High CRI and its significance

[yaesumofo]

One other quick question or two.
does CRI correspond directly to daylight color balance?

Can filters make for better CRI numbers?

It seems to me that an emitter which is filterd so the color tempature is corrected looses some of it's OOMPH.
It seems that an emitter which comes "pre corrected" (for lack of a better term) comes having lost some of it's OOMPH to begin with.

I was just wondering if the loss of lumens due to filtering is similar to the amount of loss (if you want to call it that) that the hi CRI emitters come with when compared to non high CRI emitters?
If this doesn't make any sence please just ignore.
Thank you.
Yaesumofo

 

[McGizmo]

Yaesumfo,
I believe the Nichia you have is a precursor to the High CRI and it may well have good CRI. Does it glow after you turn off the light?

I believe the phosphor used by the mfgs in their warm white and high CRI LED's is less efficient in converting the blue from the die to the additional spectrums ultimately emitted.

This may not make sense or hold water but a blue tinted LED for instance will have its way on other colors and upset the balance but if an object is red, it is the presence or absence of red light that strikes it that likely has the most effect on us determining it is red. Blue is absorbed by this object and excessive blue is just absorbed that much more?!? :thinking: :shrug:

 

[McGizmo]

Hi guys,
I decided to take some pics to see if they illustrate a difference in color rendition. It was an informative process for me in and of itself. I set my Nikon D70 white balance on "sunlight". I let the shutterspeed vary as the camera sought a reasonable exposure. These comparisons are not about relative intensity or brightness of the different light but about the color captured.

I have a backpack that is a fluorescent light green and strange purple nylon. I have noticed that this purple is all over the place under different light sources; a "problem purple".

Camera on tripod and shot in the garage, first up is the SunDrop:

Now on my computer screen, these colors all look good and the garage door in the background is about the proper tint of white.

Next up is the same scene as illuminated by a Seoul High CRI P4 that is in a custom A2 dumb head on L1 with F04 beam shaper (a mouthful in itself):

I think the colors here are also good but the purple is not quite "right" and the wall tint is a bit off. Incidentally, to my eye, there was much more of a warm tint and awareness with the Seoul than the Nichia in the SunDrop but I think this would be "filtered out" by the brain in no time.

Next up is the scene as illuminated by a Mule sporting a Cree CR-E of high color temp:

Now the lime green or whatever it would be called has the best "snap" in this image and the reds and orange look good but the purple looks blue to me now and not right.

Finally, the same objects illuminated outside in indirect sunlight and little shaddow:

Now we have some potential problems here. For starters, I noticed while viewing these images on my camera display that the daylight image of the purple was simply blue!! Now on my laptop display it does seem more true to life but somewhat washed out.

The camera makes color corrections and the software on my laptop effects the color as well as the LCD monitor itself. The compression software has its way with these colors as well. I have no idea if you will be seeing the same colors on you end as I see here while composing this. Our eyes and brains are different too.

I can state that to my eye at the time of taking the shots and even now while viewing the images in this composition that the Nichia High CRI LED in the SunDrop provided the best color rendition to my eyes and was most in harmony with what I see viewing the same objects under sunlight. Is this significant? Well we all can identify the back pak and light blue shop towel and botlte of tide and red cup in all of these images. The only false perception I had during the process and now in viewing the images is in regards to the problem purple. I would not identify it as purple under the Cree LED either in person or by the image I now view. I might guess it is purple if a hint were provided.

I would guess that those who love incandescent light sources probably have problems in proper identification of colors in the violet to green range. Those who prefer LED light sources probably have problems in the greens and reds. Are they serious problems though? If color recognition is critical then color rendition is likely important. Duh.

 

[McGizmo]

Icebreak,
Thanks for your comments and observations. One of the "testers" of one of these High CRI LED's is babydoc. I feel real bad that I requested he "sit" on his enthusiasm and findings for so long. I provided him with an Aleph McMule with a proto Nichia LE and knowing that he needed some condensing of the flood, I also sent along a 1" diameter half ball (acrylic) in a F04 beam shaper boot so he could adjust a secondary optic in front of the source and get some collimation. Hopefully he is still around and will be willing to chime in at some point.

