Lighting controls and dimming: 0-10V, DALI, and the driver match

What lighting controls are, and where the savings live

Lighting controls are the switching, dimming, occupancy, and daylight devices that decide when the lights run, how bright they run, and when they shut off. On a commercial job they are not an upgrade you sell to the client. They are required to pass plan review, and they are where the LED savings and the code compliance actually live. An efficient fixture that burns full output in an empty room all night saves far less than a modest one that goes dark when the space is empty.

Three jobs sit inside that one word. The controls turn light off when nobody is there, dim it down when daylight is doing the work, and hold it at the level the task needs instead of full output. The fixture and the light level are the design problem, covered in the commercial lighting design guide, and the source itself is the LED retrofit guide. This guide is the controls and the dimming underneath them: the protocols, the sensors, the wiring, and the commissioning that makes the whole thing work the way the permit said it would.

The mistake that runs through bad control jobs is treating the hardware as the finish line. The sensors get mounted, the drivers get wired, the system gets energized, and nobody sets the time delays, calibrates the daylight sensor, or proves the protocol matches. Then the lights flicker, the occupant disables the sensor, and the savings never appear. Every section here exists to close that gap between installed and working.

Why does the energy code require lighting controls?

The energy code requires controls because uncontrolled lighting is the cheapest energy to stop wasting, and the code writers know it. Lighting that runs only when a space is occupied, only as bright as the task and the daylight require, can cut lighting energy well beyond what the LED fixture alone saves. Layered controls, occupancy plus daylight plus task tuning plus scheduling, are documented to bring lighting energy down to a fraction of an uncontrolled baseline.

So the codes mandate them. ANSI/ASHRAE/IES 90.1, the IECC, and California's Title 24, Part 6 all require some mix of automatic shutoff, occupancy or vacancy sensing, daylight-responsive control near glazing, and multilevel dimming in most commercial spaces. A bare wall switch does not comply anymore. The specific triggers, thresholds, and exemptions vary by code and edition, so the adopted edition and the AHJ control what you actually owe on a given job.

Comfort is the quieter reason, and it decides whether the controls survive. A daylight sensor that hunts, an occupancy sensor that drops people into the dark mid-meeting, or a dimmer that flickers gets disabled by the people who live with it, and a disabled control saves nothing and fails compliance. The savings the code is after only exist if the controls are tuned well enough that nobody wants to rip them out.

The control strategies and what each one does

Lighting control resolves into a handful of strategies, and most commercial jobs use several at once because the code asks for several at once. Each attacks a different kind of waste: light in an empty room, light competing with the sun, light brighter than the task, and light left on after hours.

Occupancy and vacancy sensing kills light in unoccupied space. Daylight harvesting dims the fixtures near windows and skylights as the sun fills in. Multilevel and dimming control lets the space run at less than full output instead of all-on or all-off. Scheduling, through a time switch or a control system, sweeps the building off after hours and is the usual way the required automatic shutoff gets met. Demand response, where the utility signals the building to shed load during a grid peak, is a fifth strategy that some codes now require above a wattage threshold.

The strategies stack rather than compete. A typical office bay runs occupancy sensing for the empty-room case, daylight dimming on the window row, task tuning to cap the top end, and a time-switch sweep as the backstop. Knowing which strategy the code is asking for in a given space is the difference between a system that passes and one that gets red-tagged for a missing control type.

What is the difference between occupancy and vacancy sensors?

An occupancy sensor is auto-on, auto-off: it turns the lights on when someone walks in and off after the room reads empty for a set delay. A vacancy sensor is manual-on, auto-off: someone has to switch the lights on, but the sensor still turns them off automatically when the space empties. Same off behavior, different on behavior, and that difference is often a code requirement, not a preference.

Vacancy, the manual-on mode, saves more energy because the lights never come on for a person who only stuck their head in, and the codes lean on it for exactly that reason. Many editions require manual-on vacancy control in specific space types, offices being the common one, so installing an auto-on occupancy sensor where the code calls for manual-on vacancy is a real compliance miss even though the hardware looks identical. Most commercial sensors do both modes on a setting, so the failure is usually a setup error, not the wrong device.

