High CRI and its significance

[2xTrinity]

Hi guys,
How about another tangent. I have been meaning to see if I could capture moonlight with my spectrometer and see how it compares to sunlight. My gut tells me that the CCT would be higher but so much depends on the spectral reflectivity of the moon itself and I have no idea what that might be. I assume that IR and UV for the most part are not reflected by the moon but within the visible spectrum, are there some anomalies?

CCT is actually bumped down significantly compared to sunlight. When the moon is highest in the sky it reaches a max CCT of about 4100k, compared to 5800k for highest possible CCT from direct sunlight.

This is easy to check if you can see the moon and a metal halide or mercury vapor street lamp simultaneously (HID vehicle headlights or a 5A binned Cree LED will also work..) -- apparent CCT is basically exactly same.

 

[baterija]

Hi guys,
How about another tangent. I have been meaning to see if I could capture moonlight with my spectrometer and see how it compares to sunlight. My gut tells me that the CCT would be higher but so much depends on the spectral reflectivity of the moon itself and I have no idea what that might be. I assume that IR and UV for the most part are not reflected by the moon but within the visible spectrum, are there some anomalies?

I just did a google on moonlight spectrum hoping to stumble immediately on a spectrum so I wouldn't have to wait for the next full moon. (I was too lazy to attempt a measure of the one that just past). I came across a page that doesn't measure or graph the spectrum but it does show a cool "nerd" project for a simple spectrometer, HERE. While we are installing light triggered needles in our tongue, we could be looking in an old cereal box as well. :nana:

Nice. Pretty soon we'll all be making cereal box spectrometers. I did find this with the following quote:

[SIZE=-1]The reflection factor of the moon is twice as much for the red as for the violet; thus it has a lower color temperature at about 4,125K. The low intensity moonlight appears bluer than it actually is as in accordance with Purkinje's phenomoenon. But moonlight is extremely variable. At full moon, it is 150,000 times weaker than sunlight. Accordingly, exposures must be doubled when the moon is at 45 degrees elevation yet multiplied by seven when it is seven days before and after a full moon.[/SIZE]

[SIZE=-1]
So it looks like moonlight actually has a lower CCT. It will be interesting to see what the actual distribution is like come the next full moon.
[/SIZE]

 

[Kiessling]

I would have thought the moon was way cooler in CCT than the sun. Interesting.

Why do we think of the bluish LED light as "like moonlight" then?

bk

 

[McGizmo]

Interesting! The RGB distribution under various conditions as depicted in that link is also enlightening. The moon may be down in the 4k but it certainly seems that moonlight is cooler than that. I think what may be at play is the nature of ambient light but I am not sure I can explain it. With a point source of light like the sun and even moon, in addition to the direct light if the source is visible and not occluded, we do have significant light that has bounced off the landscape and surrounds. We have a blue sky that brings some light to us I think?!? But the blue sky gives us mostly blue light. I wonder if that big sea of air and some moisture all above us doesn't impact the scene in addition to the direct light from the source itself? I wonder what the color temp of a difuse sky is and I suspect it is higher than the source feeding it, if that makes any sense?

 

[baterija]

It's the Purkinje effect mentioned in the quote. As the light levels get low our eye's peak sensitivity shifts towards the blue end. So even though there is more red relative to the amount of blue the blue color dominates our perception.

 

[Kiessling]

As this is not true for a digicam or for film I suppose ... why don't we recognize a difference there?

 

[baterija]

As this is not true for a digicam or for film I suppose ... why don't we recognize a difference there?

It might be because of white balancing to try and make the image seem more "natural." Looking into the "emitter" in this image doesn't make it seem all that cool/blue.

 

[Kiessling]

I was looking for some pics on real film under moonlilght, unaltered, but I have none.

Oh well ... one learns something new shattering the foundation of one's reality every day.

bernie

 

[McGizmo]

4125k is still quite a bit cooler than incan light and it may well have more blue in it than an incan source. Here is a case where it would be interesting to note the CRI of moonlight and relative to a black body radiator. I still get confused when I see spectrums of various states of daylight and don't see the classic spike at the red end which is the trade mark of tungsten.

