> ## Content Index
> Fetch the complete content index at: https://abovethefader.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# Where hum actually comes from
- URL: https://abovethefader.com/where-hum-actually-comes-from/
- Published: 2026-09-01T19:25:42.000Z
- Updated: 2026-09-01T19:25:42.000Z
- Description: Power distro, grounds, and the sixty-cycle tax. A field guide to the noise your rig makes when nobody is playing.
- Author: Tal Kocen
- Tags: The Woodshed

*Power distro, grounds, and the sixty-cycle tax. A field guide to the noise your rig makes when nobody is playing.*

It's 4:50 on a Saturday and soundcheck ended twenty minutes ago. The band is at catering. You mute the board, walk the room one last time, and there it is: a low, steady drone coming off the mains, sitting under the air conditioning like a pedal tone. You didn't hear it an hour ago. Nothing changed. Except something did, because the lighting guy just brought up his rig for focus, the video wall went from standby to test pattern, and the promoter plugged a margarita machine into the same quad box as your drive rack. You solo your way through forty inputs and it's not one of them. It's everything, a little. The hum isn't in a channel. It's in the building, and your PA has volunteered to be its loudspeaker.

Every one of us has spent part of a load-in on our knees behind a rack, unplugging cables one at a time like we're defusing a bomb. This piece is about skipping most of that. Hum is not bad luck. It's physics, it obeys Ohm's law, and once you know the three or four mechanisms that actually cause it, you can usually name the culprit from the front of house before you touch a single connector.

## Sixty cycles, and what pitch tells you

First diagnostic tool: your ears, and specifically your sense of pitch. Hum is not one noise. It's a family, and each member tells you where it came from.

A pure, round 60 Hz tone (50 Hz if you're working in Europe) is the AC line fundamental, and almost nothing in a modern rig produces it except two things: magnetic induction and ground current. A wall-wart transformer sitting on a mic cable, a power amp's toroid three inches from an unbalanced patch, or line-frequency current flowing through a cable shield that's part of your signal path. If it's clean 60 with barely any harmonics, think magnetically coupled or ground loop, and think about physical proximity and cable runs.

A 120 Hz tone, one octave and change up, points inside a box. Power supplies rectify AC to DC, and full-wave rectification produces ripple at twice the line frequency. When you hear 120 Hz, some piece of gear has a failing filter capacitor or a supply that was never great to begin with. That's a repair ticket, not a wiring problem.

The distinction earns its keep the first time you use it under pressure. A festival patch, a monitor engineer waving at you, thirty seconds to make a call. Pure low drone on the bass rig: that's ground current or a wall wart, go look at the backline power. A hard-edged rasp that breathes with the video wall: that's not your problem to solve at the stage box, it's a power-separation problem, and the fix is a conversation with the site electrician, not a different DI. Getting the diagnosis right in your head before you move means you fix the actual disease instead of medicating symptoms one channel at a time.

Then there's buzz. Buzz is hum with teeth: a harmonic series stacked on top of the fundamental, reaching up into the presence region where your ear hates it most. Buzz means something is chopping the AC waveform and spraying the harmonics around. The classic offender is the SCR or triac dimmer, which works by switching on partway through each half-cycle. That sharp turn-on edge is rich in harmonics, which is why lighting buzz sounds angry rather than warm. Switch-mode supplies, LED walls, and motor speed controllers pull current in ugly non-sinusoidal gulps and do the same thing. When the buzz tracks the lighting console, when it changes as the LX op runs a fade, you already have your answer, and no amount of re-patching your own rig will fix a noise that's being manufactured forty feet away.

A word on the magnetic version, because it's the one people forget exists. A transformer is two coils sharing a magnetic field on purpose. An unshielded wall-wart sitting on your snake trunk is one coil looking for a partner, and any nearby cable loop will do. The field falls off fast with distance, which is why the fix is comically simple: move the wart. A foot is usually plenty. Rotate it if you can't move it, because the coupling depends on orientation as much as distance. This is the same physics that gave the guitar world the humbucker: two coils wound in opposition so the induced hum cancels while the string signal doesn't. Your rig has no humbuckers. It has whatever geometry you left it in at 2 a.m., which is why the hum that wasn't there yesterday often traces back to a power supply that got tossed six inches to the left during the changeover.

So before you touch anything: is it hum or buzz, and what's the pitch? Sing it if you have to. Sixty means coupling or ground current. One-twenty means a power supply. Buzz means chopped current somewhere on the same service. That one habit will save you more time than any gadget in your workbox.

