I'm trying to build a light fixture for my aquarium, but I'm having difficulty understanding the various way manufacturers represent light output.

I'm trying to compare the output of CREEXPE-ROY-3 and CREEXTE-ROY-3 and I am just completely lost.

How can I figure out the lumen output of these LEDs? Assume they are both driven at 350mA (per LED).


Luminous Flux is the measurement you're looking for, and is measured in lumens. Looking at the pages you've linked:

Tables showing Cree XP-E Royal Blue has a luminous flux of 425, while Cree XT-E Royal Blue has a luminous flux of 550

The XP-E shows a luminous flux of only 425, while the XT-E has a luminous flux of 550. Luminous Flux is measured using an integrating sphere, so it is the actual total light output of the device under test. A given light, of course, may appear brighter than another with higher flux depending on optics and observer location, but as far as measuring the total light output, luminous flux is the measurement to use.

So to answer your question specifically, the XT-E is capable of more light output than the XP-E. This also matches well with current consumption, where the XT-E consumes 1.5A at its rated output, while the XP-E consumes 1A at its rated output.

I expect that both would have similar brightness at the same current, so if you were to feed them 350mA(ie, about 116mA per discrete LED in the module) then you can probably expect to see very similar light output between the two modules. If you're going to run them under current though, I'd suggest simply buying lower light output LEDs - they'll be much less expensive.

If you can explain what you're trying to optimize - be it brightness, efficiency, cost, etc then we might be able to provide better guidance.

  • \$\begingroup\$ Can you please post a link to the datasheet where you found these numbers? I couldn't find it \$\endgroup\$ – HL-SDK Feb 28 '14 at 17:39
  • \$\begingroup\$ My point is: How can the flux be about 4x greater than the white LEDs? It doesn't seem reasonable to me. \$\endgroup\$ – HL-SDK Feb 28 '14 at 17:41
  • \$\begingroup\$ The data is from the webpage, and it looks like the royal blue LEDs are being rated using "radiant flux" while the white LEDs are being rated by "luminous flux". The difference between the two is that radiant flux measures all the radiant energy output, visible or not, while luminous flux only measures for the light output within the range of human vision. \$\endgroup\$ – Adam Davis Feb 28 '14 at 18:23
  • \$\begingroup\$ The datasheet I'm looking at is cree.com/~/media/Files/Cree/LED%20Components%20and%20Modules/… \$\endgroup\$ – Adam Davis Feb 28 '14 at 18:24
  • \$\begingroup\$ You might want to ask "Why is the flux of this blue LED significantly greater than the flux of this white LED?" as a separate question, rather than trying to combine all your questions into this one post. \$\endgroup\$ – Adam Davis Feb 28 '14 at 18:25

Between 62lm and 114lm per led. I could not find numbers for this array module.

Source: page 3 of the datasheet

The max efficacy of that series is 148lm/W at 85 degrees C. If you put in 1 watt, you should get 148 lumens out. Half a watt? Half the lumen output. This is dependent on at least temperature and chromaticity.

  • \$\begingroup\$ How are you getting that from the data sheet? The data sheet doesn't list the lm for the Royal Blue. Also, that's only for the XPE (I think), what about the XTE? \$\endgroup\$ – Malfist Jan 29 '14 at 15:26
  • \$\begingroup\$ Looks like the royal blue is specced differently (in mw instead of lumens). You might need a different datasheet or contact Cree directly for that information. \$\endgroup\$ – HL-SDK Jan 29 '14 at 17:19
  • \$\begingroup\$ Right, which was why I was asking how to get lumens from mw. \$\endgroup\$ – Malfist Jan 29 '14 at 17:55
  • \$\begingroup\$ You multiply it by the efficacy which should be specified in lumens per watt. These LEDs seem to be around 100lm/W \$\endgroup\$ – HL-SDK Jan 29 '14 at 18:57
  • \$\begingroup\$ I can't find the lumens per watt anywhere though... \$\endgroup\$ – Malfist Jan 29 '14 at 19:55

A Luminosity Function [1] is used to convert from radiant flux to luminous flux for a given wavelength. You can convert a spectrum diagram from radiant flux to luminosity, incrementally taking points on the curve, multiplying each point with the corresponding value from the luminosity function and integrating the interpolated data over the wavelengths. The Luminosity Function has its peak in the green band, because that appears brightest to the human eye, so a blue LED has much less lumen than a green one of the same radiant flux.

Depending on your goal, lumen is not an appropriate measure for the light. As noted in a comment, lumen is an indicator for perceived brightness, not actual brightness (radiation power). The meassure used in Biology is Photoactive Radiation (PAR) and only found in datasheets for lights advertised specifically for horticulture. Plants don't care for green, they cannot absorb it. That's why they appear green. Blue LEDs still seems to be easier to manufacture, because generally LEDs with blue or white with high blue output are the best in comparison for efficiency and cost. a white led has less peak radiant flux per wavelength than a single color of the same wattage, less dramatic difference in overall efficiency. thats just something to keep in mind when mixing white and color LEDs.

Interesting enough, white LEDs are far from useless for plant growth. a "white" LED emits a discontinuous spectrum that is mostly red and blue, more so if the Color Rendering Index is low.

[1] https://en.wikipedia.org/wiki/Luminosity_function


Answering only part of the question, and 6 months after it was asked, as the answer was not clear overall. Reading the answers and comments the information is provided but is easily missed.

Higher power LEDs at the blue (short wavelength) end of the visible spectrum are not rated in lumens per Watt (or in lumens) but are instead rated in mW (milliWatts) of light output. This is because lumens are weighted according to the sensitivity of the average human eye, and eye sensitivity is very low towards the blue end of the spectrum. While results in lumen could be provided they are so low that two LEDs with significant differences in output may have ratings different only by a few lumens. As people are used to lumen differences for LEDs with lumen outputs of around 50-150 lumens/Watt, the very low absolute lumen figures would produce comparative results which SEEMED incorrect.

So, the various comments talking about typical lumen/Watt figures for the LEDs concerned do not apply at the blue end of the spectrum.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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