HUSH M4 mixer

1 Safety

This is a draft document for a product in development. Do not build mains-powered equipment from it without a qualified engineer’s review and the safety approvals your country requires.

Safety standards are cited by title and catalogue page only; this draft does not claim compliance with any of them.

2 About this manual

HUSH M4 is a four-channel analogue club mixer in development. The design brief started from what DVS1 has said in published interviews about how he plays (two CDJs and two Technics decks, an EQ that can take a band out, filters, no effects, booth monitoring, long sets). Appendix C maps each tailored feature to a cited quotation and lists where the design departs from him or nothing is documented. The interviews are cited as the design brief; this manual does not claim an endorsement.

Three words are used strictly. Design target: a goal the finished unit would have to meet. Calculated: worked out here from the cited formula with nominal component values and ideal op-amps. Measured: taken from a real unit with a stated method — there are none in this document. When a unit exists, a measurement report will follow the method of IEC 60268-3^R9.

The playable mixer on the web page is a software model built with the Web Audio API^R11. Its filters and effects are not the analogue circuits in this manual, and nothing it plays is a measurement. Where the browser supports it, the model runs in an AudioWorklet with 32-bit float processing at the device’s own sample rate, a four-band EQ and a split filter with exact bypass, optional texture, oversampled soft saturation that is transparent below 0 dBFS and a true-peak limiter on the output only, and it can play lossless tracks the user loads from their own device and record the master to a 32-bit float WAV file or a dithered 24-bit one. Appendix D lists what was measured on that code in a simulation. The hardware design has no limiter or compressor in the master or booth path.

3 Controls and connections

Channel strip (×4, identical)

Master section

Connections

4 Signal flow

Each of the four strips runs input select, trim, four-band EQ, filter and fader, in that order, with a pre-fader cue tap. The strips sum into the master bus; the master isolator and the (switchable) FX loop follow, then the master fader and the balanced output drivers. The cue bus and headphone amplifier sit beside the master path. A single linear supply feeds everything. The whole path is analogue.

5 Circuit description

One short table per stage. The full schematics are in Appendix B and on /mixer/schematics.

S1 RIAA phono preamp (moving magnet)

S2 Line input and trim

S3a Isolator building block

S3b Three-band isolator (master section)

S4 Channel filter

S5 Channel fader (control-voltage VCA)

S6 FX send / return

S7 Master, booth and cue

S8 Headphone amplifier

S9 Power supply

S10 USB-C interface (optional tap)

S11 Texture (optional)

S12 Spatial FX send / return (optional)

S13 Channel EQ (four-band, per channel)

Where the S13 figures come from. The four corner and centre frequencies and the boost and cut ranges are the ones Allen & Heath publish for the stereo channels 1–4 of its Xone:96 mixer^R24 (LO 180 Hz, +6 dB to kill; LO MID 350 Hz, +10 / −27 dB; HI MID 1.1 kHz, +10 / −27 dB; HI 3 kHz, +6 dB to kill). They are used as a published reference for a familiar DJ EQ layout. The circuit, the shelf and bell shapes (shelf Q 0.71, bell Q 0.9), the knob design and the panel drawing are this design’s own; the guide gives no EQ schematic, and no part of that maker’s design, artwork or logo was copied. No connection with, or endorsement by, Allen & Heath is implied. Relation to DVS1: he is quoted as saying he does not need a 4-band EQ (q6), so this four-band EQ departs from his statement; it was added at the site owner’s request. The only related account is a writer reporting him ducking a high-mid on an Xone:96 (q26), a band a three-band isolator does not have. See Appendix C.

6 Specification (design targets)

These are goals, not measurements. The third column says whether a figure is a design target or a calculation from the formulas in section 5.

Level and rail consistency

The +28 dBu maximum output equals 19.46 Vrms across the balanced pair, or 13.76 V peak per leg. With ±18 V rails and an assumed 16.5 V usable swing per leg, the driver can reach about +29.6 dBu, so 1.6 dB of margin. On ±15 V rails the same driver would top out near +27.8 dBu, which is why the supply is ±18 V and why the swing figure must be confirmed against the chosen op-amp’s datasheet.

Consistency with the /mixer page

• The /mixer specification and this manual both describe op-amp stages: low-noise NE5534 / NE5532-class op-amps for the RIAA stages and an inverting op-amp summer. No discrete or Class A stage is claimed; a discrete input stage is an open item for a later revision.
• Both state no crossfader. The VCA fader stage could host one; it is not fitted in this design.
• The +28 dBu maximum output needs ±18 V rails, as shown above, not the more usual ±15 V.

