Can You Replace Track Lighting Heads? Yes—But the Power Supply Is the Real Problem

Here's a phone call I get more often than I'd like.

"My nursery chandelier has been flickering for a week. My wife won't let me ignore it anymore."

A few days later, another one: "The entry chandelier is fine at full brightness, but at 40% dimming it hums and strobes. The electrician says it's the LED driver."

And then there's the search-engine classic that starts a thousand rabbit holes: "Can you replace track lighting heads?" Yes, you can. Track heads unclip, twist, or slide out depending on the track system. But if you're replacing heads because they keep failing, you're treating a symptom, not the disease.

In my role coordinating rush power-supply orders for lighting integrators and electrical contractors, I sit between two very different statements: "the fixture is broken" and "the fixture is fine—the driver is what failed." That second statement is true far more often than people expect.

A Fixture That "Keeps Dying" Is Usually a Power Supply Problem

Here's the thing: when an LED fixture flickers, buzzes, fades, or dies at partial brightness, the LED module itself is usually innocent. LEDs are solid-state components—no filament to burn out, no glass envelope to crack. Given clean, correctly regulated power, a typical LED module is rated for 50,000 to 100,000 hours (check your specific module's datasheet for the actual number). The component that dies, in most cases, is the driver or power supply feeding it.

I used to think otherwise. When I first started coordinating replacements, I assumed a dead driver was a dead component—a bad batch, a manufacturing defect, bad luck. Three years of return-and-test tickets changed my mind. A meaningful share of "failed" drivers that get shipped back and tested are, in my experience, perfectly fine. The installation killed them. More precisely, the choices made during installation killed them.

The Wrong Kind of Power Delivery

LEDs need one of two things: constant current or constant voltage, depending on how the LED module is designed. Getting that wrong is the most common mistake I see.

The Mean Well LRS-350-24 is a constant-voltage supply. According to its datasheet (meanwell.com), it delivers 24V DC at up to 14.6A—350.4W of capacity for LED strips, signage modules, or under-cabinet tape that expects a 24V feed. It's a workhorse for anything designed to run on 24V.

A bare LED module—the kind that goes into a commercial fixture with no integrated driver—does not want 24V. It wants a specific current, typically something between 350mA and 1,400mA. Feed it from a constant-voltage supply without current limiting and you're asking for an early failure. That's where constant-current drivers like the Mean Well LPC or HLG series come in. They hold current steady and let voltage float within a designed window.

The practical point: you can replace a track lighting head with a perfect new head, and if the supply feeding the track is the wrong type, the new head will fail exactly like the old one. At least, that's been my experience with commercial track retrofits.

Wiring by Assumption

Then there's the wiring.

When a customer emails asking for a "Mean Well LED driver wiring diagram," nine times out of ten they have a dimmable driver in hand and aren't sure which wire does what. I understand. The jargon isn't obvious. Mean Well's 3-in-1 dimming function—used across its dimmable driver lines—accepts a resistor, a PWM signal, or a 0-10V DC voltage between the DIM+ and DIM- terminals (Source: Mean Well product datasheets). Wire the AC input into the dimming terminals, or pair a non-dimmable supply with a dimmer that doesn't match it, and no amount of fixture replacement will fix the result.

I learned that lesson through somebody else's invoice. In March 2024, a contractor was installing a track lighting system for a retail client. The spec called for a Mean Well LPC-60-700 constant-current driver. The contractor swapped in an alternate driver to save $24 per unit, assuming "same specifications" meant identical wiring. Didn't verify the pinout. Turned out it wasn't the same. The fixtures flickered audibly, the client rejected the installation, and the contractor had a very urgent problem 48 hours before the store opening.

I've lost count of how many similar stories I've heard. I wish I had tracked them more carefully. What I can say anecdotally: in 2024 alone I coordinated 80-plus rush orders for replacement power supplies—maybe 85, I'd have to check the log—and in a solid share of those, the original "failed" driver tested fine at the distributor. The issue was wiring, environment, or mismatch. Not the component.

