Hacking LEE Filters - What Pigments are Inside?

Hi all, I finally got around to writing up a project I did a few months ago. I wondered if I could apply spectral analysis techniques on the transmission curves of gels to figure out what pigments get mixed together. I originally wanted to measure Rosco gels on the spectrophotometer at my lab (Thanks for the gel book @dvsDave) but we are shut down for covid. So I used some LEE data I was able to gather instead. I'd love to share this work with you and get your feedback.

As a sidebar issue, I wonder if UV reactive pigments come into play. I remember that with Roscolene, if you looked at the edge of the swatchbook, there was a vibrant pink and a vibrant yellow/green that looked had that day-glo look. Same with Cinemoid # 8 with an orangeish day-glo look.

I've thought about that some. It's interesting. However, I would guess that the gels you are referring to are just saturated in some hue. most colors are not super saturated in the world, especially compared to the objects around us and the white point.

The reason I don't think that fluorescent pigments are really at play in gel design is because the florescent component of the radiation (light is radiation) would be isotropic. It would emit from the gel in all directions, kind of bad news for an industry where we use "spot" lights. Another reason that they wouldn't work very well is that the amount of radiation that the fluorescent dyes are emitting would be very small compared to the amount transmitting through the gel. It would have little or no observable effect on the color of the beam. Fluorescent dyes and pigments are effective in materials because most materials aren't super reflective so the fluorescent component has a good chance to shine through. And lastly, gels are, typically, meant to be used in 3200K sources, like our trusty source4s. These lamps have very very little short wavelength component that would be capable of exciting the fluorescent reaction.

It definitely occurred to me as a possibility. And actually, there is another GOOD argument for the usage of fluorescent pigments in a gel. Like it or not, the way the gel looks backed by that little white piece of paper in the book has a strong influence on the gels we like. We don't always judge gels in the light lab or through a real light source. And backed by that white piece of paper the fluorescent pigments could be effective in making me think some green gel was SUPER green.

But there WAS something about those 'Lene colors, a medium pink (not the most saturated one), and a pale yellow-green, that caused visible side emission (sadly, not nocturnal emission) to occur. If I can find my swatchbook, I'll look up the exact colors if someone doesn't beat me to it. Then it can be up to @TuckerD to 'splain it to us.

derekleffew said:

Two colors, I want to say 828 Pink and 8xx (closest to R87) would act as what we think of today as fibre optic transfer media. The edges of the perimeter would "glow" or fluoresce under ordinary incandescent light and in a frame in a fixture, well, wow. Never observed those two under UV light, but I can imagine.

Click to expand...

As for the Cinemoid colors, they also had an amber-pink that did the same thing. Until his death a couple of years ago, the lighting designer of Johnny Mathis carried with him a stock of the Cinemoid color, and collected every piece after the gig.

Okay. Roscolene 827 Bright Pink and 878 Yellow Green. Why do only these two colors have such prominent side emission? Is it also true that these two fluoresce under UV light?

I'd love to take a look at that! I have a Rosco book and can do uv-included uv-excluded measurements to see if there is anything there. Just need the gel number. And my lab to open up.

Great work. Haven't finished reading it yet, but what does “optimizer friendly” mean?
Also could/ should be a teaching tool for people who primarily use LED sources or moving lights that we all used to spend far more time choosing colors because of their specific qualities and one shouldn't be complacent in picking "blue" with LEDs because it's easy, compared to how deep a conversation it was when using gels to choose a blue that really felt right.

Yes. Anyone aspiring to be a lighting designer today must fundamentally understand that there is more than one blue, what that means, and how to access them with the consoles and fixtures that they have. Frankly, every art teacher or theater teacher I have ever personally learned from did not understand color. I mean deep down in a way that would qualify them to teach "color science". Unfortunately it is a difficult topic.

Now, with that said I know several current students who are getting a good education about how blue is not just a button you press on Eos. But there is still so much education needed in this industry. Actually. Since you bring that up, I am working my way towards writing about that and how it specifically relates to the entertainment industry. However it's going to take me a few blog posts to get there. Here is the first. I hope this gets enough of a signal boost as a reply to this topic.


Back to your first question, what does optimizer friendly mean. That is a way of writing about a problem to say "a computer can solve this quickly using matrix math" and it means something that is a bit more theoretical and mathematical than most people reading this post need to know. I really through it in as a way to signal to other scientists who might read my blog to say "I thought about this in a smart way." If you want to get more into the nitty gritty, let me know.

As for my new post about color rendering. I'm really looking forward to answering peoples questions, so please ask them!

Great article. Did you make the gifs? Something that simple to show clients when they don't understand what color shift is, is amazing.
Even in my day job, there's constant arguments about color since we're always spec-ing based on Pantone swatches. Those who don't understand don't realize that the type of light is important to how a color looks.

Yes I did make all of those animations. 99% of my code i share freely with my peers. This one, not so much. it's a lot of work to make them and I want my name on them. However, if you have some target fixture in mind. I can make one for you, PM for details.

Oh no worries, I'd just show them your website as an illustration.