DIY High powered Studio LED project.

[Alexander Sutton Hough]

Hi, I am in the middle of my own DIY project making a high-powered LED light. I am using the Luminus XNOVA-Studio Series 150w LED COB. I chose this one because it could handle high heat, is meant for use in video and photography lighting, outputs a little less than a 650w tungsten bulb and the company rates it at 5600K 95CRI. But now that I have built it I am very disappointed with the actual CRI. It's way too green and feels very lacking in the color rendition.

 

I was hoping since Luminus is an American company in San Francisco and this line of LEDs are meant for Fresnel and studio lights their CRI claims would be more honest.

 

Does anyone know a good supplier of 150w LED COB with an high CRI, maybe even ones that major brands are using?

 

[Phil Rhodes]

Google for Yuji LED. I've only seen one of their 100w modules, but it seemed okay and measured well on a Sekonic colour meter. They also offer a 500w module, which I haven't seen. Naturally it'll require a more concerted cooling effort.

[Akos Baranya]

I have a batch of these LEDs, how hot is yours running? I noticed that it is decidedly less green when running at higher temperatures.

 

(not just me, the spec sheet also has a "Change in CIEx vs. Junction Temperature" graph showing this behaviour)

 

I measured mine with a color meter and it seemed to have a nice TLCI of 96 , the CRI came out to 92, a bit of a disappointment compared to the specs.

 

I'm also interested if anybody has experience with alternative chips.

 

This was the reading I got (the test was with a just turned on chip, no warming allowed, so this is the worst it performs):

 

CCT : 5751 K
CRI(Ra) : 92
CQS : 92
Illuminance : 4709 lux
Foot Candle : 437,5 fc
CIE1931 x : 0,3268
CIE1931 y : 0,3432
CIE1976 u' : 0,2022
CIE1976 v' : 0,4778
λp : 445 nm
TLCI(Qa) : 96,3
GAI : 93,1

 

here are the pics:

http://imgur.com/a/SI1XI

Edited February 6, 2017 by Akos Baranya

[Phil Rhodes]

That R9 number is a bit low.

 

I had the Yuji 100W chip as a demo, and never got around to doing anything with it. Probably I should.

 

P

[Stuart Allman]

Alexander, Phil,

 

I created a DIY LED light with the Yuji violet based LEDs so I could get better color rendering. Unfortunately I haven't had access to a Sekonic C-700 meter to take CRI measurements yet. The light definitely does not look green, but it's hard to judge how good the light quality is without a proper measurement.

 

Here's a link to my project.

http://illuma.blogspot.com/2017/01/diy-led-light-with-high-cri.html

 

I kind of wish I had made the bulbs bi-color, but oh well...

 

Stuart Allman

---------------------------

illuma.blogspot.com

[Phil Rhodes]

Ah, you've done better there - they don't make the 100W modules I saw with the violet-based technology. What you've made is essentially state of the art, although I suspect most (all) LED manufacturers will still try to tell you they've got some sort of secret sauce. In the LED emitters. That they don't make.

 

Teehee.

[Stuart Allman]

Phil,

 

Well...I've run into plenty of LED vendors for which that is true. However, there are some folks out there like Cineo who do have custom formulations for their remote phosphor panels. The company's original came out of Intematix, which specializes in remote phosphor. I also know other companies that buy off the shelf since they don't have funding for custom R&D.

 

I produce gear reviews for Video Gear and we try to actually measure the fixture's real performance when we have access to a C-700 spectrometer. Unfortunately we haven't had the funds to invest in one for ourselves yet. I had a borrowed unit two years ago at NAB and it was easy to tell which fixtures were engineered for color and which ones were engineered purely for brightness. I miss that meter - excellent product.

 

Stuart Allman

--------------------------

illuma.blogspot.com

[Phil Rhodes]

Yes, I'm in exactly the same position. Had one on review, then they wanted it back!

 

To be fair I was never that impressed with the C700. It was just the only thing available, though there are now alternatives. Never had that much confidence in its results.

 

P

[Alexander Sutton Hough]

Has anyone tried this on led from CREE? http://www.cree.com/~/media/Files/Cree/LED-Components-and-Modules/XLamp/Data-and-Binning/ds-CXA2.pdf

[Alexander Sutton Hough]

Has anyone tried this on led from CREE? http://www.cree.com/~/media/Files/Cree/LED-Components-and-Modules/XLamp/Data-and-Binning/ds-CXA2.pdf

Well I got a hold of one tonight and very impressive. I compared it to my Kino Diva 55k, Quasar Science Par lights and defiantly 93 CRI and up. It claims 95 CRI and I wouldn't be surprised if that were true. I don't have a SpectroMaster so I am comparing other light sources that I know the true CRI from online testing. My camera in light room with a datacolor test chart 5500k with -1/+1 green and magenta and skin tones are looking good.

 

I need some bowens mount accessories such as Fresnel, a soft box and take it out in the field.

 

I don't plan on using this professionally but I had to build it since its been bugging me how expensive LED are when this projects only cost me few hundred with off the shelf parts and I am not even engineer, (though I was raised by one so that helps).