The Bench Test Lied to You

Here's the quiet killer: the system worked on the bench.

A typical setup is straightforward. Driver, LED module, wiring diagram on the table. Current is stable, brightness is even, no flicker on the scope. Then the same components go into a ceiling cavity with no airflow, a long wire run, and a breaker shared with four other circuits. Suddenly the bench result doesn't matter.

Heat is the big one. Many power supplies derate at elevated ambient temperatures. According to Mean Well's technical documentation, output current starts to roll off above roughly 50°C for many models. A driver tucked into an enclosed chandelier mount can cook itself to 60°C in July. The bench was a comfortable 25°C.

Voltage drop over long runs and inrush current when capacitors charge on power-up add to the list. All three conspire to make a correctly selected driver fail in the field. A homeowner staring at a flickering nursery chandelier doesn't see any of that. They just see a bad product.

A $24 "Savings" That Cost $3,230

Now for the part I care about most: what these choices actually cost.

Back to the March 2024 job. Sixty drivers. The contractor saved $24 per unit by switching—$1,440 on the purchase order. Then the failure:

• Rush shipment of 60 correct drivers: $180 in freight.
• Weekend electrician re-visit: $950.
• Penalty clause for missing the store opening: $2,100.

Total: $3,230 to "save" $1,440. And that doesn't count the relationship damage, the project management hours, or the contractor looking unreliable in front of the client. When I'm triaging a rush order, I ask three things: how many hours until the deadline, can the part physically arrive in time, and what's the worst case if it doesn't. In that order. The worst-case number was more than ten times the savings.

The same pattern plays out in smaller forms every week. A dimmer that doesn't match the driver's dimming range. A power supply sized to the fixture's faceplate wattage instead of its actual driver draw. A $35 extra for next-day air that only gets paid after the client calls angry. Every one of those is a total cost of ownership problem, and almost nobody runs the full calculation before purchasing.

We lost a $14,000 contract in 2023 because we didn't keep a dimmable driver in stock and the standard lead time was three weeks. That's when we implemented a policy: before quoting any project with dimming requirements, verify stock. It sounds like common sense. We'd gone six years without doing it.

Look—I'm not saying every low-cost component fails. I'm not even saying premium supplies never fail; I've expedited enough Mean Well replacements to know they're not magic. But the cheapest option on the purchase order is only cheap if it survives the real installation. Unit price is the tip of the iceberg. Underneath: freight, labor, downtime, and the cost of a phone call like the one at the top of this article.

What I Actually Recommend (Boring on Purpose)

The fix isn't clever. That's the point.

Match the supply to the application.

• LED strips, sign modules, and tape light designed for 24V: a constant-voltage supply. The Mean Well LRS-350-24 is a solid choice for a 350W-class 24V load (street pricing lands roughly between $80 and $110 based on distributor quotes I processed in Q1 2025—verify current rates). Check the derating curve before mounting it somewhere hot.
• Direct LED fixtures with a specified current: a constant-current driver. Match the current to the fixture spec and size the wattage to cover the load.
• Control panels, relays, and automation: DIN rail supplies like Mean Well's HDR series. The form factor is the point.

Read the wiring diagram. The AC input terminals (L, N, ground) and DC output terminals (V+, V-) are clearly marked on an LRS-350-24. The DIM+ and DIM- terminals on a dimmable driver are clearly marked too (mental note: our team should turn this into a public checklist). Spend ten minutes with the datasheet before you hang a ladder.

Run the total cost, not just the unit price. Real talk: a $70 driver isn't expensive when it prevents a $950 re-visit. I now calculate TCO before comparing any vendor quotes—unit price plus freight, installation, risk of failure, and cost of downtime. It takes ten minutes and saves a lot of 2 a.m. calls. If a quote is dramatically lower than the next one, asking why isn't rude. It's due diligence.

So yes, you can replace track lighting heads. And yes, chandelier drivers are worth matching carefully. But next time a light flickers, a fixture dies, or a head fails for the second time in a year, don't just replace the component. Ask what was feeding it. That's where the real problem—and the real cost—was hiding all along.