Lighting Automation

Introduction

A moving light is a lamp, a set of motors and a small computer, hanging upside down twenty feet in the air, being told what to do by a cable that has been daisy-chained through nine other fixtures before it reaches this one. Automation is the craft of making that arrangement reliable.

Most of the trouble is not in the console. It is in the cable, the termination, the addressing, the firmware and the filters that nobody has cleaned since last season. This book is about all of that.

What this book covers

What a moving light actually does while it moves. The optical path from source to lens, and why two fixtures with the same lamp look different. Colour wheels against CMY mixing. Gobos, prisms and orientation. DMX as a physical thing -- cable, connectors, topology, and the termination that causes most intermittent faults. What is actually on the wire, and why a universe is 512 slots. Wireless DMX. RDM, and addressing a fixture from the ground instead of from a lift. Ethernet, switches, IP addressing, Art-Net and sACN. Patching, personalities and building a rig on a console. Pan and tilt and the long way round. The pre-show check. Why fixtures fail -- heat, shock and age. Lamp replacement, alignment and calibration. Preventive maintenance. Firmware. And automation outside the show, in buildings that run on a time clock.

It was measured against Richard Cadena's Automated Lighting, chapter by chapter from the published contents of the third edition, so that nothing that book teaches a working automation technician is missing here.

Who this is for

Lighting technicians who can hang and patch a rig and want to understand what is happening inside it, board operators who are handed faults they cannot diagnose, and anybody who has been told "it's a DMX problem" and wanted to know what that actually means.

This Book Does Not Replace the Manual

Every make and model differs. Lamp alignment on one fixture is nothing like lamp alignment on another, service intervals vary, and the procedure that matters is the one in the document that came with the unit.

What this book gives you is the understanding to read that manual and know what it is talking about -- and to know when something is wrong before the manual is any help at all.

How to Use This Book

Read it in order. The chapters on DMX and the network are the two that most often surprise people who thought they understood the subject, and everything about diagnosis depends on them.

Keep a notebook

Write down the address of every fixture and which universe it is on, and keep it with the plot. A rig you cannot address from paper is a rig you address from a lift.

Write down every fault and what fixed it. Automation faults repeat -- the same fixture, the same position, the same cable run -- and the pattern is only visible if somebody recorded the instances.

Write down firmware versions when you update them. A rig on mixed firmware behaves differently fixture to fixture, and it is the last thing anybody thinks to check.

A note on the words

A universe is 512 slots of DMX. A channel is one of those slots, and it is also, confusingly, what a console calls a fixture. A fixture's address is where its first slot sits. A personality or profile is the file that tells the console what this model's slots mean.

Art-Net and sACN both carry DMX over Ethernet and are not the same thing. RDM is not a separate cable -- it rides on the DMX one. And "it's a DMX problem" is what people say when they mean "I do not know what is wrong".

A note on safety

Moving lights are heavy, hot and hanging above people. Every one needs a clamp and a safety bond, and the bond is not optional because the clamp looks fine. Fixtures are also mains-powered devices you will open, so they are isolated and given time to discharge before a cover comes off.

Two hazards are specific to this work. Discharge lamps run under pressure and can shatter -- eye protection and a cool-down period before a lamp change are not caution, they are the procedure. And a fixture that homes unexpectedly while your hands are in it will move with real force.

What a Moving Light Is

Motors, a source and a small computer, hanging upside down. Everything else in this book follows from taking that description literally.

The three machines in one housing

A moving light is three things bolted together. A light source with an optical path in front of it. A set of motors moving that path, or moving the whole head. And a controller listening to a data line and deciding what the motors should do.

Almost every fault you will meet belongs to exactly one of those three, and knowing which one narrows the problem before you have touched anything.

Pan, tilt and the yoke

Pan is rotation about the vertical axis, tilt about the horizontal. A typical fixture pans somewhere between 460 and 630 degrees and tilts around 270. That over-rotation is deliberate -- it lets a fixture reach the same point by two different routes.

Which matters because a fixture crossing its own zero point takes the long way round, visibly, in the middle of a show. A light that swings across the stage when you asked it to move six inches has not malfunctioned; it has run out of pan in one direction and gone the other way.

Homing

On power-up a fixture homes: it drives every motor to a known stop so it knows where everything is. A unit that does not complete homing does not know where its colour wheel or its shutter is, and everything it does afterwards is wrong in a way that looks like a control fault.

Watch the homing sequence. It is the cheapest diagnostic in the trade -- a fixture that hesitates, grinds or stops part-way is telling you something before the show does.

Head, yoke and base

The base holds the power supply, the data connections and usually the main board. The yoke carries the tilt motor and the cabling between base and head. The head carries the source, the optics and the effects.

Cable inside the yoke flexes every time the fixture moves, which is why intermittent faults that change when a light pans are almost always a yoke loom rather than anything you can see.

Three machines, three families of fault

How to do it

1. Power it and watch the whole homing sequence without touching it.
2. Note anything that hesitates, grinds or stops short.
3. Check the display for an error log before you assume anything.
4. Identify which of the three machines the symptom belongs to.
5. Only then open the manual, at the section for that machine.

What goes wrong

Practice it

One fixture, on a bench or hanging, that you can power on and off.

Inside the Fixture

Between the source and the lens is an optical path, and every part of it is doing something to the light. This is why two fixtures with the same lamp look nothing like each other.

The source

Three families. Tungsten halogen: warm, dimmable electrically, short-lived, enormously hot. Discharge or arc: very bright, fixed colour temperature, cannot be dimmed electrically at all. LED: efficient, cool, long-lived, dimmable to zero, and the colour is built rather than filtered.