Fig. 1 - What pitch tells you: hum vs. buzzLine spectra: pure 60 Hz hum (teal) vs. dimmer buzz, harmonics of 60 Hz falling off as 1/n (rust)601202505001000200040000\-20\-40\-6060 Hz humdimmer buzz: harmonic series into the presence regionfrequency, Hz (log scale)level, dB

Fig. 1: hum is a note, buzz is a chord. Pure 60 Hz vs. a dimmer's harmonic series.

## The loop nobody wired on purpose

Here's the mechanism behind most of the misery, and it's worth actually understanding rather than just fearing.

Take two pieces of gear, each with a grounded power cord. Your console at front of house, your drive rack on stage. Each chassis is bonded to safety ground at its own outlet, as it must be. Now run a cable between them. The cable's shield connects chassis to chassis, and you have just drawn a complete circle: shield from A to B, then back from B to A through the building's ground wiring. A loop. A loop of copper, sitting in a building full of transformers, fluorescent ballasts, and fat conductors carrying line current. A loop is one turn of a transformer secondary, whether you meant it to be or not, and the building induces a circulating current in it.

On top of that, the two outlets are rarely at the same potential. Ground wiring has resistance, other equipment leaks small currents into it, and so "ground" at the stage end might sit a volt or two away from "ground" at front of house. That difference drives current through your shield too, because your shield is now the lowest-impedance path between them.

None of this would matter if the shield carried no signal. In a balanced line it mostly doesn't, which we'll get to. But in an unbalanced connection the shield IS the signal return. Whatever voltage the ground current develops across the shield's resistance gets added directly, in series, to your audio.

Run the numbers, because they're sobering. Decent braided shield runs on the order of 30 milliohms per meter. A 30-meter unbalanced run is about an ohm of shield resistance. Push a modest 100 mA of circulating ground current through it and Ohm's law hands you 100 mV of 60 Hz sitting in your signal. Against a consumer-level signal of −10 dBV, about 316 mV, that hum is only 10 dB below your program. Not "audible in quiet passages." Dominant. Even a polite 1 mA of ground current gives you a millivolt, which is 50 dB down: quiet enough to ignore on a rock stage, loud enough to sink a theater piece or a broadcast feed.

Fig. 2 - The loop nobody wired on purposeTwo grounded chassis, one shield: the circuit you built by accident, V(hum) = I(ground) x R(shield)console (FOH)drive rack (stage)cable shield (R = 1 ohm / 30 m)safetyground Asafetyground Bbuilding ground wiring (has resistance; carries other equipment's leakage)circulating ground current I100 mA x 1 ohm = 100 mV of 60 Hz in series with the signal

Fig. 2: the accidental circuit. Ground current through shield resistance lands in series with your signal.

Two details worth filing. First, loop area matters. The voltage a stray magnetic field induces in your accidental loop scales with the area the loop encloses, which is why cable dressing is not just aesthetics. A signal run and its return path spread across opposite sides of the stage enclose half the venue; the same run dressed together encloses almost nothing. Keep signal and power apart where you can, and when they must meet, cross them at right angles instead of running them in parallel down the same truss. Every old hand who ever barked at you about cable discipline was teaching electromagnetics, whether they knew it or not.

Second, the ritual of unplugging one cable at a time is not wrong, it's just unordered. You're opening loops until the current loses its path. Do it with a map instead of at random: the loop needs two ground bonds and a cable between them, so start with the connections that bridge different power sources or different systems. The FOH-to-stage run. The feed to the recording rig. The consumer gear. You'll find it in three moves instead of thirty.

That's the whole trick of the ground loop. Not mysterious, not haunted. Current, times resistance, in series with your audio. Every fix you've ever seen work (transformers, ground lifts, balanced lines, one power source) works by attacking one of those three terms: kill the current, remove the resistance from the signal path, or make the voltage common to both conductors so it cancels.

## Balanced doesn't mean what the brochure says

Most of us were taught that balanced audio works because the two conductors carry equal and opposite copies of the signal, and the input subtracts them. Nice story. Also wrong, or at least beside the point, and the man who spent a career straightening this out is Bill Whitlock of Jensen Transformers. Whitlock has been beating this drum in AES papers and seminars since the mid-nineties, including a landmark paper in the same June 1995 AES Journal issue that carried Muncy's pin 1 work. That issue is arguably the most consequential thing the Journal ever printed for working sound people, and most of us have never read it. Fine. That's what this column is for.

