Flashlights, Reflectors, and XR-E

NewBie

*Retired*
Joined
Feb 18, 2004
Messages
4,944
Location
Oregon- United States of America
I grabbed a few reflectors of various profiles and did some testing.

I found the shallow modified reflectors cause an issue with a distinctly colored ring when projected on a white wall.

I found that deeper reflectors, that come up tighter to the CREE XR-E have much less, if any tinted ring effect.

Then I reversed the situation to demonstrate where the rings are comming from.

The setup:
refcree.jpg


What you see is a flashlight in the path and angle of where I saw the maximum tinted ring effect.

Then I turn on the flashlight, and bring the paper in to block the direct light, so it has to hit just the reflector, and you then see where the light is comming from:
refcree2.jpg


With deep reflectors of a narrow diameter, I don't see this, so it might be something useful to consider.


Here is another shot, which should show things a little better, it appears that on the shallower reflectors, there is an angle that hits the inside of ring wall, and then is reflected down upon the phosphor surface.

refcree3.jpg



Another item to look at is the light cast out the side:

creeangd.jpg



And another set of photos to look at:

creeanga.jpg




When I get the production Seoul P4 parts, I'll try and get a similar set of examinations, so folks that want to fiddle with this stuff can see what is going on there.
 
Last edited:

IsaacHayes

Flashlight Enthusiast
Joined
Jan 30, 2003
Messages
5,876
Location
Missouri
I noticed the same thing with the light from the reflector shining into the led, but never thought of studying the reflector from it in relation to focus of the led!
 

Norm

Retired Administrator
Joined
Jun 13, 2006
Messages
9,512
Location
Australia
It all makes perfect sense now, the light source is below the ring causing the ring to throw a shadow. Well done.
Norm
 

McGizmo

Flashaholic
Joined
May 1, 2002
Messages
17,290
Location
Maui
I have a HD45 head with three Cree 7090 LED's I got from Newbie. They are driven in series at 270 mA and the long exposure beam shot below doesn't catch the dynamic change in the colors which are all over the place. The shot is a blending of tints over time:

fountain-beam.jpg


:thinking:
.

.
.
.
.
.
.
:nana:

There is some water on the window that is part of the problem and the other part is the LED's are red, blue and green.

fountain-watercolor.jpg


Now this light is conceivably capable of better white light than many but geting the blend is a bit tough.
 

Raybo

Enlightened
Joined
Apr 27, 2004
Messages
477
Location
Illinois
McGizmo said:
I have a HD45 head with three Cree 7090 LED's I got from Newbie. They are driven in series at 270 mA and the long exposure beam shot below doesn't catch the dynamic change in the colors which are all over the place. The shot is a blending of tints over time:

fountain-beam.jpg


:thinking:
.

.
.
.
.
.
.
:nana:

There is some water on the window that is part of the problem and the other part is the LED's are red, blue and green.

fountain-watercolor.jpg


Now this light is conceivably capable of better white light than many but geting the blend is a bit tough.

How did you do that last photo Don?

:poke:

Very nice, NewBie and yourself are very efficient at photography.

Ray
 

McGizmo

Flashaholic
Joined
May 1, 2002
Messages
17,290
Location
Maui
Dang! I just lost a post I had a lot of time into. :(

Serves me right for throwing in this OT beam shot and post here. :eek:

The second image is a reasonably quick shutter shot of an arch of water from a fountain pump spilling onto the window of the HD45 head. This assembly is part of a rainbow fountain I got around to making today. With all the discussions on tints and disparities in beams as seen on a white wall, I found it interesting to contrast this with a blatantly colored flashlight beam.

In the image below, you can see the subject better.

fountain-components.jpg


The fountain in plenty of ambient light:

fountain-indoor-rainbow.jpg


What I find fascinating is the fact that the light show on the ceiling is at such a rapid pace in changing colors that you can't catch much of it with your eyes; forget about any camera image. You have the red, green and blue streams of photons directed upwards by McR-17XR reflectors but the streams are in constant alteration due to the everchanging optic provided by the water stream. The light show on the ceiling has any color imaginable flashing here and there and dancing to the tune of the water. You see white when the colors all coinside but mostly you see colors.

However, the sum of the various colors bouncing off the ceiling in turn illuminate indirectly the surrounding area in a bath of rather consistant warm white light! With a riot of colors above, I could photograph an object below that is being illuminated by these colored LED's and you would be certain I had used a white source of light for illumination. And I did, sort of. :thinking:

What a static white wall beam shot can show is is often not something we can perceive in a dynamic use of a flashlight where the beam is moving and painting the target area with brush strokes of light and your eyes are moving about as well. With a brush stroke of photons over an object, perhaps a yellow corona or ring in the beam might enhance the perception of reds within the object wheras a beam missing this ring of yellow would be less effective?