Discharge lamps run under pressure and are the reason a fixture has a cool-down period. They also have a restrike delay -- switch one off and it will not light again until it has cooled, which on a show means a fixture that is out for several minutes.

Reflector and the optical path

Behind the source sits a reflector, and its geometry decides what the fixture is. An elliptical reflector brings light to a focus, which is what a profile needs. A spherical one returns light through the source. The material and coating decide how much survives the trip.

In front, an infrared filter removes heat from the beam so the gobos and gels downstream survive. Take one out and everything after it starts to cook.

Dichroic filters, and why they are not gel

A dichroic is an optical thin-film filter: layers of coating that reflect the wavelengths you do not want rather than absorbing them. That is why it survives in a beam that would destroy a gel, and why the colour changes slightly with the angle you view it from.

They chip, they delaminate with heat and age, and a chipped dichroic shows as a coloured edge or a spot on the beam that no amount of focusing removes.

Mechanical dimming

A discharge source cannot be dimmed electrically, so the fixture dims mechanically -- a douser, an iris or a set of blades closing across the beam.

This matters practically: mechanical dimming is not linear, it is not silent, and at the bottom of its travel it often is not smooth either. A slow fade to black on a discharge fixture that steps or shudders is a mechanism, not a console.

The three sources, compared

How to do it

1. Look at the beam on a white surface, not on the stage.
2. Decide whether the problem is brightness, colour, shape or edge.
3. Brightness — source, reflector, or dirty optics.
4. Colour — a wheel out of position, a chipped dichroic, or mixed firmware.
5. Shape or edge — gobo, prism, focus, or a lens that has moved.
6. Compare against a known-good fixture of the same model before opening anything.

What goes wrong

Practice it

One fixture and a white wall.

Colour, Gobos and the Beam

Everything a fixture does to the beam after the source: colour, shape, texture and edge. These are the parameters a designer asks for by name, and the mechanisms that produce them fail in recognisable ways.

Colour wheels against colour mixing

A colour wheel is a disc of dichroic filters that rotates into the beam. It gives you a fixed set of saturated colours, it changes by stepping between them, and it can be parked between two to split the beam -- which is sometimes useful and usually a mistake.

Colour mixing puts three or four graduated dichroics in the beam and varies how far each intrudes. Subtractive mixing -- cyan, magenta, yellow -- removes wavelengths from white, which is how most discharge fixtures work. Additive mixing combines coloured emitters, which is how LED fixtures work.

The practical difference: subtractive mixing loses output as it saturates, and a deep colour on a CMY fixture is much dimmer than open white. Additive mixing does not lose output the same way, but can only make colours its emitters can reach.

Gobos

A gobo is a pattern in the beam. Metal gobos are cut steel -- cheap, durable, monochrome, and limited to shapes that hold together physically. Glass gobos are printed or laser-ablated on a coated substrate, can be full colour and far finer, and are fragile and expensive.

Orientation matters and is the commonest installation error. A gobo goes in a specific way round, and the coated side of a glass gobo faces the direction its maker specifies. In backwards, any gobo projects a mirror image, and a glass gobo with its coating the wrong way round will also cook and delaminate.

Rotating gobo wheels add indexing: the ability to hold a gobo at a known angle rather than merely spinning it. A designer asking for a window at a particular angle is asking for indexing, and a fixture that cannot index cannot do it.

Prisms, frost and iris

• Prism: splits the beam into several copies. Usually rotatable. Reads as an effect, not as a texture.
• Frost: softens the edge. The difference between a hard-edged profile and a wash from the same unit.
• Iris: shrinks the beam from the outside, keeping it centred. Not the same as zoom, which changes the angle.
• Zoom and focus: zoom changes the beam angle; focus sharpens the gobo or the edge. Two different motors, often confused.

Shutters and framing

Framing shutters cut the beam with blades, usually four, sometimes on a rotating module. This is how light is kept off a screen, a wall or a proscenium arch, and on a moving light it can be done remotely -- which is the whole reason automated framing exists.

Shutters are mechanical and they drift. A framing cut that was clean at focus and is spilling by the third week is a mechanism wearing, not an operator error.

The beam mechanisms, and what fails

How to do it

1. Open white, full, on a flat surface.
2. Step every colour on the wheel, and run CMY from zero to full.
3. Step every gobo, then index one through a full rotation.
4. Run zoom, focus, iris and frost to both extremes.
5. Run the shutters in and out, one at a time.
6. Note anything that steps, sticks, hesitates or lands off-centre.

What goes wrong

Practice it

One moving light with colour, gobo and framing.

DMX, Physically

Before the protocol, the wire. Most of what people call a DMX problem is a cable, a connector, a topology or a missing terminator -- and all four are visible to somebody who knows what to look at.

What DMX actually is

A differential serial data signal on a twisted pair, sent one way from a controller to a chain of receivers. Differential means the signal is the difference between two conductors, which is why it survives long runs and electrical noise -- and why the cable has to be an actual data cable.

Microphone cable is not data cable. It has the wrong impedance, and a rig built on it works right up until it does not, usually at the far end of the longest run, usually on a show.

Connectors and the five-pin convention

DMX is specified on five-pin XLR. Three-pin is common on cheaper fixtures and is the same signal on fewer pins, but the reason the standard chose five is precisely so a DMX line cannot be plugged into an audio one.

Adapters exist and are normal. What is not normal is a rig where nobody knows which cables are data and which are audio, because they look identical in a case at two in the morning.

Topology — a chain, not a star

DMX is a daisy chain. Out of the controller, into fixture one, out of fixture one, into fixture two, and so on. It is not a star and it is not a tree. Splitting a line by plugging two cables into one output with a Y-adapter is how rigs become unreliable.