Pick the mode by the space and the code. Auto-on suits places where a person walking in with full hands needs light immediately and safely, like a corridor, a restroom, or a stairwell. Manual-on vacancy suits private offices and similar rooms where the occupant can reach a switch and the extra savings are worth it. Confirm which the adopted code requires for each space type before you set the sensor, because the inspector checks the mode, not just the presence of a sensor.

PIR, ultrasonic, and dual-technology sensors

Occupancy sensors detect people three ways, and the right one depends on the room and how the motion looks. Passive infrared, PIR, reads the heat signature of a body moving across its field of view. It needs line of sight, it is good at catching someone walking, and it rarely false-trips on air movement, which makes it the safe default for open rooms with a clear view of the space.

Ultrasonic sensors send out sound waves and read the shift in the echo when something moves, so they catch fine motion like typing or a small posture change, and they work around partitions and into the spaces PIR cannot see. The cost is false trips. An ultrasonic sensor will read the air off an HVAC diffuser or a moving door as occupancy and hold the lights on in an empty room, so placement and sensitivity matter more than with PIR. Dual-technology sensors run both: they commonly require PIR and ultrasonic to agree before turning lights on, so a draft alone does not trip them, but accept either one to hold the lights on, so a person sitting still at a desk does not get dropped into the dark.

Match the technology to the failure you are trying to avoid. A restroom with stalls and a partition wants ultrasonic or dual-tech, because PIR loses the line of sight and the lights die on someone in a stall. An open warehouse aisle wants PIR for the clean walking detection and the immunity to airflow. A private office where someone reads quietly is the classic dual-tech case, because PIR alone times out on the still occupant and ultrasonic alone false-holds on the air handler.

What is daylight harvesting?

Daylight harvesting is dimming the electric lights near windows and skylights as daylight rises, so the room holds a steady level while the fixtures back off and save energy. A photosensor reads the light, and the control dims the daylit-zone fixtures down as the sun fills in and brings them back up as it fades. The energy code requires it in the zones where useful daylight lands, and it is one of the larger savings on a perimeter-heavy building.

The sensor runs one of two ways. An open-loop photosensor reads only incoming daylight, usually mounted facing a skylight or near the aperture, and infers how much electric light to drop. It can drive many fixtures from one sensor and suits skylit and open layouts, but it needs careful commissioning because it is inferring the room level rather than measuring it. A closed-loop photosensor reads the combined daylight and electric light at the work surface and trims to hold a target, so it measures the result of its own adjustment and is generally easier to set, at the cost of seeing only its own zone.

The setpoint and the fade rate are where these go wrong. Set the dimming target above the maintained design level, not at it, or the room reads dim every time a cloud passes. Use a slow fade, on the order of 30 to 60 seconds, so the lights do not strobe up and down as the weather moves. The crews that skip this leave the sensor at the factory setting, the room hunts or dims at the wrong level, and the occupant disables it, which kills the savings and the compliance both. Commission on a partly cloudy day if you can, because that is when the hunting shows.

0-10V dimming, the commercial workhorse

0-10V is the most common dimming protocol in commercial work, and for most jobs it is the one to reach for first. It is analog: a separate pair of low-voltage control wires runs alongside the line voltage, and the DC voltage on that pair, from 0 to 10 volts, tells the driver how bright to run. 10 volts is full output and the voltage drops toward 0 as the light dims. The wires are conventionally purple for the positive and gray for the negative, and they are polarity-sensitive, so reversing them is a real wiring fault.

Know which device drives the signal. In the standard commercial setup the LED driver sources the control current onto the pair, and the dimmer or sensor sinks current to pull the voltage down and dim the light. That is why a single low-voltage control can dim a whole bank of drivers wired in parallel, up to the control's rated driver count. Match the control to the drivers, mind that count, and keep the polarity consistent end to end.

