All Blogs

LED L70 & Lumen Maintenance Explained: How to Verify a Fixture's Lifetime Before You Buy

Why "50,000 Hours" Is Not a Lifetime Promise

Nearly every LED datasheet carries a number like 30,000, 50,000, or 100,000 hours. Most buyers treat that figure as "when the light dies." That is the first mistake, and it is an expensive one.

That number is not a time-to-failure. It is a lumen maintenance figure — the point at which the fixture's light output has faded to a stated percentage of what it produced when new. The LEDs are usually still glowing long after that; they are just dimmer. A fixture that reaches L70 at 50,000 hours is not "broken" at 50,000 hours. It is simply delivering 70% of its original light by then.

For a B2B buyer, understanding the difference is the difference between a project that meets its illuminance targets for a decade and one that quietly goes dim — and fails a re-certification walkthrough — after three years.

The Core Vocabulary: L70, L80, L90

Lumen maintenance is written as an "L" value: the time (in hours) at which output has depreciated to a given percentage of its initial value.

  • L90 — output has dropped to 90% of initial. A very strict threshold, used for spaces where light levels are critical and tightly specified.
  • L80 — output has dropped to 80%. A common commercial specification, because 20% fade is where most people begin to notice.
  • L70 — output has dropped to 70%. The industry's default lifetime figure, and the one most quoted in marketing.

So "L70 = 50,000 h" means: after 50,000 hours of operation, the fixture will still produce 70% of its original light — if it was tested at a temperature that matches your installation.

The Statistical Side: B10, B50, and the Fine Print

Lumen maintenance is also qualified by a "B" value, which tells you how many fixtures in a population are expected to fail before the rated point.

  • B50 / L70 — 50% of fixtures are allowed to fall below 70% output. This is the loosest, most common marketing figure.
  • B10 / L70 — only 10% of fixtures may fall below 70% output. This is what a serious specification should ask for.

A datasheet that says simply "50,000 hours" without an "L" and a "B" value is telling you almost nothing. Ask for the full "B10 / L70" or "B50 / L70" notation.

The Two Reports That Make a Claim Real

A lifetime number is only credible if it is backed by two documents from recognized test laboratories.

LM-80 is the test method. It measures the lumen maintenance of the LED package (or module) over a minimum of 6,000 hours, at two or three controlled case temperatures — most commonly 55°C, 85°C, and 105°C. The output is a set of data, not a lifetime number.

TM-21 is the calculation that turns LM-80 data into a projected lifetime. It extrapolates the measured depreciation curve forward, with a hard rule that the projection may not exceed six times the actual test duration. A 6,000-hour test can therefore support at most a 36,000-hour claim. If a supplier quotes 100,000 hours, ask how many test hours sit behind it.

The Temperature That Actually Kills Your Fixture

Heat, not the LED chip itself, is what shortens LED life. Two temperature points matter:

  • Junction temperature — the temperature inside the LED die. Every 10°C of added junction heat can cut lifetime roughly in half.
  • Ambient temperature — the air around the fixture. The LM-80 figure is only valid for the test temperature, so a fixture rated at 85°C case temperature will behave very differently in a hot, enclosed high-bay or a sun-baked outdoor wall pack than in an air-conditioned office.

This is the single most common source of disappointed buyers: a panel light tested at 25°C ambient, installed in a sealed recessed housing at 40°C, and fading years ahead of the datasheet. Always match the application temperature to the test temperature.

The Driver Usually Dies First

LEDs fade; drivers fail. A fixture's rated life is only as strong as its weakest component, and that is almost always the driver — specifically its electrolytic capacitors.

Electrolytic capacitors dry out with heat and time, and a failed capacitor usually means the driver — and therefore the whole fixture — stops working long before the LEDs dim to L70. When comparing suppliers, ask three things about the driver:

  1. The driver's own rated life in hours (a good driver spec is 50,000 h at its maximum case temperature).
  1. The capacitor brand and its temperature rating (105°C-rated capacitors outlast 85°C-rated ones).
  1. Whether the driver is serviceable or replaceable, which matters more for hard-to-access installations.

Warranty vs. Rated Life: Read the Gap

Rated life and warranty are two different promises. A fixture can be marketed at L70 = 50,000 hours while carrying only a 3-year warranty. That gap is where the manufacturer's real confidence shows.

A 5-year warranty on a fixture in continuous operation covers roughly 43,800 hours — close to the full L70 window. A supplier offering a 5-year warranty is, in effect, betting its margin that the product survives its own rated life. Treat warranty length, not the L70 number, as the stronger signal of build quality.



The Bottom Line

A lifetime claim without an "L" value, a "B" value, a test temperature, and an LM-80/TM-21 report is just marketing. The buyers who avoid surprises are the ones who ask for those four things before the PO is signed — and who remember that the driver, not the LED, is usually the first component to go.

At Solarix Lighting, we back our commercial, rail-transit, and industrial fixtures with LM-80 and TM-21 documentation, state lifetime in full B/L notation at realistic operating temperatures, and pair every LED with a driver sized and temperature-rated for the application. Send us your ambient conditions and required illuminance, and we will quote a fixture whose rated life matches the way it will actually be installed.

LED IP Rating Guide: IP65 vs IP66 Explained for Outdoor, Tunnel & Industrial Lighting
What Is UGR? A Plain-Language Guide to Glare Rating in Commercial Lighting