When you need branches, you use an opto-splitter, which regenerates the signal on each output and isolates the branches electrically. That isolation is doing real work -- it is also what stops a fault on one branch taking out the others.

The practical limits: thirty-two devices on one chain, and roughly a thousand feet of cable, before the signal needs regenerating. Both numbers are routinely exceeded and the rig routinely misbehaves.

Termination

The last fixture in a chain needs a terminator -- a resistor across the data pair. Without it the signal reflects off the open end of the cable and comes back up the line, colliding with what is still arriving.

The symptom is the most characteristic in the trade: fixtures flickering, jittering, or responding to commands nobody sent, intermittently, worse on longer runs. Many fixtures now self-terminate, which helps until somebody adds one more unit past the one that was terminating.

The physical layer, and what each part does

How to do it

1. Ask what changed. A rig that worked yesterday has had something done to it.
2. Check the terminator is at the actual end of the chain, not where it used to be.
3. Count the devices on the chain and the total cable length.
4. Look for Y-splits and replace them with a splitter.
5. Swap the suspect cable for a known-good one before suspecting a fixture.
6. Only now look at addressing, personalities and the console.

What goes wrong

Practice it

A short chain of fixtures you can rewire.

What Is On the Wire

The protocol itself. Why a universe holds 512 slots, what a packet actually contains, and what wireless DMX changes.

Slots, universes and addresses

A DMX packet is a start code followed by up to 512 slots, each one a number from 0 to 255. That is the whole protocol. One slot is one parameter at one level.

512 slots is a universe. A fixture's address is where its first slot sits, and it reads as many consecutive slots as its mode needs. A fixture in a 24-channel mode addressed at 100 occupies 100 to 123, and the next fixture starts at 124 or later.

Overlap two fixtures and both respond to the same commands. That is not a fault, it is arithmetic, and it is the second thing to check after termination.

DMX512-A, and what it added

The current standard. It kept everything that worked and added defined alternate start codes, which is what made RDM possible -- a way for other kinds of message to share the same wire without disturbing fixtures that are not listening for them.

It also tightened the electrical specification, which is the part that makes correct cable and termination matter rather than being advice.

Refresh rate, and why it matters

A controller sends the whole universe repeatedly, up to about 44 times a second. That rate decides how smooth a fade looks and how responsive a fixture feels.

A heavily loaded network, a badly configured node, or a long chain can slow it. The symptom is steppy fades and a rig that feels slightly behind the desk -- which people attribute to the console and almost never to the data path.

Wireless DMX

A transmitter takes the universe and sends it over radio to receivers, usually in the 2.4 GHz band. It is ordinary and it works, and it fails in ways cable does not.

Interference is the obvious one -- the same band as wifi and a great deal else, in a venue full of both. Pairing is the other: a receiver paired to last week's transmitter will sit there doing nothing while everybody checks the cable that is not there.

Wireless is a convenience for places cable cannot reach. It is not a substitute for a properly built wired rig, and on a show that matters it is the first thing to remove from the critical path.

The protocol, in numbers worth remembering

How to do it

1. Get the fixture's mode and its channel count from the manual, not from memory.
2. List every fixture on the universe with its start address.
3. Work out each one's last slot from its channel count.
4. Look for any overlap, and for anything running past 512.
5. Re-address with gaps if the rig is likely to change.

What goes wrong

Practice it

A console and three fixtures of the same model.

RDM

RDM lets the console talk back to a fixture: find it, address it, read its sensors and its error log — from the ground rather than from a lift. It is the biggest practical change in this trade in twenty years.

What RDM is, and what it is not

Remote Device Management rides on the same physical layer as DMX: the same cable, connectors, splitters and fixtures. It is not a second network. It uses the alternate start codes DMX512-A defined, so fixtures that are not listening for RDM simply ignore it.

What it is not is universal. Older fixtures do not speak it, and some splitters block it because they are unidirectional -- and a unidirectional splitter is the commonest reason RDM "does not work" on a rig where every fixture supports it.

Discovery

The controller asks who is out there, and fixtures answer with their unique identifiers. The result is a list of everything on the line -- make, model, current address, mode.

That list is worth having for its own sake. It is the fastest audit of a rig anybody has ever had: what is actually hanging there, as opposed to what the paperwork says is hanging there.

Get and set

Get reads a value from a fixture: its address, its mode, its lamp hours, its internal temperature, its error log. Set writes one: change the address, change the mode, reset the lamp hours after a change.

Re-addressing a rig from the desk is the headline, and it is genuinely transformative on a rig at height. Reading temperature and lamp hours across a whole rig before a show is the quieter one, and arguably more valuable -- it tells you which fixture is about to fail.

Practical RDM

Discovery takes time proportional to the rig, and it interrupts nothing but is best done when nothing is running. Some fixtures respond slowly or incompletely, especially on long chains.

And a warning worth taking seriously: setting a parameter on the wrong fixture is as easy as setting it on the right one. Confirm the unique identifier before you change anything, particularly mode, which can silently change how many channels a fixture occupies and shift everything after it.

What RDM gives you, in order of practical value

How to do it

1. Confirm every splitter in the path is RDM-capable, not unidirectional.
2. Run discovery with nothing else happening on the line.
3. Compare the discovered list against the paperwork and resolve any differences.
4. Read lamp hours and temperature across the rig and note the outliers.
5. Address from the desk, confirming the unique identifier each time.
6. Re-run discovery afterwards to confirm the rig is as you intended.

What goes wrong

Practice it

An RDM-capable console and a small rig.

The Network

A modern rig runs DMX over Ethernet, which means the lighting department is now running a network. This chapter is that network, in the terms a lighting technician actually needs.

