Cree XR-E tested in integrated sphere.

Erasmus

Flashlight Enthusiast
Joined
Dec 29, 2004
Messages
1,077
Location
Belgium
Cree has published a document in which 2 of their brand new XR-E LEDs are tested according to the NIST (National Institute of Standards and Technology) standards. Tests were run 5 minutes after the LEDs were powered so they are running stable in this test. The LEDs were marked as X1 and W3. I have no idea what this means, as far as I can see no flux bin or tint bin is specified.

Here it is : http://www.cree.com/products/pdf/NIST XLamp LED Document.pdf (PDF file)

Anyway, I found these results in which I am most interested :
- average voltage of 3.183 V at 350 mA (in my opinion a nice low voltage, and according to other CPF members' tests it will not raise as much as a Luxeon LED when running at 700 mA)
- average color temperature of 6045 K
- average CRI index of 78 (in my opinion good enough for home lighting, for which I have planned a project :) )
- average flux of 84.00 lumen ( :rock: )
- average efficacy of 75.4 lumen/W ( :drool: )
- relative spectral power distribution graphs can be seen in the document

Very good results in my opinion! These XR-E's do what was promised by Cree :)

Cheers!
 
Last edited:

BentHeadTX

Flashlight Enthusiast
Joined
Sep 29, 2002
Messages
3,892
Location
A very strange dark place
Erasmus said:
- average voltage of 3.183 V at 350 mA (in my opinion a nice low voltage, and according to other CPF members' tests it will not raise as much as a Luxeon LED when running at 700 mA)
- average color temperature of 6045 K
- average CRI index of 78 (in my opinion good enough for domestical lighting, for which I have planned a project :) )
- average flux of 84.00 lumen ( :rock: )
- average efficacy of 75.4 lumen/W ( :drool: )
- relative spectral power distribution graphs can be seen in the document

Very good results in my opinion! These XR-E's do what was promised by Cree :)

Cheers!

Wonder what the bin codes on those actually were? A 60 watt incan light bulb puts out 860 lumens so an 11 watt 10 XRE array puts out the same amount of light :rock: It would make a great bench light for work on... well, LED mods. Thanks for the info, it pushes the boundries on LEDs and I can see a big jump in LED area lighting.
 

paulr

Flashaholic
Joined
Mar 29, 2003
Messages
10,832
You can get an 13 watt compact fluorescent for about 2 bucks with supposedly the same output as a 60w incan.
 

slvoid

Newly Enlightened
Joined
Oct 29, 2006
Messages
182
Location
Brooklyn, NYC
Heat is the issue, it'll lose a lot of efficiency as it heats up to 11 watts.

BentHeadTX said:
Wonder what the bin codes on those actually were? A 60 watt incan light bulb puts out 860 lumens so an 11 watt 10 XRE array puts out the same amount of light :rock: It would make a great bench light for work on... well, LED mods. Thanks for the info, it pushes the boundries on LEDs and I can see a big jump in LED area lighting.
 

NewBie

*Retired*
Joined
Feb 18, 2004
Messages
4,944
Location
Oregon- United States of America
Actually, if you look at the last page, you will note that the Chief Optical Technology Division Physics Laboratory guy, Abert C. Parr, signed off on it, the testing was done at NIST, by NIST.

Even more impressive, the results were after the LED heated up, and given 5 minutes to stabilize!

Boy, look at that tiny heatsink they mounted it on (meaning, humm, definitely warmed up for the measurement). The face of it measures ~1" (25mm) by 1" (25mm), plus it was mounted on a MCPCB.


I hope this honest way of measuring LED performance continues across the industry, Nichia just pulled another trick out of the rabbit hat, and was recently pulsing their LEDs for a 1% duty cycle- 0.005 seconds (so the die was extra cool during the light pulse measurement) to get the absolutely highest possible numbers they could.

I'm even more impressed that they paid NIST to do the meaurements.


So, what does Philips LumiLEDs do?

Calibration of Luxeon Testers
All electrical/optical test parameters for the Luxeon emitters are specified in the data sheet at a junction temperature of 25°C. Thus, unlike the SuperFlux and SnapLED emitters, Luxeon emitters are tested for instantaneous luminous flux and dominant wavelength (and CCT for white) instead of thermally stabilized luminous flux and dominant wavelength. All Luxeon emitters are tested for luminous flux, dominant wavelength (except white), correlated color temperature (white only), forward voltage, and reverse breakdown voltage using a single 20 ms pulse per test.

The calibration procedure for Luxeon emitters is to measure luminous flux, dominant wavelength (except white) and CCT (white only) in a reference test system at the specified peak current with a pulsed condition with a 1% duty factor at 1 kHz. A number of calibration units are tested that span the expected range of luminous flux, dominant wavelength, and CCT of the particular type of production units to be tested. Finally, these calibration units are tested in the production test system under a pulsed condition of about 20 ms and the production tester readings are made to agree with the reference test system readings.

I've never seen any reference to how LumiLEDs does their official lm/W testing, just this, on the production units.
http://www.lumileds.com/pdfs/AB08.pdf
 
Last edited:

Kinnza

Enlightened
Joined
Nov 15, 2005
Messages
223
Location
Spain
Hope all LEDs manufacturers finish doing the same. Im tired of tricky measurements. The lacking of LED's measurement standards harm the LED market, and difficult its expansion to the general lighting market.

There was some justification for the lack of standards for indicatitors LEDs, but there isnt any to not measure LEDs for general lighting market in IS as any other light product, with the NIST standards. Now is time for a regulation for white LEDs for general lighting applications, obliging to test them in the standard waybefore they can reach the market.

If consumers refuse to buy LEDs not measured across NIST standards, sure they all finish doing it.
 

yaesumofo

Flashlight Enthusiast
Joined
Jul 21, 2003
Messages
3,701
Location
Eastern Pacific, LAX DM03 sw actual
If you read the document carefully you can figure out exactly what the actual bins are. The color tempatures are noted as are the x's and y's in the chart.
I don't have the time at this moment but if you look for those numbers and go get the cree bin chart with the little curved squares those represent a x and y on the grid. that will give you the bin. the lumens per watt is a very interesting number as well.
Waiting for the board experts to find this thread. Because I aint one thats for sure.

Yaesumofo
 

NewBie

*Retired*
Joined
Feb 18, 2004
Messages
4,944
Location
Oregon- United States of America
Kinnza said:
Hope all LEDs manufacturers finish doing the same. Im tired of tricky measurements. The lacking of LED's measurement standards harm the LED market, and difficult its expansion to the general lighting market.

There was some justification for the lack of standards for indicatitors LEDs, but there isnt any to not measure LEDs for general lighting market in IS as any other light product, with the NIST standards. Now is time for a regulation for white LEDs for general lighting applications, obliging to test them in the standard waybefore they can reach the market.

If consumers refuse to buy LEDs not measured across NIST standards, sure they all finish doing it.


I hear you, many are doing the pulse test now to get rediculous claims...
 

cy

Flashaholic
Joined
Dec 20, 2003
Messages
8,186
Location
USA
hince the huge difference between claim output VS out the front end output.
 

65535

Flashlight Enthusiast
Joined
Dec 13, 2006
Messages
3,320
Location
*Out There* (Irvine, CA)
I think third party testing should be done by one lab for every different high powered high efficiency LED, third party tests should be at a junction temp of about 60C where it would lie in most usage, and the same equipment should be used plus batch testing 5 pieces per batch. From different places in the batching.
 
Top