0-10V is simple and reliable, and it has two limits worth knowing cold. First, the low end. Many 0-10V drivers do not dim much below about 1 percent, and at 0 volts a lot of them still hold a minimum glow rather than going fully off, so true off usually needs a relay or a switch leg in addition to the dimming pair. Second, no addressing. Every driver on a control pair dims together as one zone, so the only way to get independent zones is to run separate control pairs and separate sensors. When the job needs per-fixture control or software regrouping, that is where 0-10V runs out and DALI starts. The 0-10V analog interface traces back to the lighting provisions in IEC 60929; confirm the driver's listed input range against the control.

DALI and DALI-2: digital addressable control

DALI, the Digital Addressable Lighting Interface, is the digital alternative to 0-10V, and it earns its keep when the space needs flexible zoning, feedback, or per-fixture control. Instead of one analog voltage shared across a pair, DALI runs a digital two-wire bus where each device carries an address. The controller talks to fixtures individually, groups them in software rather than in the wiring, and reads data back from them. The standard is IEC 62386, and DALI-2 is the certified version that requires DALI Alliance testing so devices from different makers interoperate.

Two things set DALI apart from 0-10V. It is addressable, so a single bus can run many fixtures and the commissioning technician assigns each one to a group in software, which means you re-zone a floor by reprogramming instead of repulling control wire. And it is two-way: a driver reports its status, lamp or driver failure, operating hours, and energy back to the controller, which turns maintenance from a walk-around into a report. A DALI line commonly addresses up to 64 control gear, and the bus itself is polarity-insensitive, which removes the 0-10V polarity trap.

The flexibility costs commissioning. A DALI system is not finished when it is wired; the addressing, grouping, and scene setup happen in software, and that programming is a real line item and a real skill. The payoff is a system you can reconfigure and diagnose without touching the ceiling, which is why DALI shows up on large, churn-heavy, or feedback-driven jobs while 0-10V holds the simpler ones. Pick DALI when the zoning will change or the feedback matters; pick 0-10V when the zones are fixed and simple.

Phase dimming: forward, reverse, and ELV

Phase dimming controls brightness by chopping part of each AC half-cycle at the line voltage, with no separate control wire, which is why it dominates residential and small-commercial work where running a control pair is not worth it. There are two kinds, and the difference decides whether an LED runs clean or flickers.

Forward-phase dimming, also called leading-edge or TRIAC, cuts the front of the sine wave. It came up with incandescent and works with magnetic low-voltage (MLV) transformers, and it is the older, more rugged technology. Reverse-phase dimming, also called trailing-edge or ELV (electronic low voltage), cuts the back of the wave and is gentler on electronic loads, so it dims most LED drivers and electronic low-voltage supplies more smoothly with less flicker and buzz. The short version: forward-phase suits incandescent and MLV, reverse-phase suits LED and ELV.

Compatibility is the whole game here. An LED on the wrong phase-cut dimmer flickers, buzzes, refuses to go low, or drops out before the slider reaches the bottom, and you cannot tell by looking, you have to check. Use a dimmer rated for the load type, confirm the LED driver is marked dimmable on that method, and check it against the manufacturer's compatibility list. On a commercial job that needs real dimming across many fixtures, 0-10V or DALI usually beats fighting phase-cut compatibility on a long row of LED.

The LED driver and dimmer must match

This is the one that generates the most callbacks, so it gets the bluntest treatment: the LED driver has to match the dimmer and the protocol, or the lights misbehave. The driver is what actually dims the LED, and it is built for one control method. A 0-10V driver wants a 0-10V control. A DALI driver wants a DALI controller. A phase-dimmable driver wants the right forward or reverse phase-cut dimmer. Cross those and the system does not work right, no matter how good either piece is on its own.

A mismatch shows up as flicker, drop-out where the light shuts off before the control reaches the bottom, no usable low end, buzz, or a dead zone in the dimming travel. The other half of the same problem is minimum load. LED draws a tiny fraction of the current an incandescent did, and many older or line-voltage dimmers need a minimum load to behave, so a handful of LED fixtures on a dimmer built for a heavier load will flicker or refuse to dim even when the method is nominally right.

The fix is to check before you buy, not after you energize. Manufacturers publish driver-and-dimmer compatibility lists precisely because the combinations are not interchangeable, and that list is the document that settles it. Confirm the driver's control method, confirm the dimmer is listed compatible with that specific driver, and confirm the load falls inside the dimmer's rated range. Finding the mismatch at commissioning, with the ceiling closed and the owner watching the lights stutter, is the expensive way to learn this.