Why Ethernet at all

One DMX line carries 512 slots. A rig of moving lights in 30-channel modes runs out at seventeen fixtures. Ethernet carries many universes down one cable to nodes that fan them out to DMX where the fixtures are.

That is the whole reason. Everything that follows is the cost of it.

Cable, switches and distance

Cat5e or better, and the hundred-metre limit is real -- it is not a suggestion and it does not degrade gracefully. Beyond that, fiber or another switch.

Switches: a managed switch can be configured and can prioritise traffic; an unmanaged one cannot. On a small rig unmanaged is fine. On a large one, or one sharing a network with anything else, managed is the difference between a reliable show and an unexplainable one.

IP addressing, subnets and DHCP

Every device needs an address and they all have to be on the same subnet to talk. A subnet mask decides which part of the address is the network and which is the device.

Art-Net conventionally lives on 2.x.x.x with a 255.0.0.0 mask. That is a convention, not a law, and a rig where half the devices followed it and half did not is a rig that half works.

DHCP hands out addresses automatically. It is convenient and it is a liability on a show: a device that gets a different address after a power cycle is a device the console can no longer find. Static addressing, written down, is the professional default.

Routable against non-routable matters mainly because show networks should not route anywhere. A lighting network with a path to the internet is a lighting network somebody's phone will find.

Art-Net and sACN

Both carry DMX universes over Ethernet, and they are not the same. Art-Net is older. Its early versions broadcast every universe to every device, and gear still set to broadcast floods a network if you let it; current Art-Net (Art-Net 4) sends each universe only to the nodes that ask for it. sACN — streaming ACN, the E1.31 standard — multicasts, so switches only send each universe where it is wanted, and it has priority built in so two sources on one universe resolve predictably.

On a small rig either works. On a large one sACN is usually the better behaved, and the reason is entirely about what the switches have to do with the traffic.

The failure mode people meet first: a broadcast Art-Net rig on a network shared with anything else, saturating it. The symptom looks like the lighting failing. The cause is the lighting flooding everything.

Art-Net against sACN

How to do it

1. Decide the protocol and use one, not both, unless something forces it.
2. Address everything statically and write the addresses down.
3. Keep every device on the same subnet, with the same mask.
4. Keep the lighting network off any network that routes anywhere.
5. On anything large or shared, use a managed switch and prefer sACN.
6. Test with the full rig running, not one fixture.

What goes wrong

Practice it

A console, a node and a switch.

Patching and Building a Rig

Where a pile of fixtures becomes something a console can control. Patch is unglamorous, it is done at the start of a long day, and it is where a large share of show-day faults are created.

Personalities and modes

A personality — or profile, or fixture library entry — tells the console what each of a fixture's slots means. The wrong personality gives you a rig where colour is on the channel the console thinks is gobo, and every symptom looks like madness.

Most fixtures have several modes with different channel counts: a basic mode, an extended one with 16-bit movement, sometimes a compact one for saving universe space. The fixture's mode and the console's personality have to be the same mode. This is the single commonest patch error in the trade.

Building the patch

Address in an order that makes sense physically — left to right, upstage to downstage, or by truss. A patch numbered in the order you happened to pick fixtures up is a patch nobody can work with at three in the morning.

Leave gaps. A rig always changes, and a patch with no room forces a re-address of everything downstream when one fixture is added.

Sixteen-bit, and why movement needs it

One slot gives 256 steps. Across a fixture's full pan that is visible stepping on a slow move. So movement uses two slots — coarse and fine — giving 65,536 steps, which is smooth.

If pan and tilt step visibly on slow moves, the fixture is probably in an 8-bit mode, or the personality is. It is not the motors.

Checking the patch before anything else

Bring up each fixture alone, in order, and confirm the right unit responds and that its parameters do what they say. Colour should change colour. Gobo should change gobo.

This takes fifteen minutes on a modest rig and it finds the wrong-personality, wrong-mode and overlapping-address faults while there is still time to fix them.

The patch errors, and what each looks like

How to do it

1. Confirm each fixture's mode at the fixture, not from the plot.
2. Select the matching personality and mode in the console.
3. Address physically — by truss, left to right — and leave gaps.
4. Write the patch down and keep it with the plot.
5. Bring up every fixture alone and confirm it and its parameters.
6. Note anything that does not match, before focus starts.

What goes wrong

Practice it

A console and several fixtures of at least two models.

Running the Rig

The show is loaded and the rig is patched. This chapter is the routine that finds the dead fixture before an audience does, and what to do when one goes wrong with people watching.

The pre-show check

Every fixture, every show, in the same order. Bring the rig to a known state, run every unit to full white, and look at the rig rather than at the console.

Then move everything: pan and tilt to both extremes, every colour, every gobo, focus and zoom through their travel. A fixture that cannot reach one end of its pan is a fixture that will be wrong in exactly one cue, and you want to know now.

Read the error logs while you are there. RDM makes this a two-minute job on a whole rig rather than a climb per fixture.

What to check that is not a fixture

• Terminators, still in the last fixture of each chain after yesterday's changes.
• Node addressing, unchanged since the power cycle.
• Firmware consistency across identical fixtures.
• Spare cables and a spare fixture, actually present rather than theoretically available.

When one fixture is wrong mid-show

This is triage, not repair. Nobody fixes a moving light during a show. The decisions are: can the cue survive without it, can another fixture cover it, or does it need parking somewhere it does no harm.

Park is the underused answer. A fixture stuck in a wrong colour in the middle of a look is more damaging than the same fixture blacked out, because the audience reads a mistake rather than an absence.

Note it, and fix it at the next break. A show is not the time, and a technician on a ladder mid-act is a bigger problem than the light.

