Zebralight MKIII projecting image of emitter dome ???

Does anyone have any advice for me on this issue? In a recent purchase of a Zebralight MKIII Cool white I have a distinct image of what appears to be the Yellow Dome from the emitter projected. If you shine the light on a white wall from 15 feet to 25 feet you can see it distinctly.

HERE is a dropbox link to a JPG and a MOVIE of this I took with my SONY NEX7 camera.. It is very difficult to see in the photograph, a little easier to see in the MOVIE, but it is VERY apparent to the naked eye when viewing the projected beam on a wall.

IS this artifact just a reality of these high power domed emitters?

Have we read of another member(s) who also had this defect? I think it needs to be returned, in less than 30 days from purchase I believe. Contact ZebraLight would be my advice.

I definitely see it in the photo and video. I had both a CW and NW SC600 III and neither had that spot in the middle.

I haven't seen this topic pop up in a while but it's not new if I understand what I'm seeing correctly. First off, I don't have a Zebralight MKIII CW at hand but looking for specs I gather that it's using a Cree XML2 and a CPC optic. Is that right? If so then I would be surprised if you didn't see a yellow dot in the projected image. You can probably do a search on this forum or on the internet at large for color separation over viewing angle for Cree XML and you will find a lot of information about what you are seeing. To be brief what happens is the blue photons emitted from (about) 45-90 degrees to the surface of the LED die travel a shorter distance through the phosphor layer and are a much cooler white having passed through less phosphor. Most of the cool white light passes right out of the end of the CPC without being refocused (some is but not much). The blue photons that are emitted at an angle lower than (about) 45 degrees will pass through more phosphor making the down-converted light more yellow / warm. The low angle light is focused by the CPC to the center of the beam making the yellow dot you see at a distance. You don't see it as much close up because the light is pretty well integrated there. The CCT of an LED is measured in an integrating sphere. To get color over angle requires a goniometer. I did take a look at the available files for the XML2 and they are monochromatic so obviously there's a problem with the files. I've probably made it less clear by trying to be brief. If I can find a couple of white papers to link I will do so.

First thing on a Google search is a Candle Power thread: http://www.candlepowerforums.com/vb/showthread.php?309724-XM-L-Color-Separation

There's a white paper and discussion in the thread and I blather on in that one too....

Harold_B said:

...I gather that it's using a Cree XML2 and a CPC optic. Is that right....

Click to expand...

Thanks Harold for your detailed information. I suspected that as you suggest that there is this Yellow dot artifact as a result of the physics and the angles of refection involved in the optics. Designing around this is the job of the engineers at Zebralight. I had hoped that they got it right, but it looks like maybe they 'hit the brick wall' on that part of the design. The emitter is actually a new Cree product, the XHP-35.. This is part of a whole new series of high output emitters from Cree. They offer it with this dome and also without. I have one of Zebralights MKIIIs that uses the de-deomed version of the XHP-35 in a 'netural white' tint and it DOES NOT show any of this type of artifact. I think it is fair to say that the artifact is a direct result of the domed device sitting where it is in the path of the optics. Humm... bummer ! I have written Zebralight to get their input and am waiting to hear their take on it.

Also it is worth noting that I have a MKII that uses the XM-L2 (domed) in cool with tint and it DOES NOT show this issue. I think there is getting to be a number of engineering trade offs between high lumen output and secondary un-intended artifacts. It's the old 90-10 rule. It takes 90 percent of the design and fabrication effort and cost to get the last 10 percent of the performance.

Harold_B said:

...
First thing on a Google search is a Candle Power thread: http://www.candlepowerforums.com/vb/showthread.php?309724-XM-L-Color-Separation

There's a white paper and discussion in the thread and I blather on in that one too....

Click to expand...

Thanks for that excellent link Harold... I did not realize that this subject had been dealt with back in 2011 in such depth.. Unfortunately many of the link to import photographs are now dead, but the discussion is still first rate.. I was particularly stuck by one posters observation that "... Yep, seen it on multiple lights. Once again Cree sacrifices beam quality for maximum output. I suppose most people don't care, and compared to a Maglite beam, anything is better, including the old 5mm LEDs we used to use back at the turn of the century..."