7 Calibration and alignment

A proposed procedure for a prototype. Nothing here has been carried out. Use an isolation transformer on the bench.

8 Measurement plan

When a unit exists, the figures in section 6 would be measured following the methods of IEC 60268-3^R9 and reported with the test conditions (source impedance, load, bandwidth, weighting, reference level). Until then the table in section 6 remains a list of design targets. AES17, the method for digital audio equipment, would apply to the optional USB-C interface (S10). Publication plan (method only): every result would be published with its test conditions, the instrument and its settings, the number of units tested and the raw files, and a figure would appear on the product pages only after it has been measured on a real unit. No result is given here.

9 Care and handling

Keep liquids away; the faders and pots are user-replaceable but the panel is not sealed. Clean with a dry cloth. Power the mixer on before the amplifiers and off after them. Do not block the ventilation slots. Replace the mains fuse only with the stated type.

10 Revision history

11 References

• [R1] RIAA equalization: time constants 3180, 318 and 75 µs (IEC 60098 / RIAA) — Wikipedia, RIAA equalization
• [R2] S. P. Lipshitz, “On RIAA Equalization Networks”, Journal of the Audio Engineering Society 27(6), pp. 458–481, June 1979 — AES E-Library record
• [R3] J. Karki, “Analysis of the Sallen-Key Architecture”, Texas Instruments application report SLOA024B — TI SLOA024B
• [R4] S. Linkwitz, “Active Filters”: LR2 and LR4 crossovers as cascaded Sallen-Key sections, Q0 = 0.71 per LR4 stage — Linkwitz Lab, Active Filters
• [R5] Texas Instruments, LM317 3-Pin Adjustable Regulator datasheet (Rev. Z): Vₒ = 1.25 V × (1 + R2/R1) + Iₐdⱼ·R2 — TI LM317 datasheet
• [R6] Texas Instruments, NE5532 dual low-noise operational amplifier datasheet (5 nV/√Hz typical at 1 kHz) — TI NE5532 datasheet
• [R7] Texas Instruments, NE5534 low-noise operational amplifier datasheet (3.5 nV/√Hz typical at 1 kHz) — TI NE5534 datasheet
• [R8] THAT Corporation, 2180 series voltage-controlled amplifier datasheet (control constant 6.1 mV/dB) — THAT Corporation datasheets
• [R9] IEC 60268-3:2018, Sound system equipment — Part 3: Amplifiers (specification and measurement methods) — IEC webstore
• [R10] IEC 62368-1:2023, Audio/video, information and communication technology equipment — Part 1: Safety requirements — IEC webstore
• [R11] W3C, Web Audio API (the browser demonstration on /mixer uses it; it is not the analogue design) — W3C Web Audio API
• [R14] ESS Technology, ES9842PRO 4-channel ADC datasheet v0.5.1 (dynamic range, THD+N, 32-bit PCM / TDM words) — ES9842PRO datasheet
• [R15] ESS Technology, ES9843PRO 4-channel ADC datasheet v0.4.1 (alternative part) — ES9843PRO datasheet
• [R16] XMOS, XK-AUDIO-316-MC USB audio multichannel reference: USB Audio Class 2.0, asynchronous, XU316 — XMOS sw_usb_audio, app_316_mc
• [R17] T. Stilson and J. Smith, “Analyzing the Moog VCF with Considerations for Digital Implementation”, ICMC 1996 — CCRMA, Stilson and Smith
• [R18] A. Huovilainen, “Non-linear digital implementation of the Moog ladder filter”, DAFx-04 — DAFx-04 paper archive
• [R20] J. Dattorro, “Effect Design, Part 1: Reverberator and Other Filters”, Journal of the Audio Engineering Society 45(9), pp. 660–684, September 1997 (plate topology, delay lengths and output taps, Fig. 1 and Table 2) — CCRMA, Dattorro
• [R21] Princeton Technology Corp., PT2399 Echo Processor IC datasheet V1.9, March 2022 (44 kbit RAM, VCO delay set by an external resistor, THD 0.5 % typ, SNR better than 90 dB typ, +5 V application circuits) — PT2399 datasheet
• [R22] Panasonic MN3207, 1024-stage low-voltage low-noise BBD datasheet (delay 2.56 to 51.2 ms for a 200 to 10 kHz clock, supply +4 to +10 V, THD 0.4 % typ, S/N 73 dB typ). Read from a copy hosted by Synthrotek. — MN3207 datasheet (hosted copy)
• [R23] R. Elliott (ESP), “Equalisers, The Various Types And How They Work”, Elliott Sound Products, March 2015, updated February 2021 (Baxandall and James tone controls; boost / cut stages with a band-pass filter between virtual-earth op-amp inputs; multiple-feedback, twin-T and state-variable band-pass options; gyrators) — ESP, Equalisers
• [R24] Allen & Heath, Xone:96 User Guide, channel equaliser and hardware specification (stereo channels 1-4: HI 3 kHz +6 dB / kill, HI MID 1.1 kHz +10 / -27 dB, LO MID 350 Hz +10 / -27 dB, LO 180 Hz +6 dB / kill). Used only as a published reference for the control layout and ranges; no circuit is published there. — Allen & Heath support site
• [R19] V. Zavalishin, The Art of VA Filter Design, Native Instruments. Cited by title and author; no link was checked for this draft.
• [R12] D. Self, Small Signal Audio Design, 3rd edition, Focal Press / Routledge (2020). Book, no link.
• [R13] P. Horowitz and W. Hill, The Art of Electronics, 3rd edition, Cambridge University Press (2015). Book, no link.

