Don't Be LED Down The Garden Path (Part 2)


Hearing a lot about "white" LEDs? So are we. And some of what we're hearing just isn't true. Especially LED claims about UV and IR. So, lets look at some actual LED test data and separate the science from the science fiction.

We started this series with a spectral power distribution for a typical "white" LED (just the center of the graph above). Typical "white" LEDs have a blue spike at 450nm and a lower phosphor curve centered around 560nm. Phosphors, manufacturers, beams and color temperatures may vary slightly. But, the basic curves remain the same.

The problem is, lighting manufacturers tend to limit their data to the visible range. Until recently UV and IR were not considered problems. Then they either extrapolate (believe without testing) that the data continues along the same curves, or if they know better, they let you believe that the data continues along the same curves.

The fact is that it doesn't. The low points at the end of the visible range are not the tail ends of the LED spectral outputs, they are simply low points (valleys) in the data. You need to know that significant UV and IR emissions continue past the data points most manufacturers show you.

All LEDs emit UV. The graph above is what the full spectral power distribution of a "white" LED looks like. Notice that the UV output matches the intensity of the LED visible light output. Notice, too, that the IR output matches that of a steam radiator.

To be fair, we found no measurable UV in the long wave 300-380nm range. That this is the only UV that most museum quality UV meters measure may just be a coincidence. Regardless, there is big difference between no long wave UV and no UV at all. (That's why the government sets UVB filtering standards for sunglasses.)

The photo above shows a UVX Radiometer using a 200nm-300nm head (UVB and UVC) in the actual testing of a major brand "white" LED luminaire. The meter shows a UV output of 3.8"W/cm2 for their "cool white" LEDS. The spectral power distribution from the manufacturer's website shows peak output for this LED to be roughly the same intensity, 3.7"W/ cm2. Their "warm white" LEDs show even worse results, a peak output of 1.9"W/cm2 and a short wave UV output of 2.9"W/cm2. The bottom line is that these fixtures put out as much UV as they do blue light.

All LEDs also emit IR. As a matter of fact, heat dissipation is a major factor in LED design and LED life. While manufacturers may say stupid things like, "all thermal energy is conducted through the housing and not radiated in the beam," the fact is that these LED luminaires stabilize around 50°F above ambient temperature. The whole unit radiates IR. No object that radiates heat can be considered IR free.

Despite some manufacturer claims, LED sources are not UV and IR free. Their high short wave UV output makes them particularly dangerous for art and artifacts. Without significant secondary UV filtering LEDs are not acceptable light sources for fugitive or fragile materials and do not meet IESNA guidelines for museum lighting.