As far as my current issue with the specific light I have in hand, well.. I think I will get an RMA from Zebra and just let the thing go. For some purely personal quirk in my brain I just can not flick this light on without all my attention turning to the Yellow artifact. Talk about ANNOYING !!!

It is your money and you should be satasfied with your purchase that is for certain. I have no idea how the engineers at Zebralight go about designing their reflectors or if they outsource that part of the project but designing for color uniformity from a specific LED source isn't trivial. it is more of a challenge the tighter the desired beam angle and the larger the die area. It's also a possibility that the CPC was intended for several LEDs and each has its own compromise.

Harold_B said:

..The CCT of an LED is measured in an integrating sphere. To get color over angle requires a goniometer.....

Click to expand...

With the aid of my iPad and little program available for it that uses the camera sensor (search "LUX" on the iStore...) I decided to try a little "hip pocket science" and projected the beam on a flat wall then "swept" the reflected light with the sensor of the iPad bolding it off 3 inches from the wall. I was able to detect a drop of about 120K in the zone that my naked eye saw as the "Yellow Zone" ,,.. I then "swept" several other flashlight samples, including the Zebralight MKIII HI version of the same light and also their H32 cool white, .. both those lights do not exhibit that variance in the K across the width of the beam.

All this proves, of course absolutely nothing! My efforts are not the result of lab grade instrumentation. But it does bring up a couple of points. One is that the Yellow Zone on the light in question is, in some sense, verifiable, and not a total figment of my poor eye-sight.

It also seems to suggest that other similar commercial examples of these flashlights seem to exhibit rather uniform color temp across their beam width. The most important point of all this is that it would suggest that it would be very possible for a standardized lab-grade test be performed to record the color map across the beam width of these high end flashlights and that information be delivered to the prospective buyer of these lights. After all, we are talking about +$100 pocket lights. This is rare air in the consumer world. Seems that those of us that are willing to spend that type of cash deserve that data. There is the ANSI/NEMA FL-1, maybe this should be in FL-2

I think what you are asking for is reasonable although I'd imagine most manufacturers would have to send their products to a lab for certification. Major manufacturers likely have similar equipment but these start at around the $70K mark give or take a bit depending on options. Your iPad app is not going to be real accurate (I've used a bunch of them before breaking down and buying a handheld spectrometer) but it will give you a more objective measure than simply viewing the spot so that was a good idea for a quick confirmation.

http://www.radiantvisionsystems.com/products/imaging-goniometers

Well I located this following interesting thread on this subject so guess I am not the only horse in the race !

http://www.candlepowerforums.com/vb...rd-Yellow-Center-Spot-Normal&highlight=yellow

I have obtained an RMA from Zebralight and will post the light to them today. I will report back in this thread in a week or two when they offer me a replacement.

PeterRamish said:

..
I have obtained an RMA from Zebralight and will post the light to them today. I will report back in this thread in a week or two when they offer me a replacement.

Click to expand...

OK,.. I said I would report back/... Zebralight replaced my Yellow Do-nut with a very nice emitter sample that they found in the "binning" Five Gold Stars to Zebralight ! They are a responsible and reliable company that stands behind their product.. I can recommend purchases from them without hesitation. They made good on a bad sample that crept into the supply chain.

Let discuss a little about how this sample wound up in the production Zebra MKIII that I was shipped. The question "How can this happen?" is a valid one.

I am going to dodge that question here by just giving you a reading list that explains it all. (homework my teachers used to call it.. :thinking

What if I told you the Yellow Dot is in some samples of MKIII CW because:
"No LED manufacturer can produce uniform color points in their white LEDs; rather, they produce batches of LEDs with varying distributions of color and flux to create inventory based on the results of the production."

If you think I just made that sentence up, then think again. It is Crees own words from the following: Cree Document : LED Color Mixing: Basics and Background: CREE Document:CLD-AP38 REV 1

So Here's the reading List:

1.) Essential reading and you can find it (the PDF) HERE or HERE
2.) Other Essential reading on all this is:
and, of course:

3.) Essential Reading: http://www.cree.com/LED-Components-and-Modules/Document-Library
...the Specific Fact Sheets and Product Pages for many Cree Lamps you find in your lights.

deleted..