LED Light Pads Panels

[augusthouse]

I like these guys (in a smaller form).

They are most likely more practical on this scale (as in image). There are a number of ways to achieve diffused lighting in photomacrography; but from an application perspective I found them interesting.

The Kelvin temp is up there with flash - which in concentrated doses could be used for highlights via FO cable(s) if applicable.

Just thinking aloud, thought you might find it interesting.

P.S. The one under the icecream is not turned on (unfortunately).

Some more info at the links below.



http://www.rosco.com/us/video/litepad.asp

http://www.luminousfilm.com/led.htm

Craig

[Charles Krebs]

These are interesting. I'm still not convinced about the color quality of LED illumination for photography. I haven't checked the more recent products, but the ones I looked at a few years ago had a discontinuous spectrum. You could probably balance for them OK, but it may not be that straightforward to augment it with electronic flash. Not sure.

[augusthouse]

These products are somewhat affordable. The Sydney office of Rosco have sent through a price list and some details (retail AUD $90.00 including transformer for the 3" x 3") . The same size is US$57.50 from Adorama.

The 'cookie jar' is empty at the moment, but when able I'll buy one of the smaller versions (3" x 3") and have a closer look.

Craig

[cannyman]

These products are somewhat affordable. The Sydney office of Rosco have sent through a price list and some details (retail AUD $90.00 including transformer for the 3" x 3") . The same size is US$57.50 from Adorama.

The 'cookie jar' is empty at the moment, but when able I'll buy one of the smaller versions (3" x 3") and have a closer look.

Craig

There seems to be a great deal of interest in these superbright LED`s. I have seen a few,but I am not impressed. They seem to show just a pool of light on a dark surface. Using my front flash close macro I am working around F27-F32. This gives me a good DOF.

[augusthouse]

A couple of Rosco LED Lite Pads have crossed my path and I am scratching my head as to what I can do with them to explore their usefulness and/or potential.

I have two of the 6" x 3" versions.

Any ideas?

Craig

[Harold Gough]

These are interesting. I'm still not convinced about the color quality of LED illumination for photography. I haven't checked the more recent products, but the ones I looked at a few years ago had a discontinuous spectrum. You could probably balance for them OK, but it may not be that straightforward to augment it with electronic flash. Not sure.

It sounds like colour correction filter(s) would be the only answer at the studio stage.

No doubt, you digital guys could do manipulation wizardry with histograms, curves or whatever.

Harold

[mgoodm3]

I have some LED lights and they do have a bit of a color shift even with a good WB setting. Should be able to be corrected a little more thoroughly in the color correction portion of Adobe's RAW editor - which can be automated with a plug-in.

All lights that use phosphors tend to be a bit rough on the spectrum - LEDs, fluorescents.

[rjlittlefield]

Craig, see http://www.photomacrography.net/forum/v ... php?t=5810 for discussion of "metamerism" and the variety of spectra that LED's can put out.

It would be nice to know how spiky the emission spectrum is, because that will determine how good of a color rendition you can get without filters.

Mask off all but a narrow line on the panel and run the "reflection in a CD" test, described by g4lab in the same topic.

--Rik

[augusthouse]

Ok,

I have masked the panel leaving a thin line unmasked; but I receive different results depending on which CD or DVD I use; whether it is blank or otherwise may also be relevant? - yes it is, see link at end of post.

I am looking for a duplication of the illuminated line in either red, green or blue or a combination - is this correct? - yes and no - see link at end of post.

*Later note: I believe I can see it now - it's beautiful. I expect the colours I am seeing have more to do with the reflective properties of the DVD rather the effect of the illumination - not necessarily; a CD/DVD is an example of a reflection diffraction grating.