The SunDrop addresses the issue of partial collimation or condensing of light for greater intensity by virtue of the sapphire lens. The lens is 15mm in max diameter and the convex surface has a 25mm diameter or 12.5mm radius. I realized that some folks might want further collimation and that was one of the reasons I had the relative deep grooves machined into the head. In addition to hand grip, the grooves provide a concentric grip for any boot that might be installed. Below is the proto with a F04 boot holding a 1" half ball against the front of the head:

To try to get an illustration of what this second optic does, I took a couple quick and uninspired beam shots in the garage. The light is about a meter from the wall. First up is the SunDrop with only the sapphire lens and you have seen that this produces a sharp circle of flood. In the image and due to the focal length of the lens, the perimeter of the beam circle is beyond the frame:

You can see the profile of the light sitting on a step stool at the left. I think this image also shows that even though the SunDrop has a sharp end to its beam, it is still a wide flood.

Below, I added the second half ball lens and camera, light and object are all in the same position:

What you can't really see in the shot is the fact that the spot beam is somewhat oblong due to the 6 dice of the LED (2x3). The intense portion of the spot inside of the warm corona is roughly 6"x7" at a distance of 1 foot. There are obvious artifacts at the edges of the beam but I believe the utility of the heart of the beam as well as its distribution of intensity is quite acceptable. I also think this dual lens combo is reasonable in terms of light loss. Reflection loss aside, the brunt of the light produced does hit the first lens and its resultant beam does hit the rear of the secondary optic. You can pull the secondary optic further from the light to the point you have a sharp and ugly projection of the 6 dice.

On the other hand, you can use the LSD disk from the F04 behind the 1" 1/2 ball and sandwiched against the front face of the SunDrop and get a smooth soft white spot beam that I think is worthy of a white wall (about 1 foot away):

 

[BigHonu]

Aside from the shop towel, the Nichia is spectacular!

It doesn't overstate the neons (green on back pack and yellow on the Tide label) like LEDs would normally do, and your concrete floor retains its warmth without leaning too much on the yellow.

Seoul is nice, but too yellow compared to the Nichia.

Would be interesting to see some dark greens with black mixed in. If the Nichia can do that well, then this LED is a WINNER all the way!

At what distance was the light to the objects?

 

[McGizmo]

BigHonu,
There is black and a dark green on the FLIR camera in the last post I made. The wider flood shot is a bit underexposed relative to the FLIR camera but in the bottom image, you can see the green and black in the button on top and flare of light off the black rubber on the side com port cover of the FLIR.

EDIT: Your shop towel comment had me thinking as it really looks a bit overstated in the daylight image. SInce the daylight image is taken in the shade and the sun not apparent in the sky above, I wonder if there might not be more emphasis on blue than otherwise? (those wayward blue photons). By the same token, the green in the pack and the purple look a bit washed out in the daylight shot. It's funny but we have items which we typically see under the same light source be it outside and natural light or inside and always artificial light of some sort. With indirect sunlight entering my house during the day, my pack is always purple as I walk by it during the day. With LED illumination in the house at night, it is a blue pack. I just remembered I need to get off the computer here and get something in the mail to you today! :nana:

 

[BigHonu]

BigHonu,
There is black and a dark green on the FLIR camera in the last post I made. The wider flood shot is a bit underexposed relative to the FLIR camera but in the bottom image, you can see the green and black in the button on top and flare of light off the black rubber on the side com port cover of the FLIR.

Saw that right after I posted up.

Thanks! That is the type of separation I was looking for.

So, I need to ask because I'm pretty sure you already have done it:

What is the Nichia like in the Lunasol 20 head, and more importantly, can you find 3mm LEDs that can complement the Nichia?

 

[McGizmo]

Check my edit, leave me alone p) and the image of the 083 is too large to be captured well in the LS20 as I recall......