What makes a balanced interface reject noise is not signal symmetry. It's impedance symmetry. The two signal conductors must have equal impedances to ground. Do that, and any noise coupled into the line (ground-potential difference, magnetic pickup, dimmer hash) appears equally on both conductors as a common-mode voltage, and the differential input subtracts it away. The signal itself doesn't need to be symmetric at all. You can drive one leg and ground-reference the other through a matched impedance, and the interface is still balanced. Plenty of pro gear does exactly that.

The measure of how well the subtraction works is common-mode rejection ratio, and here's where the brochure numbers fall apart. CMRR gets specified on a bench with perfectly matched source impedances. In the field, your line is driven by a real output with real component tolerances. A few ohms of imbalance between legs is normal; a coupling capacitor aging on one leg can make it far worse. How much that imbalance costs you depends on the input's common-mode impedance. A garden-variety electronically balanced input built from 10 kΩ resistors will lose serious rejection from a source imbalance a transformer input wouldn't even notice, because the transformer presents an enormous common-mode impedance. As a rough rule from Whitlock's work, rejection scales with the ratio of common-mode input impedance to source imbalance: raise the first or shrink the second. This is why a good input transformer, or a bootstrapped input stage like Whitlock's InGenius design, holds high CMRR against real-world sloppy sources while cheaper differential inputs quietly give up 20 or 30 dB of the rejection printed in the spec sheet.

Put numbers on it, using the approximation from Whitlock's papers: rejection scales roughly as the ratio of the input's common-mode impedance to the source's impedance imbalance. A typical active input with 10 kΩ common-mode impedance, fed by a source with 10 Ω of imbalance between legs, gets you about 60 dB. Respectable. Let that imbalance grow to 100 Ω, a tired electrolytic on one output leg, and you're at 40 dB, which is audible on a quiet stage. Now swap the input for a transformer or a bootstrapped stage presenting megohms of common-mode impedance, and the same sloppy source still yields 100 dB plus. Same cable, same noise on the line, 60 dB difference in what reaches your preamp. This is the honest argument for input transformers in 2026, decades after "transformerless" became a selling point: not warmth, not mojo, just an input that forgives the real world.

Fig. 3 - Why real-world CMRR is not the spec sheetCMRR \~ 20 log10(Zcm / dZ): common-mode input impedance vs. source imbalance (after Whitlock)0.11101001 k20406080100120140transformer / bootstrapped, Zcm = 10 Mbetter active, Zcm = 100 ktypical active, Zcm = 10 k\~10 ohm: normal output tolerancesource impedance imbalance, ohms (log scale)common-mode rejection, dB

Fig. 3: rejection vs. source imbalance for three input types, per Whitlock's approximation.

The practical takeaways: balanced lines are still your best friend, and the loop current that ruins an unbalanced run mostly flows harmlessly on a balanced run's shield, outside the signal path. But "it's balanced" is not a guarantee. It's a system property that both ends have to cooperate on, and the cheapest link sets the budget. A balanced input fed from an unbalanced source through a properly wired cable still rejects ground noise; the same input fed through a cable someone "fixed" by floating one leg rejects nothing. When a balanced line hums, check how it's actually wired before you blame the gear on either end.

## Pin 1, the ten-cent mistake

In 1994, Neil Muncy presented an AES paper (published in the June 1995 Journal of the AES) that named a defect audio people had been chasing for decades without a word for it. He called it the pin 1 problem, after the shield pin of the XLR connector, and it goes like this.

The shield of your cable is an extension of the chassis. Its job is to catch interference and dump it to the enclosure, where it flows around the outside of the metal box and never meets the audio circuitry. That only works if pin 1 connects to the chassis, right at the point of entry. But for years, designers routed pin 1 to the circuit board instead, tying it to audio signal ground somewhere in the middle of the layout, because it was convenient and it passed the bench test. Now any current flowing on the shield, and there's always some, gets escorted directly into the reference for the audio circuitry. The noise current shares a trace with your signal ground. Common-impedance coupling, hum and buzz for free, in gear that measures beautifully when tested alone.