When comparing my visual observations with images at the link below; if I select 'Sun' as the 'Source' for the Spectroscope; then that's about what I'm looking at - from 425 to 700; but there is a more pronounced, narrow, dark band in the area between 650 and 700, though this could just be a coincidence or a result of the camera, subject lighting arrangement... - not necessarily.

http://mo-www.harvard.edu/Java/MiniSpectroscopy.html

Craig

P.S. I have no idea what I'm doing; but have more of an idea now than I did before I started this exercise. I have not constructed a spectroscope. What is a spectroscope and how do I make one? Link below:
http://sci-toys.com/scitoys/scitoys/lig ... scope.html

I'm going to buy some diffraction grating film and explore this further - just out of curiosity.

My colour vision is apparently normal.

[augusthouse]

Question...

If I were to use the item below in a slide duplication type setup would it yield an accurate result in determining the spectrum, in this instance, of the Rosco LED LitePad? Obviously, the LEDs in question would be used as the light source.

These embossed holographic element (H.O.E.) diffraction gratings measuring 35 mm x 23 mm (mounted as a 35 mm slide). Ruled at 1000 lines/mm, these gratings exhibit high diffraction efficiency (most diffraction is first order) with a minimum of visual noise and large diffraction angle (1000 lines / mm has a greater diffraction angle) - excellent for all those experiments with lasers, rainbows, atomic and astronomical spectroscopy and anything else that you could think of! You can even identify mercury and sodium vapour lamps by their spectra! A handy wavelength reference is even included on the mount!

Diffraction gratings are the principal component in modern spectroscopes and are used for experiments pertaining to the study of light and colour: Project a spectrum using an overhead projector or 35mm slide projector for spectrum demonstration purposes. Use in conjunction with colored film gels for additive and subtractive colour demonstrations.

[g4lab]

It is pretty easy to find the spectra of LEDs online.
For example here Ocean Optics

I was going to keep my yap shut on this thread because I am not much of a digital photographer yet. But another interest of mine is gemological spectroscopy and gemology generally.

When looking through a stereo scope at a gem whether a diamond or colored stone, the metamerism color shifts of LED light sources is totally unacceptable.

This is in spite of the fact that various sources are starting to furnish them on gemological scopes. THis because they are new, cheap, free from heat relative to quartz halogen, efficient, relatively lower current devices. A lot going for them in fact. But not enough.
And also in spite of the fact that the previously most reasonably priced source of standardized light for these examinations is/was the socalled Daylight 5,000K fluorescent tube. There are many varieties of these each with a different color profile, color rendition index and spectrum. Not surprising because each company manufacturing has their own blend of phosphors that they call Daylight.


Pretty long ago (1989)a two part article was published on this subject which I think I already posted but will post again:
Part One
Part Two
I like this article because it mirrors my own prejudices and states them better than I could ever hope to. Of course when it was written there were
no white LEDs

Spectra for some Fluorescents as well as some rather prickly but well founded opining on the subject of gemological lighting can be found here:
At Adamas Gem Lab
the proprietor Martin Haske does not suffer fools nor BS gladly. He posts some pictures of various fluorescent lights.

When I thought to post on this I thought , well Charles has said exactly what I wanted to say so there is no reason to be redundant. But OK I will be. LEDs are not ready for prime time if you want accurate color rendition.

Similar problems exist for Metal Halide arc lamps which are widely sold to photographers now. They also have rather peaky spectra.

I don't know enough about RGB color and RAW and all that good stuff to know whether a correction can be made for such peaks properly. I suspect a skilled person can. I leave that lecture to someone else.

It is fairly easy to make your own hand spectroscope out of a broken CD and a cereal box. There are also lots of them for sale on ebay starting at about $4.00 for educational ones. You want one with a diffraction grating rather than a prism for the most part. The grating you showed is exhibiting mirror image spectra not spectra according to the scale printed on the slide holder. You need a slit (two razor blades) and a lens and an ocular or your digicam. It is possible to view crude spectra using a digicam but a black and white sensor is better because it eliminates all the Bayer pattern arithmetic after which I would not trust any spectral readout much. But you can get a nice histogram spectrum from photoshop and similar programs.