Flicker: where it comes from and why it matters

Flicker is the rapid variation in light output that an LED can produce when the driver, the dimmer, or the power feeding it is not clean. Most field flicker traces to two causes: a driver and dimmer that do not match, and the low end of the dimming range, where a marginal driver loses its grip and the output starts to pulse. A driver that runs steady at full output can still flicker badly at 10 percent, which is why you sweep the whole dimming range at commissioning, not just full bright.

It is more than an annoyance. Flicker is linked to headaches, eyestrain, and migraine for some people, and at certain frequencies and depths it is a documented trigger for photosensitive seizure in a small population, which is why the better drivers are rated for low percent flicker. You will not always see it directly, because the eye adapts, but the body still reacts, and the complaints that follow an install are real even when the meter says the light level is fine.

The camera is the other tell, and on some jobs it is the spec. Flicker that the eye misses shows up as rolling bands on video from a phone, a security camera, or a broadcast setup, because the camera's shutter samples the pulsing light. On studios, conference rooms with video, retail with cameras, and any machine-vision line, low-flicker drivers are a requirement, not a nicety. When you chase a flicker complaint, start at the driver-dimmer match and the low-end setting before you suspect anything exotic.

Wireless lighting controls

Wireless controls move the signal off a wire and onto a radio, which makes them the natural answer for a retrofit where there is no control wire and no appetite to open the ceiling to pull one. The fixtures, sensors, and switches talk over a radio mesh, commonly Bluetooth mesh or a proprietary mesh, and the system gets commissioned through an app instead of a control panel.

The win is the absence of control wire. In an existing building, pulling 0-10V or DALI cable to every fixture and sensor is the expensive part of adding controls, and a wireless system skips it, so the labor drops and the disruption drops. Many wireless sensors are battery-powered or energy-harvesting, so even the sensor needs no wiring. That is the retrofit case in one sentence: controls where running cable would cost more than the controls.

The tradeoffs are radio and software. The mesh has to be reliable through the building's walls and racks, the commissioning lives in an app and a database that someone has to own, and the system depends on the vendor's platform in a way a simple 0-10V pair does not. For new construction with the ceiling open, hard-wired 0-10V or DALI is often the steadier choice. For a retrofit into a closed building, wireless is frequently what makes the controls affordable at all.

Networked and luminaire-level controls

Networked lighting control, NLC, is the step up from standalone sensors and dimmers to a system where the fixtures, sensors, and controls all talk on a network and get managed together. Instead of a sensor wired to a few fixtures in one room, the whole building's lighting reports to and takes direction from one platform, which makes scheduling, task tuning, daylight response, and energy monitoring central rather than per-room.

Luminaire-level lighting control, LLLC, is the densest version: every fixture ships with its own integrated occupancy and daylight sensor and its own controller, so each luminaire senses and dims itself and coordinates with its neighbors over the network. The granularity is the point. Light follows people fixture by fixture, daylight response happens per fixture by its own window exposure, and the energy data comes back at the fixture level.

The reason to care on the estimate is the rebate. Utilities pay extra for NLC and LLLC on top of the fixture rebate, commonly on the order of 30 to 50 dollars per fixture, and the systems that qualify are the ones on the DesignLights Consortium NLC qualified products list. The DLC list is the gate for most programs, so confirm the system is listed and confirm the specific utility program before you price the rebate into the job. The added control also stacks real energy savings on top of the LED, which is what justifies the system in the first place.

Class 2 control wiring

The control wiring for 0-10V, DALI, and most low-voltage lighting control is a Class 2 circuit under NEC Article 725, and that classification carries rules that are easy to skip and easy to fail on. Class 2 is a power-limited circuit, which is what makes the low-voltage control safe to handle, but the code still governs how it is run, separated, and rated.