Muncy did something else in that paper: he drove cable shields with a known current, 100 mA at 60 Hz, 600 Hz, and 6 kHz, and measured what leaked into the signal pair of various cables. He called the result shield-current-induced noise, and found it rose with frequency and depended heavily on cable construction, since it comes from imbalanced magnetic coupling between the shield and the two inner conductors. Jim Brown's follow-up work showed foil-and-drain cables generally behave worse than braided shields here, because the drain wire is closer to one conductor than the other. So even with perfect equipment, shield current isn't entirely free; the cable itself converts some of it into differential noise, and it does so more enthusiastically at buzz frequencies than at hum frequencies.

Fig. 4 - Shield-current-induced noise rises with frequencyAfter Muncy (1995): noise from 100 mA of shield current rises with frequency; cable build matters606006 k20 k\-120\-100\-80\-60\-40foil + drain: asymmetric, couples morebraid: more symmetric, couples lessshield current frequency, Hz (log scale) - test points at 60, 600, 6 kHznoise on signal pair, dBV (illustrative)

Fig. 4: shield-current-induced noise rises with frequency, and foil-with-drain cables couple more of it (after Muncy, 1995).

The industry's answer became a standard: AES48, first ratified in 2005, with Whitlock among the working group chairs. Its core demand is exactly what Muncy prescribed: the shield contact shall have a direct, lowest-possible-impedance connection to the shielding enclosure, preferably to the outside of the chassis, at the point of entry. Shield to chassis. Both ends. Not to the board.

There's even a proper test for it, and it's beautifully crude. The "hummer" described in the wake of Muncy's paper is just a wall transformer and a resistor arranged to push a known 60 Hz current into a connector's shield pin while you listen to the gear's output. Compliant gear stays silent, because the current runs around the chassis and back out. A pin 1 offender sings immediately. Ten dollars of parts, and it will map your entire rack in an afternoon: which boxes you can interconnect carelessly, and which ones need their shields handled like evidence.

You can't redesign someone else's product at half past four on a show day, but you can know the symptom: a single piece of gear that hums or buzzes only when certain cables are attached, in ways that don't respond to the usual fixes, is a pin 1 suspect. The Rane library documents the test and the workarounds. And when you're speccing gear, "AES48 compliant" in the manual is worth more to your noise floor than most of the numbers on the same page. The maddening part of the pin 1 story is that the defect is invisible to every spec: a box with a pin 1 problem has pristine THD, pristine noise, pristine everything, alone on the bench. It only misbehaves in a system, which is where you live and the bench doesn't.

## Three legs and a neutral

Walk to the back of the building and look at what feeds your show, because a lot of "audio" problems are born here.

Most North American event power is 120/208 V three-phase wye. Three hot legs, each 120° apart in phase, plus a neutral from the center point and a ground. Hot to neutral gives you 120 V; hot to hot gives 208 V, which is 120 times the square root of three, not the 240 some rental gear expects. The camlock feeder order (green, white, black, red, blue) and the ritual of ground first on, last off exist because this stuff will kill you, and the licensed electrician tying in is not being fussy.

Fig. 5 - 120/208 V three-phase wyeThree legs 120 degrees apart: 120 V leg-to-neutral, 208 V leg-to-leg (120 x sqrt 3)leg Aleg Bleg C208 V leg-to-legN120 Vleg-to-neutral: 120 Vleg-to-leg: 120 x 1.732 = 208 V(not 240 V - check your rentals)balanced legs: neutral currentmostly cancels......except triplen harmonics,which add. See Fig. 6.

Fig. 5: the wye. Three legs 120 degrees apart, 120 V to neutral, 208 V leg to leg.

Two properties of that system matter to your noise floor. First, balance. The three legs share the neutral, and if the loads are balanced, the return currents largely cancel. Load one leg with the amp racks, leave another nearly empty, and the neutral carries the imbalance. A loaded neutral with real resistance means the "zero volts" reference at your outlets is not zero and not the same everywhere, which is exactly the ground-potential difference that drives loop current through your shields.

Second, and nastier: harmonics. Dimmers, LED fixtures, video walls, and switch-mode supplies don't draw smooth sine-wave current. Their harmonic content includes the triplens (3rd, 9th, 15th), and triplens have a party trick: being zero-sequence, they arrive in phase on all three legs, so instead of canceling in the neutral they add. A wye neutral can carry up to 173% of phase current from this effect even with the legs perfectly balanced in amperes. That's why modern practice sizes neutrals at full or even double phase ampacity for electronics-heavy loads, and it's why a shared service full of LED product can polka-dot your noise floor with buzz even when every meter reading looks fine.