Aside from warm white 3 mm's I don't believe the high CRI phosphors have been used in the lamp style LED's to my knowledge. I just got some 3 mm samples from a company in Taiwan which are "golden" white and Warm white but haven't taken them out of the bag.

As _js_ has commented elsewhere, it seems that ample flux from these newer LED's comes a long way in allowing for reasonable color recognition. You take too great of a loss in flux going to High CRI and you are stepping backwards a bit perhaps if you don't deliver enough intensity to target?!? :thinking: :wave:

 

[BigHonu]

I had a feeling that the total flux may be a limiting factor in that application. Bummer that the beam can't be collimated well with the LS20...at this point.

Edit checked...and unmounting as I type...

 

[yaesumofo]

No glow. Color rendition is still pretty good. Very neutral when compared to other emitters if that makes sense.
I think I like the idea of the sundrop and the use of the 083 emitters for photographic work. I imagine that macro work of nature would benefit from the use of them to light. I have a NIKON LED ring light which does a very good job of allowing natural colors to come out. This is OLD technology compared to the 083's. I would bet that they are some form of pre pre nichia emitters.
I like where this is going.
first perfect the hardware next prefect the light then put them together.
Yaesumofo

Yaesumofo,
I believe the Nichia you have is a precursor to the High CRI and it may well have good CRI. Does it glow after you turn off the light?

I believe the phosphor used by the mfgs in their warm white and high CRI LED's is less efficient in converting the blue from the die to the additional spectrums ultimately emitted.

This may not make sense or hold water but a blue tinted LED for instance will have its way on other colors and upset the balance but if an object is red, it is the presence or absence of red light that strikes it that likely has the most effect on us determining it is red. Blue is absorbed by this object and excessive blue is just absorbed that much more?!? :thinking: :shrug:

 

[Icebreak]

McGizmo -

Still ahead of me but I'll still attempt to be mildly contributive.

I think it is the ratio of blue in a beam of light that plays havoc with red tints. Blue absorption will produce dark color, or black meaning added slate color in the return image. Before your beamshots (pain in the neck [empathetic regards]) were posted I was going to say it would seem that the blue spike seen in the spectrum of the Nichia 083 would not be enough to cause a rendition problem when gauged against all that pretty 620nm/660nm. Remember the blue fin in your Surefire A2 spectrum shot? Not a problem.

Let's look at blood. The white envelope is a little blue because of the blue security pattern on the inside. The poster board is a creamy white. The objects are red ink, a bike hook, a drop of mostly dry blood and a drop of wet blood.

I'm still learning this point and shoot Canon SD 850 so sorry about the out of focus. Here's an unnecessary shot where the illumination is household incandescent diffused and bounced:

Here's direct sunlight through a window.

Here's an incandescent Welch Allyn 1499:

Here's a Malkoff Q5 using a Cree clear optic: I have another shot around here somewhere that shows the Q5 doing a little better.

Let's detune the color by moving the the beam to the side and invading it with a little 5mm blue Nichia:

Let's tune the color by moving the beam to the side and invading it with a little 5mm red Nichia:

When I look at the red cup in your shot using the Nichia 083 it looks just right. It looks just a tiny bit better than the outside shot which I believe you've correctly identified as being due to the light source being a blue sky.

Side note: I like the looks of the Sundrop quite a bit.

Back to collimation.

Wow. That's just really nice light management. Sure, there is a little separation of a small portion of dark red and blue but it appears that majority of the original spectrum is in the main beam. For image return, the important part, I'm still guessing that I'd be able to recognize Willy the water snake 30 feet away.

That baby doc is enthusiastic speaks volumes.

This is really exciting.

 

[BabyDoc]

Icebreak,
Thanks for your comments and observations. One of the "testers" of one of these High CRI LED's is babydoc. I feel real bad that I requested he "sit" on his enthusiasm and findings for so long. I provided him with an Aleph McMule with a proto Nichia LE and knowing that he needed some condensing of the flood, I also sent along a 1" diameter half ball (acrylic) in a F04 beam shaper boot so he could adjust a secondary optic in front of the source and get some collimation. Hopefully he is still around and will be willing to chime in at some point.

