Project desk: +1 732 919 3119 · [email protected]Global project coordination

Dialight Lighting Notes

Why I Diagnose the Circuit Before Swapping the Switch (And What That Means for Legrand, Zigbee, and Your Outlets)

2026-11-06 · Tomasz Zielinski · Lighting

I've been running lighting emergencies for a small commercial outfit for eleven years now—which is a fancy way of saying I've spent a lot of nights figuring out why a conference room is dark. And here's the thing I'd tattoo on every install van if I could: the switch is almost never the problem.

I know how that sounds. Like the guy who opens every call with "well, actually." But I can back it up. I've handled over 400 emergency callouts since 2014, and I've started tracking what we actually find. In about 40% of "bad switch" calls, the switch was either fine or was a symptom, not the cause. Replacing it didn't fix the room. The first guy just sent a bill.

Saving five minutes on diagnosis costs you two hours on a callback. Every time.

Can a bad light switch cause outlets not to work? Yes. But it's not the common one.

I want to actually answer this question, because I get asked it constantly by facility managers who are standing in a dark hallway with a multimeter and a bad attitude.

The short answer: yes, a failed switch can kill downstream outlets—if the circuit is wired that way. In older commercial and residential wiring, the switch box and the outlet box often sit on the same branch circuit. If the switch has an internal contact failure or a loose neutral, everything wired downstream of that switch box loses the loop. That's real. I've seen it.

The longer answer: when I actually trace these calls, three other causes show up more often:

None of those get fixed by pulling a Legrand dimmer out of the box. So let's talk about what to spec instead.

On Legrand dimmers: the right ecosystem, the wrong default assumption

I've installed enough Legrand products to have opinions. The Radiant series is solid, predictable, and I've had almost no returns. The adorne line is where things get interesting—the soft-touch paddle and the configurable frame options are genuinely useful in client-facing spaces where the room is part of the pitch.

But the Legrand adorne LED dimmer has a load compatibility floor that I wish came with a sticker. It performs beautifully on dimmable LED loads above roughly 25W and rated for forward-phase dimming. Feed it a cheap non-dimmable fixture (which is exactly what a lot of retrofit jobs end up with), and you'll get flicker or drop-out. The client blames the dimmer. The dimmer is fine.

I watched an installer replace the same adorne dimmer three times over a two-month span because a client's bargain downlights wouldn't dim below 15%. By the time he called me, he'd eaten about $180 in dimmers (rough estimate based on late-2024 retail; verify current pricing with your distributor) and was still going to need the fixtures swapped anyway. The right dimmable LED lamps would have cost under $50.

That's the whole prevention-over-cure argument in one job. Test the load before you buy the hardware. A 10-minute compatibility check at spec time beats a 2-hour callback in an occupied space every time.

I still kick myself for not documenting a load calculation on a job in 2022. The client had mixed dimmable and non-dimmable fixtures on the same circuit and neither I nor the installer caught it until the ribbon-cutting. We reprogrammed, replaced four lamps, and the client will never quite trust that room again. One sheet of paper would have prevented it.

Zigbee integration: SmartThings, Yolink, and what actually holds up

Now the part people email me about: Zigbee SmartThings and Yolink Zigbee interoperability with Legrand devices.

Legrand's Zigbee commitment is, on paper, the right call. Open protocol means you're not locked to a single app ecosystem, and for B2B clients with existing building automation, that flexibility matters. In practice, the results depend heavily on how you build the gateway layer.

Honestly, I'm not 100% sure why some Legrand Zigbee devices need an extra pairing cycle on third-party hubs while others don't—I suspect it's firmware handshake differences, but I've seen inconsistent behavior on identical switch models from different production runs. If someone has a definitive answer, I'd genuinely like to know.

The one recommendation I'll stand behind: pick your protocol strategy before you pick your hardware, not after. The most expensive mistake in smart lighting isn't the device—it's the reconfiguration.

"But what if the switch really is bad?"

Fair pushback. If you've actually diagnosed the switch—tested it under load, traced the circuit, confirmed a failed contact—then yes, replace it. I'm not suggesting you refurbish a dead switch out of principle. That's absurd.

My point is narrower and, frankly, less flattering to our trade: replacing the switch is what we do first out of habit, not last out of evidence. And every time you replace a part on a guess, you're charging the client for a second visit you already knew was coming.

There's something satisfying about getting the diagnosis right the first time—finding the burnt backstab, catching the half-hot someone forgot, tracing that neutral back to the junction box. It's not glamorous work. But it's the difference between a callback and a referral.

Write the circuit map. Test the load. Check the neutral. Then—and only then—replace the switch. That order is the cheapest insurance policy in the lighting business, and it's the one thing I wish someone had tattooed on my forehead in 2014.