Handing over

If somebody else runs the show, they need the patch, the known faults, the spare locations and the one instruction that matters: what to do if a specific fixture misbehaves.

A rig handed over without its list of known quirks is a rig where somebody spends an interval diagnosing a fault you already knew about.

The pre-show sequence

How to do it

1. Decide in seconds whether the audience can see the problem.
2. If not, note it and carry on.
3. If yes, ask whether another fixture can cover the moment.
4. If nothing covers it, park the fixture dark rather than leave it wrong.
5. Write down what happened and at which cue.
6. Fix it at the interval or after the show, never during.

What goes wrong

Practice it

Any rig you can run for twenty minutes.

Why Fixtures Fail

Heat, shock and age, in that order. Almost every failure in this trade is one of the three, and knowing which one you are looking at tells you whether it is a clean, a repair or a replacement.

Heat

The largest cause by a distance. A fixture makes a great deal of heat and lives or dies by getting rid of it. Blocked filters, dirty fans and clogged optics all raise internal temperature, and everything electronic ages faster hot.

The symptoms arrive in an order: the fixture runs hotter, then it logs over-temperature warnings, then it thermally shuts down mid-show, then something fails permanently. Every one of those stages is a chance to catch it, and the first two are free if anybody is reading the logs.

Shock

Fixtures are transported in trucks, dropped onto decks, and knocked on trusses. Shock breaks lamps, shifts optics, loses gobo indexing and cracks solder joints on boards.

Shock damage is characteristically intermittent, because a cracked joint makes contact until it is warm or until the fixture moves. An intermittent fault on a unit that has toured is a mechanical suspicion before it is an electronic one.

Age

Lamps lose output long before they fail — a discharge lamp at the end of its rated life can be substantially dimmer and shifted in colour while still working perfectly. A rig where three fixtures look wrong next to the others is often three lamps at different ages.

Belts stretch, bearings wear, filters delaminate and gaskets harden. None of these announce themselves. All of them are found by inspection, which is the argument for preventive maintenance.

Reading the error log

Codes are manufacturer-specific and the manual is the only authority. But the pattern is general: temperature warnings, motor or home errors, lamp strike failures and communication errors are the four families, and each points at a different one of the three machines.

The three causes, and how each presents

How to do it

1. Read the error log and note the family of fault.
2. Check internal temperature against a working unit of the same model.
3. If temperature is high, clean first and re-test — many faults end here.
4. If the fault is intermittent and movement-related, tag and swap.
5. If output is low, check lamp hours before suspecting anything else.
6. Record what you did, so the next instance shows a pattern.

What goes wrong

Practice it

A rig with RDM, or several fixtures you can interrogate.

Maintenance

The work that decides whether a rig is reliable. It is unglamorous, nobody sees it, and it is the difference between a rig that works and one that is being re-lamped at four in the afternoon.

Working safely inside a fixture

Isolate it and give it time. Power supplies hold charge after disconnection, and a fixture that has just been running is hot enough to burn through a glove.

Discharge lamps run under pressure and can shatter. Eye protection and a cool-down period before a lamp change are the procedure, not extra caution. Do not touch the glass of a new lamp — skin oil creates a hot spot and shortens its life dramatically.

And isolate before your hands go in, because a fixture that receives a home command while you are inside it will move with real force.

Cleaning

Filters and fans first, because they cause the heat that causes everything else. Optics second, with the correct materials — a dichroic or a front-surface mirror is not glass and is damaged by ordinary glass cleaner and by wiping dry.

A front-surface mirror has its reflective coating on the outside, where it can be scratched by exactly the cloth you would reach for. When in doubt, blow rather than wipe.

Lamps, alignment and calibration

Lamp replacement is the commonest service task. Afterwards two things must happen: alignment, so the lamp sits at the reflector's focus, and resetting the lamp hours, so the record means something.

Misalignment shows as a hot spot off-centre, uneven field, or simply less output than the fixture beside it. It is adjusted with the fixture running, which is why it is done carefully and with eye protection.

Calibration — of pan, tilt and the effects wheels — is what keeps identical fixtures actually identical. A rig where one unit's gobo is a few degrees off is a rig that has not been calibrated since it was knocked.

Firmware, spares and records

Update firmware as a set, not a unit. Mixed firmware across identical fixtures gives you a rig where two units respond differently to the same command, and it is the last thing anybody checks.

Carry spares that match: lamps, filters, a fuse set, and where budget allows a whole spare fixture. And keep records — hours, cleans, faults, firmware — because maintenance without a record is just cleaning.

A maintenance interval that works

How to do it

1. Isolate it from power and from data, and let it cool.
2. Wait for supplies to discharge before removing a cover.
3. Clean filters and fans first, then optics with correct materials.
4. Replace the lamp without touching its glass; align it running, with eye protection.
5. Reset lamp hours and note the date.
6. Update firmware to match the rest of the set, then re-calibrate.
7. Record everything you did.

What goes wrong

Practice it

One fixture you are permitted to open, and its manual.

Automation Beyond the Show

Lighting is automated in buildings too, on a different timescale and for different reasons. The protocols overlap, the mindset does not.

What installed automation is for

A show rig is built, run and struck. An installed system runs every day for years with nobody watching it, and its job is to be invisible: the right light at the right time, adjusting to daylight and occupancy without anybody touching a control.

That changes every priority. Reliability over flexibility, maintainability over capability, and a strong preference for anything that fails into a usable state rather than into darkness.

Scenes, presets and time

An installed system runs scenes: stored states recalled by a button, a schedule or a sensor. A meeting room has presentation, meeting and cleaning; a foyer has day, evening and event.

