3
\$\begingroup\$

I recently purchased a 365 nm UV LED, and set up the proper circuit. When I turn the power on, the LED lights up very dimly. Is this white light (that's the color I saw) being emitted from the LED? If so, are there any methods (preferably inexpensive) for blocking out only visible light?

The specs for the UV LED can be found here: http://www.mouser.com/ds/2/228/LZ1-00UV00-257812.pdf

\$\endgroup\$
3
  • 3
    \$\begingroup\$ ... How are you observing this, your eyes may perceive some UV, it will appear with a strange purple tint in some cases that may appear whitish. It is not a good idea to look at it directly, even when diffuse but especially if its narrow angle. The datasheet shows a relatively narrow band that does not bleed significantly into visible. If the spectrum is appropriate for you I wouldn't worry about it \$\endgroup\$
    – crasic
    Commented Jan 28, 2016 at 6:56
  • \$\begingroup\$ I was looking it from an angle wearing UV-protective glasses. Is that okay safety-wise? It was for no more than a second. \$\endgroup\$
    – ArKi
    Commented Jan 28, 2016 at 7:06
  • \$\begingroup\$ Most glass will block UV adequately, but the color you observe may be due to a variety of reasons. Ideally you could characterize a spectrum, a crude way to do this is to use a prism and see if you can see some visible lines on dark paper. You can compare with a white LED to get a rough estimate of how likely it is that you have visible frequency modes \$\endgroup\$
    – crasic
    Commented Jan 28, 2016 at 7:24

1 Answer 1

6
\$\begingroup\$

Figure 5 of the linked datasheet shows there is a small amount of emission in the >380 nm range.

I digitized the spectrum and computed how much relative radiant flux there is greater than 380 nm.

\begin{gather} \Phi_\text{visible} = \frac{\int_\text{380 nm}^{\infty} \Phi_{\lambda} \text{d} \lambda}{\int_{0}^{\infty} \Phi_{\lambda} \text{d} \lambda} \approx 0.087 \end{gather}

So a bit under 10 % of the radiant flux is being emitted in the visible spectrum. I don't know what color this would look like in practice, though I suspect it might be have blue-violet hue since about 88 % of this is in the 380–400 nm range. Assuming 1050 mW total radiant flux, this means there's about 92 mW emitted in the visible spectrum.

You can find "reasonably priced" (~30 USD) shortpass and bandpass filters which transmit UV while blocking visible wavelengths, for example this one. This would reduce the visible radiant flux to about 10–30 mW (eyeball estimated). If you have more money available, there are of course better filters.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.