While moving and setting up my new studio, I (wince) knocked my lamp off of a table and broke the bulb that’s used to light my palette, which lead me down the rabbit hole of searching for a more accurate light bulb for color matching.
After hours of scouring the web, I was unable to find an LED or compact fluorescent bulb the had better specs than the one that I’d been using for the last 5 years.
For years I’ve been very happy with my easel and palette lighting set-up, but because of price and the negative environmental impact of compact fluorescent bulbs, I was hoping to find something comparable in an LED. But I can’t complain much about the price, after all, it lasted 5 years and who knows how long it would’ve lasted if I didn’t break it.
These would be the ideal lighting specs for temperature and color matching:
100 CRI (Color Rendering Index)
These are what I currently use:
(I modified the lamp to work better for my needs by removing the 90° elbow)
These are excellent bulbs for color matching and do the best job of replicating the temperature of midday sun that I’ve found. Although the CRI and Kelvin ratings are slightly different, the difference is barely perceptible when moving paint between the palette and easel.
If anyone knows of bulbs that are equal to or better than the ones that I’ve listed, I’d love to know about them.? If not, I would highly recommend these.
Here’s a photo of my current lighting setup…
I have the professional artist lamp and it’s good but not big enough to light all of a big painting. The other thing is if you don’t have a Single mast easel, it doesn’t work well. I have a big Hughes easel and I had to get another single mast easel next to my Hughes easel and clamp the light on that easel. What a problem because I don’t have that much room to have a double easel set up with a huge Hughes easel , Which is very big and very wide. I am looking for an alternative. For right now I have my easel set up next to my north east light window with that light over it and on the other side I have another Smith & Victor lamp that I have for night time to have a double light set up so that it hasn’t even light spread. I’m curious about the pallet lamp so I will try that. Thanks for this post.
I know Paul MacCormack in his studio uses TL 950 by Phillips 5,000k CRI 98 and wescott bulbs that are 50 watt on the model and a 30 watt on still life, with t90+ CRI and a 5,500k. I have not tried those but thinking of changing the ceiling fan in my studio that was a bedroom!
A while back, Marvin Mattelson told us about lumichrome by the light. 98 CRI and 6,000k which is too cool! I alternate with a daylight spectrum that’s warmer. Marvin was my teacher in the beginning of my painting journey and he knows allot about this. He is using different lights now.
Thanks for your suggestion of the Philips TL 950 tube. They look like a good option for whole room/studio lighting. I agree that 6000K lighting would be too far to the cool side of the spectrum for accurate color matching. Knowing that galleries mostly use halogen or equivalent lighting, which tends to be warmer, I believe that 5300K might be ideal for painting, color matching and comfort in knowing generally how the colors will translate in warmer gallery and home lighting.
It’s hard to tell from the photo, but my easel is a modified model #3000 Hughes Easel. I built a table for the front that holds my palette, brush cleaner and my easel lamp is clamped to it. I also removed the part of the easel that holds the canvas and replaced it with a rigid surface that is better suited for smaller panels (the gray surface shown in the photo). The alterations are not permanent and the canvas and panel holders and interchangeable for different needs. I wanted to point this out to show that, because the easel is basic and wooden, you can modify it to fit your particular needs. If you need something to clamp your light onto, there might be a way to add the option to yours.
Thank you for the awesome post Slade!!!
My choice of lighting has always been a pragmatic one. I often fell in line with which bulb seemed to be most commonly-sold bulbs commercially (the thinking here is that I would be likely be working under the same light that would eventually illuminate the work in the home of a collector.)
After a 2014 Color workshop with artist Douglas Flynt we became interested in taking a closer look at the different types of lights that populate our studio. One aspect of our lights that we had not considered much before Doug’s visit was CRI (Color Rendering Index). Color rendering describes how a light source makes the color of an object appear to human eyes and how well subtle variations in color shades are revealed. The Color Rendering Index (CRI) is a scale from 0 to 100 percent indicating how accurate a “given” light source is at rendering color when compared to a “reference” light source. The higher the CRI, the better the color rendering ability. Light sources with a CRI of 85 to 90 are considered good at color rendering. Light sources with a CRI of 90 or higher are excellent at color rendering and should be used for tasks requiring the most accurate color discrimination.
It is important to note that CRI is independent of color temperature.