Astronomical time clocks are the characteristic tool — the system knows sunrise and sunset for its own location and date, so exterior and daylight-linked lighting follows the actual sky rather than a fixed clock that drifts across the year.

Sensors

Occupancy and daylight sensors do most of the automation in a building. Occupancy is the one people notice, usually because it has turned the lights off on somebody sitting still.

Tuning them is the work: sensitivity, coverage and time delay. Every one of those is a compromise between energy saved and somebody sitting in the dark, and the complaint always arrives long after the commissioning engineer has left.

DALI, and the protocol the installed world actually uses

Everything so far in this book has been DMX and the networks that carry it. Walk into a building instead of a venue and the control system is very often DALI -- the Digital Addressable Lighting Interface, standardised internationally as IEC 62386.

It is a different shape of system. A DMX universe broadcasts 512 slots and every fixture listens for the ones at its address. A DALI bus is a two-wire loop where each piece of control gear has a short address and is spoken to individually, and it can answer back as a matter of course rather than as an extra like RDM.

The standard is published in parts, restructured in 2014 into three families. The 1xx parts are general -- Part 101 defines the system components, Part 102 the control gear at the driver end, Part 103 the control devices, meaning input devices and the application controllers that make the decisions. The 2xx parts cover particular kinds of control gear and the 3xx parts particular kinds of input device.

The practical translation is short. In a venue you patch a fixture to an address on a universe. In a building you commission a driver to a short address on a bus, and the addressing is done once, by somebody who then leaves.

0-10V, phase dimming and the installations you will still meet

Not every building is DALI. A great many installations run on 0-10V analogue control -- one pair of wires per channel carrying a voltage that means a level -- or on mains phase dimming, where the dimming happens by chopping the waveform.

Both are simple, both are old, and both are still being installed because they are cheap. What they are not is addressable: a 0-10V channel controls whatever is wired to it, so changing the grouping means changing the wiring. That is the reason DALI exists.

Phase dimming brings one specific trap worth knowing. An LED driver on a phase dimmer that was designed for incandescent load will frequently flicker, buzz, or refuse to dim below some level, and it is a compatibility problem rather than a fault in either part.

Input devices, and lighting that reacts to the building

A show rig does what it is told. An installed rig reacts, and the things it reacts to are standardised too -- the 3xx parts of IEC 62386 cover push buttons in Part 301, absolute devices like switches and sliders in Part 302, occupancy sensors in Part 303 and light sensors in Part 304.

That list is worth reading as a map of where installed lighting goes wrong, because almost every complaint about a building's lighting is about one of those four. The button that does not do what its label says. The slider nobody can find. The occupancy sensor that has turned the lights off on somebody sitting still. The daylight sensor dimming a room that is already dim because a cloud crossed.

None of those is a fixture fault, and none of them is diagnosable by looking at a light. They are commissioning decisions -- of sensitivity, coverage, delay and grouping -- made once and then lived with for a decade.

Why it is maintained differently

Nobody does a pre-show check on a building. Faults are found by a monitoring system, or by a complaint. Fixtures are on ceilings that need access equipment and a permit, so a lamp change is scheduled rather than immediate.

The practical consequence: installed systems are designed for long service intervals and remote diagnosis, and the documentation matters far more than on a show, because the person maintaining it in five years will never have met whoever commissioned it.

Who owns it, and who gets called

On a show the chain is short and everybody is in the building. On an installation it is long and mostly absent. There is a designer who specified it, a contractor who installed it, a commissioning engineer who addressed and programmed it, a facilities team who now own it, and a tenant who complains about it.

Almost none of those people talk to each other, and the commissioning engineer has usually gone. Which means the first question on any installed fault is not technical -- it is who is actually responsible for this system, and does anybody hold its documentation.

If the answer is nobody, that is the finding. An installation with no owner and no records is not a fault to be fixed in an afternoon, and saying so early is more useful than spending two days rediscovering a commissioning decision.

The control systems, and how each is addressed

IEC 62386, the parts worth knowing

Show against installed

How to do it

1. Find the documentation before touching anything.
2. Establish who owns the system, and stop if the answer is nobody -- that is the first thing to report.
3. Identify which control system it is — DALI, 0-10V, phase dimming or proprietary — because the diagnosis differs entirely.
4. Identify what drives each area — schedule, sensor, manual, or all three.
5. Find out what the system does on power failure and on control failure.
6. Change one thing at a time, and record it.
7. Leave the documentation better than you found it.

What goes wrong

Practice it

Any building with automated lighting — an office, a venue, a foyer.

What Lighting Automation Is

Programmed, controlled, self-running light.

Automated lighting

Lighting automation is lighting that is programmed and controlled to run automatically — fixtures and systems that move, change, and execute lighting changes on cue, automatically, rather than being adjusted manually in real time. This includes show lighting (automated moving lights and cues that run the lighting of a performance) and beyond (architectural and installation lighting that changes and runs on schedules or triggers). The lighting is programmed and then runs itself on cue. Lighting automation makes light run automatically on cue.

This builds on lighting control (Lighting Basics and Lighting Intermediate — fixtures, consoles, programming) toward the automated, programmed end: lighting that executes complex, precise sequences automatically. It's central to modern shows (automated lighting is how complex shows run) and increasingly to architectural and smart lighting (buildings whose lighting runs automatically). Lighting automation is programmed light that runs on cue across shows and buildings.

Where it's used

Lighting automation appears across contexts. Shows and entertainment — automated lighting cues and moving lights run the lighting of concerts, theatre, events, and broadcast (programmed sequences executed precisely on cue). Architectural lighting — building and landscape lighting that runs on schedules, scenes, and automation (changing through the day, for events, or by program). Installations and smart spaces — automated lighting in installations, retail, and smart buildings (lighting that responds and runs automatically). Wherever lighting is programmed to run itself, lighting automation is at work. Lighting automation spans shows, buildings, and installations.

