higher the CRI, higher the red region?

Hi guys,

I am wondering, looking at the XM-l2.. comparing the CRI, does it mean the higher the CRI, the more red the led will output? I notice there datasheet is very vague, and doesn't show a good graph depicting more detail.

I noticed there cobbs, the higher the CRI, the more red region it goes into, so I thought it would be the same for XM-L2? ( didn't know if I can link the datasheet here? )

if so, is it also why they put out less light, as the reds are harder for them to increase power in???


anyone have a spectrophotometer? I have chips someone can have if they want to test them ( 3000k, and 2700k XM-l2 ) ( there on sinkpads )



Thanks to anyone, I always appreciate it

Hello,

CRI is measured using a set of test colors. Standard leds (blue led with a phosphor layer) are pretty ok with most of the other test colors except red, so, you could say that adding more (and deeper) red makes CRI better. Check this picture for a better reference (it is a high cri Nichia and the effects of different test colors are also given).

Adding more red doesn't decrease the led output (except for some little phosphor losses) if you measure the output in watts. If you measure it in lumens, then it does. This is because lumens are measured agains human eye sensitivity, and eyes are not very sensitive to red. If you had a 100% efficient green led, it would give you more than 600 lumens, because green is the color that eyes are most sensitive to. If you had a 100% efficient red led, it would give less than 100 lumens. Both of them would still be 100% efficient.

You might want tot check Wikipedia for more information about perception, cri, lumen etc., as well as the dedicated threads here in CPF.

thanks guys..

thanks esko for explaining that. I needed to take into account comparing lumens vs. watts


now, only if I could get my hands on a spectrophotometer to see exact wavelengths..

What data do you need as far as the spectrum and what do mean by "exact wavelengths"? If you know the LED you can always look at the manufacturer specification. There is typically but not always a graph of the typical spectrum profile for a color temperature range (cool, warm, etc) but not specific CCT and CRI. Close enough?
A used USB spectrophotometer goes for about $1K on eBay. Here's an Ocean Optics currently listed: http://www.ebay.com/itm/Ocean-Optic...d=100011&prg=1005&rk=1&rkt=5&sd=251333350432&

Also, if you are looking for a specific spectrum you might request it here and someone might have it on file or be able to get it.

harold, thanks.

looking for a XM-l2 CRI 90 ( 2700k ).

what else do I need to get that spectrophotometer running?

Here is a long and interesting thread on "CRI of White LEDs".

http://www.candlepowerforums.com/vb/showthread.php?273878-CRI-of-White-LEDs&highlight=high+cri

Some where else I read that High CRI means, take % black body radiator at 3000K of the emitted spectrum. So a High CRI of 82 for 3000K means the LED emits 82% of the black body spectrum at 3000K.

pepperdust said:

harold, thanks.

looking for a XM-l2 CRI 90 ( 2700k ).

what else do I need to get that spectrophotometer running?

Click to expand...

No problem. We just bought the same unit from this seller. I'll check with the engineer setting up the equipment and post what has been done. As far as getting a spectrum we don't use that XM-L2 but I can check into getting a couple. I should have the data in my library for modeling just to be current anyway.

Esko said:

CRI is measured using a set of test colors.

Click to expand...

I would add that CRI is merely an average of scoring for 8 color samples named R1, R2...R8. (The simplicity of the system is why CRI is often criticized). There are 6 (7 in Asia) additional special color samples from R9-R15 which are not used in the CRI calculation but are important scores to look at when determining light fidelity as many manufacturers deliberately leave these values out. For example, take a look at Beijing Yuji's high CRI LEDs and how they have high R9-R15 values as well.

To answer the original question, in a high CRI LED, sufficient amounts of each CRI test sample color must be present in the emission spectrum. Since the red color sample (R9) is not included in the R1-R8 color samples, in theory, a manufacturer could get away with producing a high CRI LED that nonetheless renders reds poorly.

pepperdust - having talked with the engineer that is setting up our spectrophotometer like the one I linked the suggestion I would make is to contact the seller, ask what is included in the sale, get the s/n, and then call Ocean Optics support. They can tell you the current configuration and anything that you will need to purchase separately from that. We already use their equipment and so have the software installed and have fibers on hand, etc.

I'll look into getting an XM-L2 next week and post an update when I have one.

nanakuli said:

For example, take a look at Beijing Yuji's high CRI LEDs and how they have high R9-R15 values as well.

Click to expand...

I like what they've done with the backgrounds on that page RE: scrolling, clever and effective, can't say I've seen it done before (as an ex- site developer)

i guess i need to look at the spd of modern high cri led's
seems most white led's are also poor in the cyan area.

well, according to the definition of the CRI, if you prefer to emphasize a particular colour, like the red, you should use low color temperature light source

Look to the Film and Lighting industry for your High CRI references. Ocean optics is one of the best companies for photo spectometry equipment. Check out Mole Richardson they have the best CRI ratings for their lights. their graphs are on their site.