Fig. 6 - Why the neutral hums along: triplens addFundamentals 120 degrees apart cancel in the neutral; third-harmonic currents arrive in phase and add to 3xfundamental (60 Hz): sum = 0neutral current: zero3rd harmonic (180 Hz): in phase on all legsneutral current: 3x per-leg harmonic

Fig. 6: fundamentals cancel in the neutral; triplen harmonics arrive in phase and add.

What we do about it: get audio its own leg at minimum, its own feeder if you can, and its own transformer in a perfect world. Keep lighting and video on the other legs, and understand that a shared neutral means their harmonic trash still shapes your reference, just less of it. Feed the entire audio system, front of house included, from the same distro; the cable you run to FOH for power costs less than the afternoon you'll spend chasing the loop created by the "convenient" outlet at the mix position, which is on a different panel, in a different century of the building's wiring.

Installed venues have their own vocabulary for the same idea. Those orange "isolated ground" receptacles run their ground pin on a dedicated insulated conductor all the way back to the panel, instead of sharing conduit and box grounds with whatever else lives on the wall; the bond still happens where the code says it must, but the path your gear references is cleaner. Studios and broadcast plants go further with technical power: dedicated transformers for the audio systems, sometimes symmetric 60-0-60 V "balanced power" that cancels a chunk of leakage-current hum by the same common-mode logic as a balanced audio line. You don't need to build any of this on a Tuesday club gig. You do need to know it exists, so that when the install you're guesting in has a wall of orange outlets, you use those and not the janitor's receptacle behind the curtain.

And generators: when the show runs off a genny, the neutral-to-ground bond and the grounding electrode are the electrician's problem and the law's interest, not a place for creativity. A generator is its own separately derived system; whether and where it bonds neutral to ground changes how fault current gets home. If the power distro person did their job, your noise floor benefits. If you're tempted to do their job with a screwdriver, put it down.

## What actually fixes it

Now the triage, in the order that respects both physics and your schedule. We'll be blunt about what doesn't work first: power conditioners with impressive faceplates do nothing for ground loops, because the loop isn't riding on the hot and neutral they filter. Ferrite beads help with radio, not with 60 Hz. And the ancient move of scratching around for "a better ground," a water pipe, a beam, a rod driven behind the club, is superstition; the loop doesn't care how good your ground is, it cares that there are two of them with a cable in between.

Start at the power, because one fix there beats twenty fixes downstream. Everything audio on one service, one distro, one leg where practical. This shrinks ground-potential differences across your whole system at once. Then walk the noise: mute the master, bring up sections, find the boundary where the hum enters. It's almost always a boundary between systems: your rig to the broadcast truck, the DJ's controller to your console, the guitarist's amp to your split.

At those boundaries, unbalanced consumer gear is the standing offender, and the tool is a transformer. A DI box for instruments; a 1:1 line-level isolation transformer for playback rigs, laptops, and video feeds. The transformer passes signal magnetically and breaks the metallic path, so loop current has no circuit. That little switch labeled "ground lift" on the DI disconnects the shield continuity between the two sides; flip it and listen. Half the hum in bar gigs dies right there, and the guitarist's amp buzz usually goes with it. Every quality hum-stopper box you've ever seen is just this: a decent transformer in a can. Buy the decent one; the ten-dollar version earns its price in distorted low end, because cheap iron saturates exactly where your kick drum lives.

A worked example, because this one plays out weekly somewhere. The DJ's laptop feeds your console through a headphone-jack-to-RCA cable, and the moment their charger goes in, the system hums. The charger's switching supply leaks a little current to its ground; the laptop's "ground" now sits at a different potential than yours; the unbalanced cable's shield is the only path between them, and it's also the signal return. Every term from our Ohm's law story, assembled in one convenient package. Battery power hides it, which is why it "worked fine at home." The fix is not a better cable and it's not asking the DJ to perform unplugged. It's a stereo isolation transformer between their output and your input, thirty seconds to patch, and the charger can stay in.

For balanced lines that still hum, the polite fix is lifting the shield at one end, traditionally the receive end, so the shield still drains interference but no longer completes the loop. Telescoping shields, in the old Rane Note 110 language. Modern AES48 thinking prefers shields landed at both ends on compliant gear, but in the field, with a mystery rack of vintage and misc, a one-end lift is a legitimate tool. Make yourself two or three lift adapters, label them in a color you can see in the dark, and count them out at load-out, because a lift cable that sneaks into general stock becomes next month's "why does this line pick up radio" ticket.