Keep it separated from the line voltage. Article 725 requires Class 2 conductors to be kept apart from electric light and power conductors, commonly at least 2 inches, unless one side or the other is in a raceway or otherwise enclosed so a line-voltage fault cannot energize the control circuit. The point of the separation is exactly that: a 277-volt fault finding its way onto a 0-10V pair destroys the drivers and creates a hazard. Run the control cable in its own path, or in the conditions the code allows, not zip-tied to the line conductors.

Rate the cable for where it runs. Control cable in an air-handling plenum has to be plenum-rated, marked CL2P or CMP, for the fire and smoke performance the code requires in that space, and the cheaper riser or general-purpose cable is not allowed there. And on 0-10V specifically, the control pair is polarity-sensitive, so the separation and the rating are not the only wiring discipline that matters: get the purple-positive, gray-negative polarity consistent end to end, because a reversed pair is its own failure. The Class 2 fundamentals are covered in the low-voltage and Class 2 wiring material; confirm the adopted code edition for the current separation and rating rules.

What controls does the energy code require?

The energy code requires a set of controls in most commercial spaces, and the core list is steady across the codes even though the details differ: automatic shutoff so the lights turn off after hours, occupancy or vacancy sensing in enclosed and intermittently used rooms, daylight-responsive control in the zones near glazing, and multilevel or dimming control so the space can run below full output. Recent editions also reach into receptacle control and, above a wattage threshold, demand response. The commercial lighting design guide covers which control lands in which space type; this is the wiring and commissioning side of the same requirement.

The governing standard depends on the jurisdiction. Most of the United States runs on ANSI/ASHRAE/IES 90.1 or the IECC, which references 90.1 as a compliance path, while California runs the stricter Title 24, Part 6. The control list and the trigger numbers have tightened almost every cycle, so the edition matters, and the adopted edition plus any local amendments control what you owe. Title 24, for example, requires demand-responsive lighting capable of responding to an OpenADR signal in buildings at or above a stated installed-lighting-power threshold.

Treat the code list as the minimum to pass, not the design. Confirm the adopted edition with the AHJ, confirm which control type each space requires, and confirm the manual-on versus auto-on requirement per space, because that is the detail the plan reviewer checks and the field most often gets wrong.

What is lighting controls acceptance testing?

Acceptance testing is the documented functional test that proves the installed controls actually work, and on many jobs it is required before the building gets its certificate of occupancy. Installing the controls and proving the controls are two different milestones, and the code increasingly demands the second one in writing. A system that is wired but never tested is the single most common reason the promised savings never show up.

California makes it formal. Title 24 requires acceptance testing performed by a certified acceptance test technician under the CALCTP-AT program, who runs a plan review, a construction inspection, and a documented functional test of each control type, occupancy and vacancy, daylighting, automatic shutoff, demand response, and outdoor controls, before final occupancy. ASHRAE 90.1 and the IECC carry their own functional-testing and commissioning requirements that have grown each cycle, so even outside California the controls usually have to be demonstrated, not just installed.

The test checks behavior, not presence. The occupancy sensors have to time out and shut off, the daylight sensors have to dim the fixtures as light rises, the time switch has to sweep the building off, and the dimming has to reach its levels without flicker. The acceptance report records the time delays, the dimming ranges, and the daylight calibration, so there is evidence on file that the controls do what the permit said. Budget the test and the time to pass it, because a failed acceptance test holds the occupancy permit.

Commissioning the controls

Commissioning is the work of turning installed hardware into a system that behaves, and it is where most of the real performance is won or lost. The fixtures and sensors come out of the box on factory defaults that almost never match the space, so commissioning is the step that sets the time delays, the sensitivity, the daylight setpoints, the fade rates, and the trim to the actual room.

Set the occupancy time delay long enough that people are not dropped into the dark during a normal pause, and short enough to still save energy; many spaces land somewhere around 10 to 20 minutes, with the code and the space driving the number. Set the sensor sensitivity so it catches the occupants without false-tripping on airflow or holding an empty room on. Set the daylight setpoint above the maintained design level and the fade slow, so the room does not hunt or strobe as clouds pass. Then walk every zone and confirm the behavior with the lights, not just the settings screen.

