L70: Can we accurately predict L70?

By Wade Johnson     16 Sep 2019     Reading Time: 3 minutes


This is Part 3 in a 4-part series.

The short answer is that we can’t. Worse is we don’t even fully know the variance. The good news is that LEDs will last longer than traditional light sources but nobody knows exactly how long.

Making this more difficult is that the industry standard for predicting the decay of LED past their “tested” lifetime doesn’t consider aging factors in the rest of the system, such as lens and reflector losses over time.

L70 prediction is based on test data from a specified amount of time (usually 6,000 to 10,000 hours) and multiplying the result by 6. Why 6? I haven’t been able to track that down. I am sure this is buried in some DOE report somewhere, but it is probably outdated like all the other methods. LED technology moves too fast.



It is not practical to test to the full expected lifetime of the fixture, as that test data would be technologically irrelevant before the test was even concluded.


In 20 years in the lighting industry, I have gained a lot of respect for the ability of engineers to meet product requirements and continually decrease cost. If any LED chip engineer is given a scope of work that calls for the least expensive chip that is efficient enough to be qualified for DLC Premium and provide 10,000 hours of data, how much effort and expense will be put into making sure the LED continues to produce the same light output for longer than 10,000 hours? I am concerned that LED chip vendors have designed the perfect 10,000-hour chip that fools these L70 predictions.

Emergency LightingIf this is the case, or will be the case in the future, emergency egress light levels may be at significant risk. Most facility managers know that they need to push the button on exit signs once a year to make sure the batteries are still working but very few are maintaining a 1 foot-candle path of egress. With traditional light sources this was not a concern because routine lamp replacement would increase the light level near the original design parameters. With LED, this is not the case, as explained in Part 1 of this blog series.

Frankly, to most purchasers of LED lighting, this doesn’t matter. What matters is when the purchaser will need to budget to replace the lighting. To know the truth, you have to look at LED light fixture warranty language. Most manufacturers warrant the percentage of LEDs that are out but not the amount of light they will continuously produce. As the LED’s age, they will continue to consume the same amount of energy but produce less light each hour. Therefore, the lumens per watt listed on the spec sheet may only be valid for the first day.

The only solution I can see to this dilemma is to require someone other than the end user to pay for more fixtures and the cost to install them if they do not live up to their required function of providing a specific amount of light. Only then will the claims on the spec sheet equal the belief of the manufacturer.

Don’t lose sight of the reason you are purchasing lighting. Require a specified amount of light for a specified amount of time. A ten-year warranty is not guaranteeing ten years of adequate light.


Finally, we need to consider one last question

  • When LED lights reach L70, what do we do with them? (Part 4)
    Available end of September
Wade Johnson
Wade Johnson
Sr. Business Development Manager
Flex Lighting Solutions