Interview and rider sources for the design brief are in Appendix C.

A Derivations and component values

All values come from tools/mixer_values.py; running it prints the same numbers.

RIAA

Playback curve H(s) = (1 + s·318 µs) / ((1 + s·3180 µs)(1 + s·75 µs))^{R1, R2}. U2 provides the 3180 / 318 µs shelf: for G = 1 + Zf/Rg with Zf = Rf in parallel with Cf, the pole is at Rf·Cf and the zero at Rf·Cf/(1 + Rf/Rg), so Rf·Cf = 3180 µs and Rf/Rg = 9 give a 318 µs zero. A passive R·C = 75 µs follows. A non-inverting stage cannot fall below unity gain, which is why the 75 µs roll-off is passive and the gain is made up in U3.

Sallen-Key and Linkwitz-Riley

For equal R and C: f0 = 1/(2πRC), Q = 1/(3 − K), K = 1 + Rf/Rg^R3. K = 1.586 gives Q = 0.707 (Butterworth); two such sections in cascade make an LR4 filter with the −6 dB point at f0^R4. With C = 10 nF: R(300 Hz) = 53.05 kΩ → 53.6k; R(3 kHz) = 5.305 kΩ → 5.36k.

Power

Vout = 1.25 × (1 + R2/R1) with R1 = 240 Ω and R2 = 3.24k gives 18.29 V^R5. Reservoir ripple ΔV = I/(f·C) with f = 100 Hz for full-wave rectification.

Levels

0 dBu = 0.7746 Vrms. +28 dBu = 0.7746 × 10^(28/20) = 19.46 Vrms. Balanced peak per leg = Vrms × √2 / 2 = 13.76 V.

C Design rationale and DVS1 sources

Each tailored feature is mapped to something he has said in public, with the source. Where the design departs from what he has said, or where nothing is documented, the table says so. The interviews are cited as the design brief, not as an endorsement, and DVS1 has not reviewed or approved the M4.

Where the design departs, or nothing is documented

• FX loop. He says “I don’t need effects” (DJ Times, 2014), yet his own riders ask for stereo cables for an effects box that he brings, on aux 1 (2016 rider, q17). The M4 therefore has an FX send and return, hard-bypassed and off until it is switched in. This departs from the 2014 statement and follows the riders.
• USB-C interface. He says “I don’t think a laptop belongs in a DJ booth” (q7). The optional USB-C record / playback interface departs from that: it is a tap for recording, off by default, with no DJ software and no screen, and nothing in his documented setup asks for it.
• Spatial FX. He says “I don’t need effects—I don’t need to fluff my mixes” (q4) and his riders ask only for a plain effects send and return that he fills with his own box (q17). DVS1 favours minimal effects, so the M4 spatial FX (tape echo, spring, plate, BBD ensemble) are the designer’s addition, not something he asked for. They are optional, off with the return at zero, and hard-bypassed in hardware by SW3. No source reviewed mentions spring, plate or tape echo.
• Four-band channel EQ. He is quoted as saying “I don’t need a 4-band EQ” (q6). The M4 channel EQ has four bands anyway (HI, HI MID, LO MID, LO), which departs from that statement; it was added at the site owner’s request. The only related account is a writer reporting him ducking a high-mid on an Xone:96 (q26). The frequencies and ranges follow the published Xone:96 specification, not anything DVS1 said.
• Texture section. No source reviewed mentions resonant, ladder, comb or bitcrush effects, and he says “I don’t need effects” (q4). The texture section is the designer’s addition; it is off by default and exactly bypassed when off.
• Four phono stages. The documented setup has two turntables (q1, q2), and the 2019 festival rider lists none. Four phono stages go beyond what he has said he needs; they are there so any deck can use any channel, and could be cut to two to save cost.
• Layout. He says he does not like the layout of the Xone mixers and does not say what he dislikes (q3). The M4 strip layout is the designer’s guess. Nothing documented says it is what he wants.
• Crossfader. None of the sources reviewed says whether he uses a crossfader. The M4 has none; that is a design choice, not something he asked for.
• Rotary or linear faders. No statement about rotary-fader mixers was found. The M4 uses linear faders (see above).
• Master isolator and VU meters. No source reviewed mentions either. A writer reports him ducking a high-mid on an Xone:96 (q26); that is the only related account.
• Output level and headroom. The +28 dBu maximum output and the ±18 V rails are an engineering choice. He has not said anything about mixer headroom.
• Pitch and tempo. He pitches records on Technics MK5G decks (q24). Pitch is on the turntable, so the M4 hardware has no tempo, sync or BPM function. The browser app adds BPM detection, tap tempo, a pitch fader, loops, hot cues and an optional sync as tools of the software model; they are a departure from this, not something he asked for, and sync is off until pressed.
• His riders name a mixer. Both riders ask for an Allen & Heath Xone (92, then 96) and say “STRICTLY NO OTHERS” (q30). The M4 would not meet his rider, and nothing here suggests he would use it.
• Not found. No usable interview text was found from Resident Advisor, Mixmag or Pioneer on mixer layout, rotary mixers, headphone cueing or booth monitoring. A 2016 Slam Academy video interview exists but was not transcribed, so nothing from it is used.

