HDR: Part 7 - LEDs - For Better For Worse
If we could get lights capable of full color mixing at the same price as conventional white-only ones, would we?
This article was first published in 2020. It and the entire 'HDR Series' have been immensely popular, so we are re-publishing it for those who missed it first time around.
Color Mixing Lights for All
It’s easy to imagine a fairly imminent future in which all lights might be capable of full color mixing, and that seems like a great idea. It’s faster, it saves buying and carrying gels, and makes matching practicals and other manufacturers’ lights easier than any combination of gels. It lets us drop in a splash of bright, saturated color when we need it, too, although the party colors are perhaps seen less often on a film set than a trade show floor. Probably the most important thing color mixers can do is variable white, but sure, nobody’s going to turn down the option to have any color of the rainbow at the press of a button if there are no compromises.
Lower Power Levels
In reality, compromise is inevitable, and the most common compromise in color mixing lights is sheer power level. If the light is rated at 100 watts, that might represent peak available power. Even if we think about a light with variable color temperature (not full color mixing), there might be 50 watts of cooler light, and 50 watts of warmer light, meaning that the output power is only the advertised 100 watts when both sets are on together. That only happens when we select an intermediate color temperature matching neither tungsten nor daylight.
Certainly, this is not the case with all lights. The peak power is often limited by the ability of the light to keep the LEDs cool and dissipate their heat, so it would be quite possible to build a 100-watt light and have 100 watts of both cooler and warmer emitters. It might be bulkier, though it could maintain the full 100-watt rating regardless of the selected color temperature, which is arguably a better design if we’d prefer the output not to vary with selected color. Some lights even offer a choice – maximum output mode, where intermediate color temperatures are brighter, or constant output mode, where brightness remains constant as color varies.
Either way, it’s quite possible that a variable color temperature light might have lower output, at least sometimes, than an otherwise identical light with fixed color output.
Lights capable of full color mixing may have much the same problem. Most people understand that full color mixing requires red, green and blue emitters, and if we ask for a deep red, then the green and blue emitters aren’t going to be contributing much. Again, if there are enough red, green and blue emitters to fulfill the light’s nameplate rating, great, but that’s far from a common design approach. Your hundred-watt light is unlikely to be a hundred-watt light if we ask it for a deep, saturated color.
Output isn’t the biggest issue, though. The long running battle with LED lighting has been color quality, although to be fair it is more or less a solved problem in 2020. Certainly, some early products were less than ideal, and LED lighting picked up a reputation for colorimetry problems which has caused a huge push towards improving things. Huge strides have been made in the last five or ten years, and while nothing’s perfect, many LEDs are now better than trusted technologies like HMIs ever were.
Or at least, some LEDs are that good. White LEDs are built using blue LEDs to make a yellow-emitting phosphor glow, which we call a phosphor-converted LED. Putting that phosphor on the front of the LED means that it will get hot as the LED does, and the spot of phosphor will be inundated with a huge amount of very intense blue light. It’s a workable approach – that’s how most LED movie lights work – but the absolute best designs, remote phosphor designs, put the phosphor on a separate panel and illuminate it with the blue LEDs, saving the phosphor from the worst of the heat and brightness.
The problem is that there’s often no easy way to use multiple emitters to implement color mixing with remote-phosphor designs. That’s not a complete disaster, though; there are perfectly competent phosphor-converted white LEDs around which very high quality lights can be made. If there’s a question, it’s why a light with red, green and blue emitters needs white emitters at all – couldn’t we just turn on the red, green and blue emitters all at once, with the same result?
Yes, if those RGB emitters added together added up to white light. With most LEDs, they don’t.
Traditionally, to make colored light, we’d filter white light to remove colors we don’t want. That way, it’s possible to have yellow light with a spectrum that includes a broad spectrum of wavelengths around 580nm, which most people would see as a bright canary yellow. LEDs, though, don’t have that broad spectrum. Combine red and green LED light, and the result will look yellow to humans, but it might not contain any 580nm light at all. It will contain red light around 680nm, and green light around 530nm. On a white surface, the result will seem yellow. On a yellow surface, though, one that only reflects light in a narrow band around 580nm, the light might look unsaturated and dark, despite that being exactly the opposite of what we’d expect.
That is, to be fair, an extreme example. Certainly colored LEDs can create very saturated colors – a narrow output spectrum, technically – but it’s not common for real world objects to have such saturated colors that things look really wrong. It’s possible, though, and it’s also possible for subtler errors to happen in less extreme scenarios. Most lights that offer color mixing therefore need at least four sets of emitters: red, green, blue and white. Then, if we want white light, the bulk comes from high-quality white light LEDs, and the RGB LEDs can be used to adjust the color. This has a lot of advantages: not only can the light produce variable color temperature, but it can also produce plus- or minus-green shifts, as well as bright and saturated effects colors.