These applications share the core idea (lighting programmed and automated to run on cue/schedule/trigger) but differ in context (a precisely cued concert vs. a building's scheduled lighting vs. a responsive installation). Lighting automation is used wherever light is programmed to run automatically.

Lighting automation is lighting programmed and controlled to run automatically — fixtures and systems that move, change, and execute lighting changes on cue automatically rather than being adjusted manually in real time — including show lighting (automated moving lights and cues running a performance) and architectural/installation lighting that runs on schedules or triggers; the lighting is programmed and then runs itself on cue, building on lighting control toward the automated end. It's used across shows and entertainment (automated cues and moving lights running concerts, theatre, events, broadcast), architectural lighting (buildings/landscapes on schedules and scenes), and installations/smart spaces (responsive, automatic lighting) — wherever light is programmed to run itself.

Automated Fixtures & Control

Controllable lights and how they're controlled.

Automated fixtures

Lighting automation uses controllable/automated fixtures — lights that can be controlled and can change automatically. These include moving lights (moving heads — fixtures that pan, tilt, change colour, gobo, and beam, all controllable), controllable LEDs (colour-changing, dimmable LED fixtures), and other automated fixtures. Unlike a fixed, manually-switched light, these fixtures respond to control signals and can execute programmed changes (movement, colour, intensity) automatically. Automated fixtures are the hardware of lighting automation. Understanding automated fixtures is fundamental. Automated fixtures are controllable lights that change on cue.

These fixtures (especially moving lights) are what make automated, dynamic lighting possible — they can do complex, programmed movements and changes that fixed lights can't. Automated, controllable fixtures are the building blocks of automated lighting.

Control and protocols

Automated fixtures are controlled via lighting control systems and protocols. The control comes from a console or controller (which sends control signals to the fixtures), over a control protocol — most commonly DMX (the standard lighting control protocol that carries control data to fixtures), and increasingly lighting-over-network protocols (like Art-Net or sACN, carrying lighting control over networks for large/complex systems). The controller sends commands and the fixtures respond. Understanding lighting control and protocols is central to lighting automation. Control systems send signals over protocols (DMX, networks) to the fixtures.

This control layer (console/controller + protocol + fixtures) is how automated lighting is driven: the controller runs the programming and sends control data (via DMX/network) to make the fixtures do what's programmed. The control system and protocols drive the automated fixtures.

Lighting automation uses controllable/automated fixtures — lights that can be controlled and change automatically: moving lights (moving heads that pan, tilt, change colour, gobo, and beam), controllable LEDs (colour-changing, dimmable), and other automated fixtures that respond to control signals and execute programmed changes, unlike fixed manually-switched lights; these (especially moving lights) make complex, dynamic, programmed lighting possible. They're controlled via control systems and protocols — a console/controller sends control signals over a protocol, most commonly DMX (the standard lighting control protocol) and increasingly lighting-over-network protocols (Art-Net, sACN) for large/complex systems — the controller runs the programming and sends data to make the fixtures do what's programmed.

Programming & Automation

Cues, timecode, triggers, and schedules.

Programming and cues

Automated lighting is programmed — the looks and changes are built into cues (programmed lighting states and changes), recorded on the console/controller, so they can be played back to run the lighting (building on lighting programming from Lighting Intermediate). The programming defines what the lighting does and when — the sequence of looks and changes (the cues) that make up the lighting. Programming the cues is the foundation of automated lighting — it's where the automated lighting is created. Understanding programming and cues is central to lighting automation. The lighting is programmed into cues that run it.

Good programming builds the lighting (the looks and their sequence/timing) into the system so it can run automatically and precisely. The cues are the programmed instructions the automation executes. The programmed cues are what the automated lighting runs.

Automation methods

Automated lighting runs its programming through various automation methods. Manual cue execution (an operator pressing 'go' to advance cues — semi-automated). Timecode — the lighting cues triggered automatically by timecode (synced to music, video, or a show's timeline, so the lighting runs itself perfectly in sync — common in shows and broadcast). Triggers — cues triggered by events or inputs (a sensor, a signal, an action). And schedules — lighting running on a time schedule (common in architectural/building lighting — changing by time of day, etc.). These methods make the lighting run automatically. Understanding automation methods is central to lighting automation. Timecode, triggers, and schedules run the lighting automatically.

The method depends on the application: timecode for precisely synced shows, triggers for responsive/interactive lighting, schedules for architectural/building lighting, or operator-run for flexible live shows. The automation method executes the programmed cues automatically. The lighting runs automatically via timecode, triggers, schedules, or cue execution.

Automated lighting is programmed — the looks and changes are built into cues (programmed lighting states and changes) recorded on the console/controller so they play back to run the lighting (building on lighting programming); the programming defines what the lighting does and when, and the cues are the instructions the automation executes. It runs its programming through automation methods: manual cue execution (an operator pressing 'go' — semi-automated), timecode (cues triggered automatically in sync with music/video/timeline — common in shows and broadcast), triggers (cues fired by events, sensors, or signals), and schedules (lighting on a time schedule — common in architectural/building lighting) — the method chosen for the application makes the lighting run automatically.

Lighting Automation in Practice

From show cues to smart buildings.

Show lighting automation

In shows, lighting automation runs the lighting of a performance — programmed cues (built by the lighting designer/programmer) executed precisely, often by timecode (for perfectly synced shows) or by an operator on cue, with moving lights and controllable fixtures doing complex, automated moves and changes. This is how modern concerts, theatre, and broadcast achieve their sophisticated, precise, repeatable lighting — programmed and automated rather than manually operated. Show lighting automation is a core application. Show lighting is programmed and automated to run precisely.