Additionally, even with moon light, you have a whole sky that is diverting some of that moonlight as secondary to the source light and that light is likely skewed to the blue side. Below is a relatively long exposure of some moon lit clouds I took prior to an eclipse. Notice the color of the sky.

I have noticed over here that at times the night sky can be seen to have a deep blue VS black color to it.

 

[McGizmo]

To the subject of High CRI and its significance, right or wrong, good chef or bad chef, I really wanted to improve on the illumination over my BBQ as the Cree LED's provided more than ample light but the meats and vegetables on the grill were not rendered well in color, IMHO.

To wit:

The solid copper bar was milled with a pocket to receive a 3x 083 High Cri MCPCB as well as a piece of half round optic.

I used some really slick Viega copper crimp connectors and a length of copper tubing to complete the fixture. The optic is bedded in a clear UV cure epoxy and the Viega fittings are plumbing fittings so the LEDs are well sealed from the elements:

I contemplated taking a shot of the color checker on the grill under daylight as well as at night under the Nichias but I couldn't find the motivation energy required.

The optic is great at concentrating the light on the BBQ and its immediate surrounds.

 

[js]

4125k is still quite a bit cooler than incan light and it may well have more blue in it than an incan source. Here is a case where it would be interesting to note the CRI of moonlight and relative to a black body radiator. I still get confused when I see spectrums of various states of daylight and don't see the classic spike at the red end which is the trade mark of tungsten.

Additionally, even with moon light, you have a whole sky that is diverting some of that moonlight as secondary to the source light and that light is likely skewed to the blue side. Below is a relatively long exposure of some moon lit clouds I took prior to an eclipse. Notice the color of the sky.

I have noticed over here that at times the night sky can be seen to have a deep blue VS black color to it.

God Don! I just love this picture! Nice.

And really cool 3x083 setup in your post above mine!

 

[McGizmo]

Thanks Jim. It is kind of a trip to see stars in a blue sky.

 

[Kiessling]

Stars! oo:

I thought that would be dirt on the lens

 

[griff]

To the subject of High CRI and its significance, right or wrong, good chef or bad chef, I really wanted to improve on the illumination over my BBQ as the Cree LED's provided more than ample light but the meats and vegetables on the grill were not rendered well in color, IMHO.

To wit:

The solid copper bar was milled with a pocket to receive a 3x 083 High Cri MCPCB as well as a piece of half round optic.

I used some really slick Viega copper crimp connectors and a length of copper tubing to complete the fixture. The optic is bedded in a clear UV cure epoxy and the Viega fittings are plumbing fittings so the LEDs are well sealed from the elements:

I contemplated taking a shot of the color checker on the grill under daylight as well as at night under the Nichias but I couldn't find the motivation energy required.

The optic is great at concentrating the light on the BBQ and its immediate surrounds.

Very nice Don!

 

[2xTrinity]

Here is a case where it would be interesting to note the CRI of moonlight and relative to a black body radiator. I still get confused when I see spectrums of various states of daylight and don't see the classic spike at the red end which is the trade mark of tungsten.

It's very possible that the reflectance curve of the moon may be like a CTO filter, that is, a color temp stepdown filter. This is a filter that reflects red almost completely, then as the wavelengths decrease, reflects less and less. As long as this reflectance is a particular smooth, continuous curve and the reference light source (the sun) has blackbody spectrum, the output will simply correspond to a different blackbody color temp. However, I have no idea if this is actually true in the case of the moon.

Interesting! The RGB distribution under various conditions as depicted in that link is also enlightening. The moon may be down in the 4k but it certainly seems that moonlight is cooler than that. I think what may be at play is the nature of ambient light but I am not sure I can explain it. With a point source of light like the sun and even moon, in addition to the direct light if the source is visible and not occluded, we do have significant light that has bounced off the landscape and surrounds. We have a blue sky that brings some light to us I think?!? But the blue sky gives us mostly blue light. I wonder if that big sea of air and some moisture all above us doesn't impact the scene in addition to the direct light from the source itself? I wonder what the color temp of a difuse sky is and I suspect it is higher than the source feeding it, if that makes any sense?

The sky tends to scatter blue light that is already present. In the case of the moon, very little blue light is reflected, thus there is not much blue left that can be scattered by the atmosphere.