Diagram 2 - The diagram on the left shows an LED with significant distribution in the red and yellow but only partial distribution in the green spectrum. Using reflected light, the apple would look desaturated in a camera as there is not much green from the light source. However, the apple on the right would look highly saturated as there is significant green in the distribution. To the human visual system, both distributions would look the similar, but to a camera there would be a difference in saturation.
The Fundamental Problem with Color Mixing
Mixing together the various emitters to create useful results is complicated, and these lights involve a lot of complicated color science. Even a four-emitter design, though, doesn’t always produce quite the best results available. Some designs might use a white emitter designed to emit cooler light, closer to daylight, and use the RGB emitters to create warmth in the color if the user needs tungsten-balanced output. That works, but it means that the warmer the light, the more output is coming not from the white emitters, but from the RGB emitters. Color quality might be noticeably worse with warm whites than cooler ones. That, fundamentally, is the problem.
There are a number of possible solutions. First, we can use even more combinations of emitters. Some of the best full color mixing lights in early 2020 use five, with two shades of white to create variable color temperature, and RGB emitters to make very small adjustments to the output color. A huge variety of approaches are possible, and crafty designers may take a number of approaches in designing the electronics to mix various emitters together in various ways. Some lights even use six emitters, excluding white in favor of lime green and amber colors.
Phosphor-converted Colors and Ultraviolet
We might expect this approach to raise the same problem as with simple red, green and blue LEDs: they produce narrow spikes of color that may not illuminate saturated objects very well. The solution to this problem might actually be an answer to almost all of the questions we’ve raised so far: phosphor-converted LEDs designed to produce red, green, blue, or other colors. Instead of converting a blue LED to white, a phosphor can also convert it to red or green, creating a less monochromatic and saturated light. Create phosphor-converted red, green and blue, and the combination might add up to a more reasonable white light.
As ever, there are compromises. Less saturated light is – well – less saturated, and a phosphor-converted red LED is visibly more pastel than a conventional red LED. This isn’t often a big problem since most cameras have a limited ability to see very saturated colors anyway. More crucially, phosphor conversion is not a lossless process. It reduces the efficiency of the light over non-converted LEDs, and efficiency is what we wanted LEDs for in the first place. Finally, the biggest problem is that phosphor-converted blue LEDs are difficult. Phosphor conversion can only shift a color toward red, never toward blue (strictly, it can increase wavelength, not decrease it.) So, we can’t create a phosphor-blue LED with a conventional blue LED; we must start with an ultraviolet LED, which in early 2020 is a difficult, expensive technology with low output and short lifespan.
It probably won’t be long before the UV LED problem is solved and it becomes possible to make LED lights with entirely phosphor-converted emitters. Some very high-end lights reputedly use UV emitters, though manufacturers are often tight-lipped about their latest technology. In the meantime, it’s a good idea to ask probing questions of anyone selling a color-mixing LED light about exactly how it’s built, because the effects of those choices can be far-reaching in terms of color quality, output, and efficiency. Pick any two, perhaps.
You might also like...
The Peril Of HDR: Just Because You Can Doesn’t Mean You Should
There is a disturbing and growing consensus among viewers that many movies and TV shows today are under illuminated or simply too dark. As DOPs, we surely share some of the blame, but there is plenty of blame to go…
The Back Of The Brain May Soon Rule The Roost
If industry reports are to be believed, Apple is poised to release a mixed-reality headset at some point in 2023. Of course, it’s anyone’s guess when Apple’s Reality Pro will actually see the holographic light of day, but one t…
Learning From The Experts At The BEITC Sessions at 2023 NAB Show
Many NAB Shows visitors don’t realize that some of the most valuable technical information released at NAB Shows emanates from BEITC sessions. The job titles of all but one speaker in the conference are all related to engineering, technology, d…
Magicbox Puts Virtual Production Inside An LED Volume On Wheels
Virtual production studios are popping up across the globe as the latest solution for safe and cost/time-effective TV and movie production. This method replaces on location shooting and, by utilizing all-encompassing LED walls (often called “volumes”), is fundamentally changing the…
Celebrating BEITC At NAB Show
As we approach the 2023 NAB Show in the NAB centenary year, we celebrate the unique insight and influence of the Broadcast Engineering & IT Conference that happens alongside the show each year.