This connects to the Lighting Design and Lighting Direction & Leadership books: the designer programs the automated lighting, and it runs (via timecode or cues) to deliver the show's lighting precisely and repeatably every performance. Automation is what makes complex show lighting possible and consistent. Automated lighting delivers precise, repeatable show lighting.

Architectural and smart lighting

Beyond shows, lighting automation runs architectural and smart lighting — building, landscape, retail, and installation lighting that runs automatically on schedules, scenes, and triggers (lighting that changes through the day, sets scenes for different uses, responds to occupancy or events, and runs without manual operation). This overlaps with smart buildings and control systems (automated, integrated building lighting). Architectural lighting automation is a large, growing application (commercial, civic, and residential lighting increasingly automated). Lighting automation runs buildings' lighting automatically.

This shows lighting automation's range beyond entertainment — into the automated lighting of the built environment (overlapping Smart Building AV and control systems). The same core (programmed, automated lighting) applies, in a building/installation context. Combining what lighting automation is, fixtures and control, programming and methods, and these applications is lighting automation. The professional programs and runs automated lighting across shows and buildings. This is lighting automation in practice. Automated lighting runs both shows and the built environment.

In shows, lighting automation runs the lighting of a performance — programmed cues (built by the designer/programmer) executed precisely, often by timecode (perfectly synced) or by an operator on cue, with moving lights doing complex automated moves — how modern concerts, theatre, and broadcast achieve sophisticated, precise, repeatable lighting (programmed rather than manually operated). Beyond shows, it runs architectural and smart lighting — building, landscape, retail, and installation lighting on schedules, scenes, and triggers (changing through the day, setting scenes, responding to occupancy/events, running without manual operation) — overlapping smart buildings and control systems, a large, growing application across the built environment.

Lighting Automation Understood

Programmed light across entertainment and the built world.

Lighting automation summarized

Pulling it together, lighting automation is programmed, controlled lighting that runs automatically — controllable/automated fixtures, driven by control systems and protocols (DMX, networks), programmed into cues, and run automatically via timecode, triggers, schedules, or cue execution, across shows, architecture, and installations. Lighting automation is how modern lighting achieves complex, precise, repeatable, self-running results, in entertainment and the built environment. Understanding lighting automation is increasingly valuable as lighting becomes more automated everywhere. Lighting automation understood is grasping programmed, self-running light. Lighting automation is the programmed, controlled future of lighting.

Lighting automation spans the lighting world (from concert moving-light shows to automated smart-building lighting) and is growing as lighting becomes more controllable, networked, and automated. Understanding it — fixtures, control, programming, automation methods — is valuable knowledge across show and architectural lighting. This book has introduced that field. Lighting automation is the programmed, automated direction of lighting across entertainment and buildings.

Developing in lighting automation

Developing in lighting automation means building lighting fundamentals (Lighting Basics and Lighting Intermediate — fixtures, control, programming), learning the control systems and protocols (consoles, DMX, networks), getting hands-on programming and running automated lighting (cues, timecode, automation), and understanding the application area you're interested in (show lighting, or architectural/control systems). It connects to lighting design/direction (shows) and control/smart buildings (architectural). The best in lighting automation combine strong lighting and control knowledge with programming and automation skill, built hands-on. Deliberately developing these is the path forward.

From here, growing means working on more and more sophisticated automated lighting (complex shows, large architectural/smart systems), deepening your programming and control skills, and building toward lighting programming, design, and control/automation roles. It's a growing field as lighting automates across entertainment and buildings. The foundation this book provides — how automated lighting works — plus lighting and control skills is the entry point; growth comes from hands-on programming and specialization. Keep working with automated lighting, keep building your control and programming skills, and you grow in lighting automation.

Lighting automation is programmed, controlled lighting that runs automatically — controllable/automated fixtures, driven by control systems and protocols (DMX, networks), programmed into cues, and run via timecode, triggers, schedules, or cue execution, across shows, architecture, and installations — how modern lighting achieves complex, precise, repeatable, self-running results in entertainment and the built environment. It's growing as lighting becomes more controllable, networked, and automated, developed by building lighting fundamentals, learning control systems and protocols, getting hands-on programming and running automated lighting, and specializing toward show lighting programming/design or architectural control/automation roles.

Glossary

Practical sign-off

Lighting automation is learned programming real systems. The learner identifies fixtures and control, programs cues, and uses an automation method. A mentor signs off when the learner can program and run automated lighting.

• Identified automated fixtures and control protocols (DMX/network)
• Programmed lighting into cues
• Used an automation method (timecode, trigger, or schedule)
• Programmed and ran an automated lighting sequence

What the gear actually looks like

Every device named in this book, in one place. Recognition is the whole point of Chapter 4 — use this page to attach a picture to each name before you meet it on a dock at six in the morning.

Audio

Dynamic microphone

Condenser microphone

Wireless handheld

Radio rack

Mixing console

Stage box / snake

Power amplifier

Main loudspeaker

Subwoofer

Stage monitor wedge

In-ear monitors

DI box

Video

Camera

Video switcher

Scaler / processor

Projector

LED wall panel

Confidence monitor

Lighting & control

Conventional fixture

Moving-head fixture

Lighting console

Dimmer

Show controller

Media server

Networking & infrastructure

Network switch

Patch bay

Equipment rack

Power distribution

Cable tester

A properly coiled cable

Connectors, close up

XLR

TRS / TS ¼-inch

SDI on BNC

HDMI

Cat5e / Cat6 (RJ45)

Fiber optic

DMX (5-pin XLR)

AC power — not signal