The failure to plan for is the daylight sensor left at default. Dropped in and never calibrated, it dims at the wrong level or hunts, and the occupant disables it, which kills the savings and can put the building out of compliance. Most callbacks on controls are not the hardware. They are the commissioning nobody finished. Capture the final settings in the record, because the next person to troubleshoot a zone needs to know what it was set to and why.

High-end and low-end trim

Trim sets the ceiling and the floor of the dimming range, and both are commissioning settings that matter more than people expect. High-end trim caps the maximum the lights are allowed to reach. Low-end trim raises the minimum they are allowed to drop to.

High-end trim, also called task tuning, is the bigger energy play. A space is often designed and installed brighter than the task actually needs, partly to cover lamp aging that LED depreciation has already mostly removed. Capping the top end, say holding a space at 80 percent of full output where the task is fine at that level, cuts energy directly and the occupants generally never notice, because they never saw the missing 20 percent. Task tuning is documented to save a meaningful share of lighting energy on its own, and it is free once the controls are dimmable, so leaving the fixtures at 100 percent is leaving savings on the table.

Low-end trim solves a different problem: the bottom of the dimming range is where a driver flickers, buzzes, or drops out, so raising the floor to the lowest level the driver runs cleanly keeps the dimming smooth all the way down. Set the low end above the point where the specific driver misbehaves and you stop the flicker complaint that lives at the bottom of the range. Both trims get set at commissioning and recorded, because they are invisible later and a future tech needs to know they exist.

Why are the lights flickering or not dimming right?

When dimmed LEDs misbehave, the cause is almost always one of a short list, and they rank by how often they bite. The first suspect is the driver-and-dimmer mismatch: a driver on the wrong protocol or a dimmer not listed for that driver flickers, buzzes, or refuses to go low. Check the compatibility list before you suspect anything else, because this is the most common cause by a wide margin.

The low end is the second suspect. Flicker and drop-out that only appear when the lights are dimmed down point at the bottom of the range, where a marginal driver loses control or the dimmer falls below its minimum load. Raise the low-end trim to where the driver runs clean, or confirm the load meets the dimmer's minimum. A light that goes dark before the control reaches the bottom is drop-out, and it is the same family of problem.

On 0-10V, reversed polarity is its own failure, so when a 0-10V zone runs full bright and ignores the dimmer, or behaves backward, check the purple-positive and gray-negative pair before anything else. Sensor problems are the other big category: an ultrasonic sensor false-tripping on HVAC airflow holds an empty room on, and an over-tight time delay or low sensitivity drops people into the dark. Walk the symptom back to the strategy. Lights on in an empty room is a sensor or a delay; flicker on dimming is the driver match or the low end; a 0-10V zone stuck bright is usually polarity.

Office and data center cases

The office is the bread-and-butter controls job, and it stacks the strategies: manual-on vacancy sensing in the private offices where the code wants it, daylight dimming on the window rows, task tuning to cap the top end, and a time-switch sweep as the after-hours backstop. The open plan adds the wrinkle that one sensor must not hold an empty neighbor's lights on, so the control zones stay small enough that occupancy in one area does not light the next.

The data center white space is the opposite environment and a strong controls case for it. The room is mostly unoccupied, and every watt of lighting is heat the cooling has to remove, so occupancy control that holds the aisles dark until a technician walks in pays twice, once on the lighting and once on the cooling. Zone it aisle by aisle so the light follows the work, and coordinate the install with the facility's change control, because a live white space does not let you open the ceiling whenever you want. The lighting-design guide covers the light levels for both cases; this is the control layer on top of them.

What to document

A control system that nobody documented is one nobody can commission, troubleshoot, or claim a rebate on. The record is what the acceptance technician verifies, what the next electrician reads when a zone goes dark, and what the rebate application needs. Build it zone by zone so every control ties to a location.

Capture the control type and the dimming protocol for each zone, the sensor technology and mode, the driver and its control method, and the commissioned settings that are otherwise invisible: the time delay, the daylight setpoint, the high-end and low-end trim. Note the 0-10V polarity convention used, record the acceptance test result, and keep the cut sheets and the DLC listing for any networked system. The table below is the minimum spine of that record.