Context

• Browser demonstration. The Web Audio demonstration on this page has an output limiter and soft saturation so that nothing clips in a browser. The hardware design has no compressor or limiter in the master or booth path.
• Not applicable. His remark about keeping “a maximum of seven channels” in DJ Mag (2018) is about his studio, not his DJ mixer, and is not used.
• Dates. The sources run from 2013 to 2019. His setup has changed over that time (for example Xone 92 to 96, CDJ 2000 to 3000), so these are statements from their date, not today.

Sources

1. [DT] Jim Tremayne, “DVS1: Back to the Old School”, DJ Times, 5 February 2014 (archived copy) — open
2. [DM] Rob McCallum, “DVS1: The techno purist rages against the machine”, DJ Mag, 31 January 2018 — open
3. [LWE] Steve Mizek, “DJ Debriefing: DVS1”, Little White Earbuds, 21 August 2013 (archived copy) — open
4. [TK] Tobias Fischer, “Interview with DVS1”, Tokafi (undated on the page) — open
5. [XR] “Ask the Experts: DVS1”, XLR8R, 24 March 2017 — open
6. [R16] Triangle Agency, artist tech rider, DVS1 – DJ (PDF, 2016 upload) — open
7. [R19] Triangle Agency, artist festival tech rider, DVS1 (DJ) (PDF, 2019 upload) — open
8. [MD] “DVS1 Wall of Sound dominates Los Angeles”, Magical Dancefloors. A writer’s account of a show, not DVS1’s words — open

Quotations are word for word from the linked pages (checked by script against the page text). Everything else on this page is the designer’s own wording. They are cited as the design brief, not as an endorsement, and DVS1 has not reviewed or approved the M4.

D Browser model audio specification (simulation)

These are simulation results for the browser model, not measurements of the analogue mixer and not hardware measurements. They were produced by running the same AudioWorklet code in Node at 48 kHz (Float64 inside the filters, Float32 at the module boundaries) with the script tests/dsp_test.js. They describe the software on /mixer; the circuits in this manual are design targets and have no measured figures.