Yes, I am still around, and I am relieved that I can finally talk about the Nichia HIGH CRI light that Don sent me. Let me back up a bit and tell everyone that I am no color or lighting expert. However, as a pediatrician I saw the potential that LED's had as an intense and efficient examination light source, but was very disappointed when I actually tried using a variety of available Cree and Rebell lights. All of them, no matter how warm the tint, failed to show the subtle shade differences in the pink and red region which would allow me to properly diagnose pharyngititis, and gingival disease. In many instances these less accurate LEDs normalized abnormalities I might have seen had I used an incandescent source with a more balanced color spectrum. I had started a discussion on the forum a few months ago about this disappointment, and Don came to my rescue and sent me his prototype Nichia HIGH CRI light. The only problem is I was silenced until now to tell anyone about it and how wonderful this light is.

This light has been everything I could have hoped for in terms of color rendition for my medical purposes, but I might add it has advantages , too, as a general use light. Articificial light is supposed to illuminate the real world in a real way, and not make the world look artificial. Would anybody want a high definition TV set with inaccurate colors? I doubt it once they have experienced the real thing. Until now, LED flashlight people have tended to focus their attention on lumens without worrying about color except perhaps for the tint preferences in the color of the beam when shined upon a white wall. The first attempts with big TV's were disappointing with large pictures, with poor color and definition. Still, people bought them. With newer technologies, color accuracy and definition improved, and so did the user and his appreciation for those advancements. I think we are at the same stage with LED technology. Like big screens, we have big enough light producers for most people's needs. They might not think they need or even want accurate color rendition in their lights, but once they experience the difference with these HIGH CRI LED's, they won't want to turn back the clock.

 

[Icebreak]

Hello, BabyDoc.

I have a few questions. The spectrograph shows some good production in the blue area peaking at 450nm. It seems to be more usefully balanced than other white LEDs we are used too. For dark blue and dark purple/blue veins, does this seem to help? I'm guessing with the strong yet nicely sloping reds even into and past the dark 750nm frequencies that rendition is excellent which helps definition as well. This following question may have much to do with perception but here it is. Would you say that this Nichia 083 might actually be better for your purposes than traditional, high quality incandescent light you've experienced?

Is it possible that all my questions could be answered by the following coupled with your High Def television metaphore?

This light has been everything I could have hoped for in terms of color rendition for my medical purposes, but I might add it has advantages , too, as a general use light. Articificial light is supposed to illuminate the real world in a real way, and not make the world look artificial

Thanks for checking out McGizmo's prototype. As I've mentioned in the other thread, that not only helps his project but helps the rest of us as well as we watch the advancement of lighting technology.

 

[Edwood]

Wow. That is impressive color rendition. Now I really really want some Nichia 083's for a Video Light. Maybe mod a Zebralight H30 or something for a vertical orientation. Or find a way to strap a Sundrop or two to my Camcorder.

 

[BabyDoc]

Hello, BabyDoc.

I have a few questions. The spectrograph shows some good production in the blue area peaking at 450nm. It seems to be more usefully balanced than other white LEDs we are used too. For dark blue and dark purple/blue veins, does this seem to help? I'm guessing with the strong yet nicely sloping reds even into and past the dark 750nm frequencies that rendition is excellent which helps definition as well. This following question may have much to do with perception but here it is. Would you say that this Nichia 083 might actually be better for your purposes than traditional, high quality incandescent light you've experienced?

Is it possible that all my questions could be answered by the following coupled with your High Def television metaphore?



Thanks for checking out McGizmo's prototype. As I've mentioned in the other thread, that not only helps his project but helps the rest of us as well as we watch the advancement of lighting technology.

It was actually an honor for me to be involved with Don's (McGizmo's) prototype trials. He seemed to be one of the few people here who heard my concerns about color while most others couldn't understand the fuss. (They reminded me of people decades ago with their black and white TV's who couldn't imagine why you had to have a color TV.)