Common mistakes

• Putting an LED driver on the wrong dimmer or protocol, so the lights flicker, buzz, or drop out.
• Reversing the 0-10V control pair, so a zone runs full bright or behaves backward.
• Installing auto-on occupancy where the code requires manual-on vacancy control.
• Skipping the required acceptance test, so a failed control type surfaces at the occupancy permit.
• Leaving the daylight sensor at the factory setpoint, so it hunts or dims wrong and gets disabled.
• Running the control cable as ordinary wire instead of Class 2, separated from line voltage and plenum-rated where required.
• Energizing the system and never commissioning the time delays, trim, and setpoints.
• Pricing a dimming job without checking the driver-and-dimmer compatibility list first.

Field checklist

Standards and references

The energy code is what makes the controls mandatory. ANSI/ASHRAE/IES 90.1, the IECC, and California's Title 24, Part 6 set which controls are required, the multilevel and daylight and shutoff provisions, and the functional and acceptance testing. The required control types, the trigger thresholds, and the demand-response and acceptance-testing rules vary by code and edition, so the adopted edition and any local amendments control, and the AHJ enforces. Title 24's acceptance testing runs through the CALCTP-AT certified technician program.

The dimming protocols have their own standards. The 0-10V analog interface traces to the lighting provisions in IEC 60929, and DALI and DALI-2 are defined by IEC 62386, with DALI-2 adding the certification that makes multi-vendor devices interoperate. The driver-and-dimmer compatibility list published by the fixture and control manufacturers is the document that settles whether a specific pair works, and it governs the decision more reliably than any general rule, so check it before you commit to a combination.

The wiring sits under the NEC, NFPA 70. Low-voltage lighting control is a Class 2 circuit under Article 725, which governs the power limit, the separation from line-voltage conductors, and the cable rating, including plenum-rated cable in air-handling spaces. Cite the standard that controls the specific point, confirm the section and edition against what the jurisdiction has adopted before relying on a number, and let the project specification and the manufacturer's instructions override a rule of thumb when they are stricter. Two things are worth stressing every time: match the driver to the dimmer, keep the 0-10V polarity right, and prove it with the acceptance test.

Units, terms, and conversions

Lighting controls carry a stack of acronyms, and the same idea reads differently across a control submittal, a driver cut sheet, and a code section. Keeping the protocol straight from the strategy is half the battle.

The protocols are 0-10V (analog, a 0 to 10 volt DC control pair), DALI and DALI-2 (digital addressable, IEC 62386), and phase-cut, which splits into forward-phase, also called leading-edge or TRIAC, and reverse-phase, also called trailing-edge or ELV. The strategies are occupancy and vacancy sensing, daylight harvesting, multilevel dimming, scheduling, and demand response. The sensor technologies are PIR (passive infrared), ultrasonic, and dual-technology. NLC is networked lighting control and LLLC is luminaire-level lighting control. Trim is high-end (task tuning) and low-end (the flicker floor). The control wiring is Class 2 under NEC Article 725.

0-10V

Analog dimming over a low-voltage control pair, 10 V full to about 1 percent floor, polarity-sensitive

DALI / DALI-2

Digital addressable lighting interface, IEC 62386, with per-fixture addressing and two-way feedback

Forward / reverse phase

Leading-edge (TRIAC, incandescent/MLV) and trailing-edge (ELV, LED/electronic) line-voltage dimming

Occupancy vs vacancy

Auto-on/auto-off versus manual-on/auto-off; code often requires manual-on vacancy

PIR / ultrasonic / dual-tech

Heat-motion line-of-sight, sound-echo fine motion, and both combined to cut false trips

Daylight harvesting

Photosensor dimming of daylit-zone fixtures, open-loop or closed-loop, set above the design level

Task tuning / trim

High-end trim caps the maximum for savings; low-end trim raises the minimum above the flicker floor

NLC / LLLC

Networked lighting control, and luminaire-level control with a sensor in every fixture, rebate-eligible on the DLC list

Related tools

Calculators and readiness checks for this work

Compare your options

FAQ

People also ask

• How many footcandles do you need for an office?
• What is a ballast bypass LED tube?
• How long must emergency lights stay on?
• What is a fleet safety program?
• What is BMS commissioning?