Here is the website of my friend and spectroscopy fanatic Ioannis Galidakis in Athens.
http://ioannis.virtualcomposer2000.com/ ... index.html
He is also a light source aficionado. Isn't the internet wonderful where people with the most arcane of interests can find each other. (I sent him the Xenon XBO short arc lamp he shows on the website.) He has instructions and analysis (He is a mathematician among other things) of the cereal box spectroscope and others including a monster one he built himself. Great website. Most of the spectra he shows were taken with a digital camera. I don't think he has started on LEDs yet. His links page has everything you could ever want to know about lighting and spectroscopy including some links to pages I wrote most of. (Thank you Ioannis )

[rjlittlefield]

I don't know enough about RGB color and RAW and all that good stuff to know whether a correction can be made for such peaks properly. I suspect a skilled person can. I leave that lecture to someone else.

Nope, can't be done.

Very briefly, the problem is that what the camera senses for "red" is basically just the total energy making it through the camera's red filter. That energy is the integral of the product of three spectra: emission from the light source, reflection from the subject, and transmission through the filter. Likewise for green and blue.

Depending on how the bumps in the spectra line up, two materials can give identical RGB energies under one light source, and different RGB energies under a different source. No amount of Photoshop magic can correct for that difference because not enough information was collected by the camera.

In many cases the colors could be properly corrected if the camera recorded a complete spectrum instead of three aggregated values. But such cameras are expensive and specialized, to say the least.

--Rik

[g4lab]

When I said I would leave that lecture to someone else I wuz kinda thinking of you Rik.

What cameras record the whole spectrum??? I have never run across such a thing.

What about if you take a picture of something like a Macbeth color checker.
Can that make enough info extractable to make a correction?

[Harold Gough]

What cameras record the whole spectrum??? I have never run across such a thing.

Perhaps not the answer but this may be of interest:

http://fujifilmusa.com/products/digital ... index.html

http://fujifilmusa.com/products/digital ... index.html

http://www.advancedcameraservices.co.uk/html/ir.html

Harold

[rjlittlefield]

When I said I would leave that lecture to someone else I wuz kinda thinking of you Rik.

I figured you were, but that really wasn't a lecture, more like a sidebar. (Lectures last much longer, and take even longer to write!)

What cameras record the whole spectrum??? I have never run across such a thing.

The search phrase is "hyperspectral imaging". Quoting from Wikipedia,

The primary advantages to hyperspectral imaging is that, because an entire spectrum is acquired at each point, the operator needs no a priori knowledge of the sample, and post-processing allows all available information from the dataset to be mined.

The primary disadvantages are cost and complexity. Fast computers, sensitive detectors, and large data storage capacities are needed for analyzing hyperspectral data. Significant amounts of data storage is necessary due to the fact that hyperspectral cubes are large multi-dimensional datasets, conceivably exceeding hundreds of megabytes.

The simplest form of hyperspectral camera is a recording spectroscope with a very small input footprint, scanned over a 2D area.

What about if you take a picture of something like a Macbeth color checker.
Can that make enough info extractable to make a correction?

Nope. Calibrating on a Macbeth ColorChecker will do an essentially perfect job of putting the ColorChecker's 18 sample materials exactly where they ought to be in color space. Whether it will also put other materials in the right place depends on luck of the draw regarding bumps in the spectra. Therein lies the catch. It's very easy for other materials to look just the same as the ColorChecker's under natural daylight, but different under lamps. Same RGB under one light, different under another. If all you get is RGB, you're stuck.

The reason color correction filters work is that they get applied at a place where all the wavelengths are still available to be operated on separately. In that case, what the camera sees is energy integrated over a product of four spectra: illumination, subject, color correction, and sensor. If illumination_B * color_correction = illumination_A all across the spectrum, then the color correction is perfect and any materials whatsoever will be properly rendered. If illumination_A is smooth and continuous, while illumination_B is very spiky or worst of all has zeros in it, then building the required color_correction filter becomes difficult or impossible.

--Rik