What the engine does

• Sample rate. The audio context runs at the device’s rate by default (48 kHz or 96 kHz where the device offers it) or at a rate the user picks; nothing is resampled inside the engine. The latency hint is the user’s choice (interactive, balanced or playback).
• Precision. 32-bit float at the module boundaries. The recorder writes 32-bit float WAV at the context rate; the optional 24-bit file uses triangular (TPDF) dither and reports any samples that were over full scale.
• Nonlinear stages. The texture drive, the strip saturators and the master saturator are oversampled 4× (2× in the efficient mode) with half-band filters. The saturator is the identity below 0 dBFS, so below that level nothing is shaped and the bypass is bit-exact; above it the curve bends smoothly towards +18 dB of internal headroom. The spatial FX’s own tape, spring and BBD saturators are not oversampled yet; they run at the base rate inside feedback loops.
• Smoothing, DC and denormals. EQ gains and filter positions are smoothed over about 12 ms; the texture wet path has a 5 Hz DC blocker; the texture filter states are flushed to zero below 1e-25 so no denormals arise.
• Limiter. A look-ahead true-peak limiter at −1 dBTP on the output only, using an 8× detector. It does not touch the signal below its threshold. Channel strips have no limiter.
• CPU. The master watches whether render callbacks keep up with real time. If they fall behind three windows in a row the engine switches itself from 4× to 2× and says so on the page. Phones and machines with four cores or fewer start at 2×.
• Track decks. Each of the four decks takes a lossless file from the user’s own device: WAV (16, 24, 32-bit integer or 32-bit float), AIFF or AIFF-C (PCM, little-endian “sowt” or float) and FLAC. Anything else, including MP3 and other compressed formats, is refused by reading the file’s header, not its name. WAV and AIFF are decoded by the page’s own reader into 32-bit float without rounding (a 16 or 24-bit sample divided by 2^bits−1 is exact in Float32); FLAC is decoded by the browser. The file stays in the browser tab: nothing is uploaded and the page makes no request that carries it. A loaded deck replaces its test signal and runs through the strip’s EQ, filter, texture and saturator, the FX sends, the master and the recorder. The strip processes a stereo track with two identical chains. The deck’s panner is centred for a track, so a stereo track enters the master at unity on each side, while the mono test signals sit at −3 dB per side. When a file’s sample rate differs from the engine’s, the browser converts it as it plays (the deck says so); there is no “no resampling” claim for decks. Rates from 8 to 96 kHz are accepted.
• The radio. The site’s radio player uses the highest-quality stream each provider officially offers; some provider previews are lossy and the site does not re-host them.

Limits of this table: a simulation of the code, a single noise seed, ideal 32-bit and 64-bit arithmetic rather than a browser’s audio stack, and no converter or analogue stage. Nothing here supports a claim about how it sounds.

E Browser app: DJ tools, sessions and offline (simulation)

These are checks of the browser app on synthetic test signals made for the test. They are not measurements of the analogue mixer, they use no real music, and they say nothing about how the app sounds. The app runs only in the user’s browser; nothing is uploaded.

What it adds to the browser model

• Install and offline. A web app manifest and a service worker limited to /mixer. After one online visit the page, scripts, fonts and images of /mixer load without a network. The other pages of the site, the documents and the fan room are not cached by it. The cache is versioned and older versions are removed when the worker updates.
• Saved sessions. The mixer, EQ, FX, hi-fi and deck settings are kept in the browser’s localStorage and restored on reload, can be saved under a name, and can be exported and imported as JSON (settings and cue points only, no audio). Tracks are stored only if the user ticks the opt-in box, then in the browser’s IndexedDB, and can be removed again.
• Per-deck tools. Tempo detection, tap tempo, a pitch fader of ±8% or ±16%, nudge, loops of 1/4 to 16 bars on the detected beat grid, four hot cues and a sync option that is off until pressed. Pitch and tempo move together, like a turntable: key lock (changing tempo without changing pitch) is not available in this version.
• Library, MIDI, keys. A searchable list of the tracks loaded in the tab. Web MIDI learn and map with save and load, where the browser has Web MIDI (not Safari). Keyboard shortcuts with a help sheet, touch gestures, and a first-run tour that can be skipped.

The hardware design is unchanged. Appendix C records that DVS1 pitches records on the turntable (q24) and that the M4 hardware has no tempo, BPM or sync function. The app’s tempo tools are an addition of the software model, not something DVS1 asked for, and the sync option is off by default.

Tempo detection on synthetic tracks

Run in Chromium through the real app. In addition, 80 synthetic tracks (click and kick plus hat, 80 to 180 BPM, 44.1 and 48 kHz, 40 s) were run through the same detector in Node: 78 read within 0.05 BPM, worst error 0.01 BPM, worst first-beat error 1.67 ms. The misses are octave errors: kick 80 read as 160. The detector looks at the first 150 s of a track, searches 70 to 190 BPM and assumes a steady tempo; real recordings with swing, tempo drift or no clear kick were not tested.

Loops, hot cues, pitch and sync

Limits

• Live playback at a pitch other than 0% is positioned by the browser’s resampler; the sample-exact figures above are for rate 1.0 (loops, cues) and for the offline position check at +5%.
• The loop and cue figures are for an offline render of the same source node settings the decks use; the audio device and the browser’s scheduling add their own latency on a live deck, which was not measured.
• MIDI learn takes absolute controllers (knobs and pads). Relative encoders, pickup and 14-bit controls are not supported. It was tested with a scripted controller, not hardware.
• Only Chromium was tested. Safari has no Web MIDI; other browsers were not run.
• Interaction timing was measured on a build machine without a graphics card, so it says little about phones.