Let me try and answer your questions. Being a pediatrician, I don't very often need to use the light to look at blue veins. I, would imagine though that this light will do very well with them. As far as the question of whether I prefer this HIGH CRI LED over my traditional medical incadescent lights, I would say a resounding YES!
Color aside, the bigest advantage of the LED light for me as a pediatrician is the bright and even illumination of the oral cavity. The average physician either uses a simple incadescent flashlight or uses a Welch Allen otoscope to look at the mouth. Either of these devices have too narrow a hotspot and a less intense spill area, making it necessary to reaim the light a few times as you look around the light looking for details. Young children are not often cooperative enough to let you take your time to do this oral examination leisurely. With the fisheye lens Don fitted my LED light with, and the resulting wide hotspot, I get instant even illumination of the mouth with just one aim. I can take a single mental snapshot of the entire mouth with the LED light, rather than try and piece together several images as I reaim the less intense incadescent light when the child gives me the time to do so. Truthfully, however, you can not put color accuracy aside. Without it, no matter how bright the image of the mouth is, all the subtle abnormalities get washed out if the light source isn't balanced in its color spectrum. (The problem is analagous to a stereo system where the listener has turned up the bass to the point he can't hear the triangles any more. The trebble is no longer audible being drowned out by the bass.) Traditional LED's emphasize blue to the point it blocks out your ability to see the shades of red that are there. What brought this problem home to me was a child I examined using a SF L1. While that beam was a nice wide hotspot, it failed to show the subtle injection of the child's tonsils even though I could see white exudate on the tonsils. Had there not been exudate on the tonsils, I might have called the throat normal. Seeing exudate and no redness made me question the light source. Sure enough, using an incadescent, the redness of the throat lit up. Don's HIGH CRI light also showed the redness (pinkness) of the throat. I have been more than satisfied with Don's light and had little need to go back to my Welch Allen until recently when I attempted to fix a loose contact in the light and damaged the switch trying to do so. It isn't easy going back to the the Welch Allen otoscope. Not only do I miss all of the advantages I have enumerted of the HIGH CRI LED, but I now see that incadescent lights actually give a yellow tint to everything, including pink mucous membranes. This presents no problems, but its just funny how I never noticed this until I used a more color neutral LED.

 

[cy]

sure wish Newbie was still around... he had access to some of the nicest spectrometers around....

Again, I don't consider myself expert or even proficient with the spectrometer and its software so these graphs can't be taken as rock solid.

 

[Icebreak]

Terrifically helpful and clear report, Doctor. Thank you.

 

[Erasmus]

I'm also doing some projects with warm white XR-E's. Very nice warm white tint, and quite bright at 83lm@350mA each. Yesterday I quickly made a 'portable desk lamp' with 4 of these LEDs, total power consumption of 5W, comparable to output of a 40W incandescent. On 9AA NiMH's it gives me plenty of runtime to study outside in the garden during the night More info on this later today, when my neighbour returns home and I can borrow his camera

Here I am again I just took some pictures of my lamp :

The portable LED desk lamp :






Comparison of beamshots (note : all shots were taken at the same manual setting on the digital camera) :



Feel free to comment and discuss on this light @ https://www.candlepowerforums.com/threads/199248 Thanks!

These warm white LEDs open a whole new world of applications and LED projects!

 

[mejesster]

I know this is distracting from the high CRI LEDs coming out, but I thought I'd add a bit of info from a different discipline.
It's fairly common practice in stage lighting to use lights of two colors, one warm and one cool, often amber and blue, to balance color rendition. Has anyone considered or tried multi-LED/color filtering solutions to get better color reproduction?

 

[KrisP]

I'm using two Cree P3(or P4?) 7A tint with one Cree Q2 WG as three 7A's were too warm and three Q2 WG are too cool. I haven't done any CRI tests but to my eyes colours look more accurate and vibrant.