That said, I propose an experiment for you to easily 'test' the Purkinje effect described earlier -- that is, our eyes' sensitivity curve actually changes in low-level lighting conditoins -- so we are MORE sensitive to blue. I notice this with my LEDs at 4000k in my LF2x -- at night time on very low setting (say 0.2% output, similar illuminance to moonlight) the 4000k will appear very cool. However, if i crank on my 4xT8 shoplight with 4000k lights in it, all of a sudden that looks a lot warmer. Yet, if I look at my flashlight and the T8s side-by-side, they are identical. The difference is that it is the daylight (photopic) sensitivity curve that is now at play.

I wonder what the color temp of a difuse sky is and I suspect it is higher than the source feeding it, if that makes any sense?

Diffuse sky is basically more intense at very short wavelength. I've seen values ranging anywhere from 8,000k to infinite K*.

*Color temp does have an abrupt limit, as seen in the graph above. Any higher temperature beyond about 15000k will not change the apparent color, just the ratio of UV/visible will go up.

 

[gewoodfo]

Hi All:

I am new to this forum. I am so glad that i have found this. After reading through a couple of the threads, I firmly believe that some of you on this thread will be able to help me with identifying a unique light/fixture for a project that I am working on.

I am trying to design a mobile table top studio for taking high resolution photos of coins and other small objects. When I say small mobile, I mean that the entire studio including camera, laptop, lights, camera stand, illuinated copy stand, power, cables, diffuser tent and blackout tent will all fit into a case that will meet the most stringent airline carry on baggage restrictions in regards to size and weight.

With this, one of my biggest challenges is finding lights that were small, lightweight, low heat output, and have high photographic qualities ( full spectrum natural daylight color temperature of 5000 - 5600 kevin and a high CRI index of 90+.

I have been looking for LEDs lights and have not been able to find any that have the desired photographic and natural daylight spectrum qualities.

I see this kit needed several differnet type of lights that meet thesee specificaltion. some for the base illuminated copy stand, some general floods to cover up to a 11x17 inch area ( Note: this will probably be smaller in size when I get around to making it all fit into a case.) and some spot lights for getting the light close to the coins when using a macro lens with a digital SLR or feature on a Point and shoot digital camera.

I should say that I am not a hands on type of guys nor real technical, so please forgive me if I appear to be asking redundant or trying to accomplish something that is not technically feasible or cost fesible at this time.

Thanks,

G.E.

 

[Kiessling]

HI G. E: and welcome ! :wave:

I think this question will do better with its own thread, for example in the "General" forum. It will get buried here.

Feel free to start a thread there.



bernie

 

[ElectronGuru]

I did try out the white balance eye dropper in PhotoShop on a photographed white surface I captured just today underwater:

Daylight white balance set on the Nikon:

And then the eyedropper white balance in PhotoShop was directed to sample the bright white in the whale's pectoral fin:

There are variations of this same approach. Taking another shot from earlier:

Here is the 'auto' button from Photoshop/levels:

There are probably 1000 variations of light color underwater, and ways to cope with them on a standard basis. Artistically, the trick would be getting the color more accurate, while still leaving enough of the original tint to feel natural.

 

[McGizmo]

Electron Guru,
If you are trying to shame me into going back and color correcting all my old images, fat chance! :nana:

You are right of course and since I upgraded to PhotoShop CS4, I have been tweaking my under water images when they appeared too far off from what I recalled seeing in person. I could go out with a white card and go for a manual white balance adjustment on the camera to deal with the lighting at the time but it varies considerably depending on where the sun is relative to you and the subject. I agree with your take on simply working the image after the fact, in photo shop. Within reason, I can treat most of the images from a shoot with the same color temp and tint adjustment.

A white card is not white under water but it is whiter to the eye than the camera renders.

 

[ElectronGuru]

As coincidence would have it, I was talking with a local diver not two hours ago about this.

Showing him the ExpoDisc, it might be possible to bring some form of 'white balance disc' into the water as a faster/easier alternative to grey cards. Assuming its effective (and you can push the needed buttons on the camera), it would not only be more accurate PS tweaking, but much less work. The disc would live outside the dive housing and you'd just put it in front of the sheltered lens to calibrate.

I'm all just guessing here. I've been a shooter for many years and used to swim, but I've never done both.