And now the thing we say slowly, every time, because someone new is always reading: never lift the AC safety ground. Not with a grey three-to-two adapter, not with tape over the pin, not "just for the gig." The third pin is what turns a chassis fault into a tripped breaker instead of a musician completing the circuit through a microphone and their lips. Rod Elliott's earthing article walks through loop-breaker circuits that reduce hum inside DIY gear while keeping the safety bond intact, and his position is the industry's position: hum is an inconvenience, electrocution is permanent. Every legitimate fix in this piece breaks the loop in the signal wiring. None of them touch the power cord.

## What you can take back to the desk

Some of this is checklist, so treat it like one.

Learn the pitches. Sixty is coupling or ground current, 120 is a power supply, buzz is chopped current from dimmers or switchers. Diagnosis by ear, before you move a single cable.

Power the whole audio system from one source, and run power to FOH from your own distro. Balance your legs, keep lighting and LED off audio's leg, and be suspicious of any outlet you didn't put there.

Carry transformers. Two good DIs and one stereo line-level isolator will solve more shows than any plugin you own. Use the ground lift on the DI freely; it lifts the shield, not the safety.

On stubborn balanced lines, lift the shield at one end. On stubborn single boxes, suspect pin 1 and check whether the gear follows AES48.

Respect the neutral, respect the third pin, and put the margarita machine on lighting's leg. They earned it.

If you want homework, it's cheap: build the hummer, an afternoon and pocket change, and test your own rack before a show day does it for you. Read the June 1995 AES Journal pieces, or at least the Rane notes that translate them into field English. And next load-in, spend ninety seconds at the distro asking which leg you're on and what else is on it. The electrician will tell you. They're usually delighted anyone asked.

And one habit for the long game: keep a noise log for rooms you work regularly. Which outlet at FOH is clean, which leg the LED wall lives on, which house line has the lifted shield. Half of hum-chasing in a familiar room is re-discovering what you already fixed last time. Write it down once and the 4:50 drone becomes a two-minute fix instead of a ritual.

The bigger thing to take away is the mental model. Hum is a current, flowing through a resistance, landing in your signal path. Every mystery buzz you'll ever chase is those three terms wearing a costume. Find the current, break its path or get it out of your audio, and go get some catering before the good stuff is gone.

*Above the Fader*

## Sauce

- [Rane Note 110: Sound System Interconnection](https://www.ranecommercial.com/legacy/note110.html?ref=abovethefader.com)
- [Rane Note 151: Grounding and Shielding Audio Devices](https://www.ranecommercial.com/legacy/note151.html?ref=abovethefader.com)
- [Jim Brown, SCIN: Shield Current Induced Noise (Rane Note 166)](https://www.ranecommercial.com/legacy/note166.html?ref=abovethefader.com)
- [Bill Whitlock, Design of High-Performance Balanced Audio Interfaces](https://sound-au.com/articles/balanced-interfaces.pdf?ref=abovethefader.com)
- [Bill Whitlock / Jensen Transformers, Understanding, Finding & Eliminating Ground Loops](https://www.jensen-transformers.com/wp-content/uploads/2014/08/generic-seminar.pdf?ref=abovethefader.com)
- [Sound on Sound, Ground Loops Explained](https://www.soundonsound.com/techniques/ground-loops-explained?ref=abovethefader.com)
- [Rod Elliott (ESP), Earthing (Grounding) Your Hi-Fi](https://www.sound-au.com/earthing.htm?ref=abovethefader.com)
- [AES48 standard, grounding and EMC practices (revision notice)](http://www.aes.org/standards/blog/2019/1/call-for-comment-on-revision-of-aes48?ref=abovethefader.com)
- [ProSoundWeb, The Pin 1 Problem Revisited](https://www.prosoundweb.com/the-pin-1-problem-revisited/?ref=abovethefader.com)
- [eGauge, Three-phase 120/208 wye systems and triplen harmonics](https://kb.egauge.net/en%5FUS/three-phase-120208-or-277480-wye?ref=abovethefader.com)

*Featured photo by [Samuel Ramos](https://unsplash.com/@idgeek?ref=abovethefader.com) on [Unsplash](https://unsplash.com/photos/mixing-console-KVhEugDlAYM?ref=abovethefader.com).*