1 mS to 2.5 S, the release from 0.1 mS to 2.5 S and the make-up gain is adjustable from -20 dB to +20 dB switched in 0.5 dB steps. This mode allows settings to be created that were not possible on the original units. When linking multiple Liquid Channels together, the Digital Link Bus setting specifies how many machines are to be linked. Up to eight Liquid Channels may be chained via the Digital Link Bus connectors. When digitally linked, the Liquid Channel unit's controls can be set to follow any changes made to the master Liquid Channel's controls. (The slave unit's controls become inactive.) This means mic-pre and compressor replicas called up on the master unit will also be called up on the slave unit(s). Optionally, the link mode can be configured where the user can continue to edit the slave's controls independently, even though the master and slave units' gain reductions are linked.


In Use


After a lot of hype, both good and bad, I was anxious to plug the Liquid Channel in and see if the box could truly deliver. I'm happy to report that it can and it does do an amazing job of duplicating the sound of a long list of vintage equipment and while the sound isn't identical in every way, it is extremely close and never disappointing.


I initially put the box to work on a new album by the Appalachian Christmas Quartet. I used the Liquid Channel on a variety of instruments including bass, acoustic guitar, mandolin and flute and was always pleased with my results. Due to scheduling restrictions, my assistant Chris Brown ended up doing a lot of the recording. This typically would have required extremely accurate documentation of my recording settings as he was quickly switching from one instrument to another but since I saved every one of my settings, at the push of button he was able to recall my exact preamp model, its settings, the EQ settings, the compressor model and its setting. All he had to document was the microphone used (most often a Royer Labs SF-1) and its position.


I made use of the Liquid Channel in a wide variety of instances over several months and always had positive results. I used the box along with an AKG D112 on kick drum and had fantastic results. I used it along with a Shure SM57 to record snare drum and again had great results. I also had good results using the box with a Royer R-122 to record electric guitars and with an Earthworks SR-77 to record acoustic guitar. The box worked extremely well as a vocal chain. I used it with the Brauner VM-1KHE, the BLUE Cactus and the Sony C-800G and had wonderful results in every instance.


The Liquid Channel has so many options that I found it easy to be to focused on the options rather than on making music. After a fair amount of experimentation I found that I have the best results when I initially set a rough mic pre level and then quickly scroll through the mic pre options while the vocalist is singing or the musician is playing. Once I come across a mic pre that sounds considerably better than any of the others I move on to compressors and follow the same procedure. Next, I rough in the EQ and then return to the mic pre to do the final tweaking. This is followed by the final tweaking on the EQ and then finally the final tweaking on the compressor. If the sound is still not completely there I might tweak a bit more where needed but typically this isn't the case. The only time I have varied from this scenario is when I've been trying to duplicate a sound recorded on the actual gear the Liquid Channel emulates. In this case I simply recall the exact models and match the settings. The few instances when I've done this, the sound has been so close that a subtle harmonics or EQ adjustment has been able to match the sound completely.


During a mix session, I routed a rather sterile vocal back into the front end of the Liquid Channel feeding both the mic preamp and compressor convolution engines and had wonderful results reworking the vocal sound. It felt a bit awkward at first, but I found that being able to sonically tweak the tracking settings during a mix is a pretty amazing tool.


I found the Session Saver to be a nice feature. Many experienced artists I work with expect to walk into the studio, sing their vocal down two or three times and then be done. This means every take counts and often there isn't a pass to get an adequate vocal sound. The Session Saver feature notices when digital overs are in danger of occurring, and automatically reduces the analogue pre-amp gain by 1dB for any level above 0 dB FS.





I still believe that there is no substitute for the real thing when it comes to microphone preamps, equalizers and compressors but if you can't have the real thing the Focusrite Liquid Channel is the next best thing.