Our visual perception uses integration of infirmation as well as differentiation of information.

With colored LED's we may be able to mix in a blend of full spectrum light that comes closer to natural light than any other source we have seen before.

Newbie has done a lot of work with RGB I believe and he can shed more light on this.

If I was using a LED to illuminate a white wall, I would much rather its tint be uniform throughout than having rings of color. If that light were used while in motion, its tint artifacts would likely be much less of an issue or concern.

Enough of this ramble.... :eek:
 

Concept

Flashlight Enthusiast
Joined
Sep 1, 2006
Messages
1,403
Location
Townsville, Australia.
Wow now you make wicked multicoloured light fountains too! :)

So how do you mix the colours to create white? Some type of prismatic diffuser?
Could you break the colours into many separate beams and then recollimate them in another reflector?
 

McGizmo

Flashaholic
Joined
May 1, 2002
Messages
17,290
Location
Maui
Concept,

I am no expert on this stuff but do know how to play with it.:eek: I would think you could take a prisim or defraction grating and break out the various wave lengths of light from a white source to view its spectral composition. This is what an integrating sphere does with a difraction grating sampling the light. You can recombine the spectrum to get the white. In the fountain sample above, I am using reflectors on each LED and the combined beam has artifacts of tint and intensities but for the most part the combined beam from these is a pink tinted white. There are 3 and 4 channel LED drivers now that will allow you to control the current to each of the RGB or RGGB LED's so you can dial in any resultant color (from integration) you want. I was in a hurry to finish this project up so I didn't experiment with the light head very much. I suspect a sheet of LSD film over the window might mix the light suficiently to provide a homogeneous distribution of the combined light in a beam. I didn't want the light combined in this fountain as I wanted the water optic to have access to the raw output from the three LED's. I would guess that the red should be driven at a slightly lower current than the blue and green if I wanted a better white when all three color streams overlap.

RGB clusters in flashlights are probably not that far off. WIth 3 channel control, you can dial in the color you want. You could add UV and IR LED's to the cluster and have full spectrum available. Instead of filtering out light you don't need, you dial in just that which you do need! Integration and mixing of the various photon streams will be an interesting challenge optically.

The giant video screens on buildings and in ball parks are RGB LED sources I believe now? I have heard that ColorKinetics has "owned" the RGB technology to some extent but this whole field is growing and we will see it in more subtle use as time goes by I would suspect. With RGB, there are no tint issues like with the white LED's. You will be able to control the tint.
 

mahoney

Enlightened
Joined
Jan 7, 2002
Messages
603
Color Kinetics does not own the concept of RGB color mixxing, but they did manage to patent (and legally prevail in enforcing that patent) using PWM to control color changing RGB (and CYM, or any others) LED light sources.

This was despite considerable prior art and I predict that this will stifle development of color changing lighting in the USA to some extent, and open the way for inovation from manufacturers in other countries.

There is some research that indicates that you can get a much wider range of colors and a higher quality white light by starting with 7 colors rather than three when mixing white from LEDs. If anyone's really interested I'll dig out the reference.
 

Concept

Flashlight Enthusiast
Joined
Sep 1, 2006
Messages
1,403
Location
Townsville, Australia.
In regards to the large RGB blue screens I have seen one where each pixel is represented by 1 red, 1 green and 2 blue 5mm leds. Interesting to say the least. I think they were then broken up into panels with about 18 x 24 pixels in each and then these panels were about 12 wide & 12 high to make up the entire screen. Nothing but multicoloured dots up close but from about 80 meters away it looked great. Whole great big mess of wires behind linking each panel to various "black boxes" and a PC.
I could never afford a house big enough to fit that sucker in! :)
 

NewBie

*Retired*
Joined
Feb 18, 2004
Messages
4,944
Location
Oregon- United States of America
Avago (was HP/Agilent) makes a chipset with integrated sensor for sampling the light and PWM adjusting RGB LED drive to maintain a somewhat constant white point. Last I looked they had the price down to 40 dollars.
http://www.avagotech.com/products/product-detail.jsp?navId=H0,C1,C5231,C5889,C5949,C5749,P96305


They also offer a decent line source module of R, G, B, not all that efficient, but it is pre-made, so you just bolt it in:
http://www.avagotech.com/products/highlight-detail.jsp?navId=H0,C1,C5231,C4941,C1201&id=1201
 
Top