In addition to carefully reviewing what the bulbs state on their packaging and on their respective websites, here at the AAAW we thought we would examine the color makeup of our lights for ourselves (remember–always check for yourself if possible!). We made an aperture for the various lights, grabbed a large prism and started splitting the lights up into component colors. (We should mention that a few lights are included here that we do not normally use in the studio but felt they added some breadth to the experiment. In addition we are missing some CRI information that we just could not find (Mighty Bright Orchestra Light, iPhone 4 Light, etc…)
This image shows what we saw. No images were enhanced other than a general softening/de-speckling to remove noise from the photographs. (click to enlarge)
The bulbs that I currently use are Cree LED bulbs. 93CRI - 800 Lumens (60 Watt replacement (13.5 Watts))
You can get them here: https://www.amazon.com/Cree-Equivalent-White-2700K-6-pack/dp/B00WU2JRLO/ref=sr_1_1?ie=UTF8&qid=1452381368&sr=8-1&keywords=cree+60w+soft+white+led+bulb+tw+series
Additionally, if you would like a little help in choosing the right bulb for you I would recommend this great primer on buying light bulbs from CNET:
Thx for this reply Tony. I check out your suggestion andcwe t to Amazon… if I use 5 of the bulbs you suggest, 4 in my ceiling fan…so I do t have to dismantle the fixture, and one in my Smith and Victor, would it work with my easel pro artis easel lamp that Skade and I have. I hate tvatblamp as it only works in a single mast and I’m dying to get rid of the other easel next to it. I have to take a shot of my studio to show you what I’m talking about. But your idea seems best with the CRI
Sorry spell check gone wild lol in a car
Thanks for the reply Anthony! I loved seeing your experiment results!
I read up on the Crea lightbulbs when I was researching LED bulbs and after seeing that their temperature in 2700K, I moved on. I’m sure that, knowing the quality of yours and your students paintings, you would have switched to another option if your lighting was hindering your efforts, but I do have questions about the choice…
Because the color temperature is so low (well within the yellow side of the spectrum), do you find it difficult to discern slight variations in blues or other cool hues?
Do you find that the paintings lose their warm tones when lit by mid morning to mid afternoon sun?
It would be interesting to see how the lights that I’m currently using would fare in your experiment.
Here’s something that you might find interesting. It isn’t really related to this thread but only an interesting group of analyses regarding CRI.
You are 100% correct Slade. There is a 5000K LED from Cree:
I should also share that I gravitated to the 2700K as I had painted under regular incandescent lights for many years prior to the switch to LED (as did many people who had incandescent lights in their home/work.) I don’t think that I have much trouble discerning changes in the short wave range, but I guess it would require objective testing to be sure.
One argument that could be made for leaning into the 2700K range is that work may seem slightly cool when completed (and viewed in a 5000k range of illumination). Considering that many paintings will yellow over time, one might say my work will come closer to its intended appearance over time. LoL!
Hmmm interesting thought. True about yellowing. Do you find that the paintings look too neutral under a cooler light? Also, Liquin yellows as weel. So, that might be something to consider in choosing the right temperature for myself. I need to add more bulbs in my studi for night time. Day is fine ,but night time is tricky in my set up!
Not that I have noticed Nanci. I have access to paintings that have been done about 20(+/-) years now and they do not appear to be yellowing just yet. Keep in mind that my use of Liquin is SUPER conservative.
Some of mine have darkened, but I think it’s the damar with turp and stand oil. Do you think that’s what caused a little darkening?
I did opt for another bulb comparable to Cree with lumens a little higher than yours. I purchased the 5000k 93cri 60 watt to put in place of fluorescent tubes . I have natural light coming in with the pro artist lamp over my easel.
I like the mix of warm and cool lights as I haven’t had a problem with different lights.
I totally understand why you using 2500k as work is not represented in north light but incandescent.
I’m not sure I’m thinking right, from your scientific logical point of view. Analytically, I’m thinking my north east light window against the warmer light might produce your effect.
Any thoughts on that?
Btw… loving this forum and… my friend Amy that goes to ARA in Boston will be joining. I bragged about you lol. Hoping she joins the Skype group! Pam said she had to watch her grandson but should be in your next session:-) hoping my spell check hasn’t gone wild on this iPhone
It could be Nanci!
The only resin recommended by Natural Pigments’ George O’Hanlon is synthetic. He recommends that organic resins, like copal, mastic, and damar, should be avoided as they darken and yellow significantly.
However, Damar IS reported to be the least yellowing of the natural resins. It is the one natural resin that is said to be truly comparable in quality to the synthetic alternatives.
It is also important to understand that an oil painting’s appearance will be affected by its environment. The color of a drying oil in a painting (linseed, safflower, walnut, and even alkyds) changes in response to the amount of light that it is exposed to. Light exposure bleaches oil, and in the dark, drying oils increase in color. However, extreme conditions aside, for the vast majority of paintings and the vast majority of display and storage conditions these changes are relatively imperceptible.
I use 5 different bulbs (2 cool temp, 2 warm temp, one middle temp; with different methods of emitting light i.e. incandescent, LED, CFL) in one of those floor lamps that has 5 tentacles that receive 1 bulb each. My hope is that, while each of the bulbs emit a different spectrum, that the total spectrum received by my surface and palette will be pretty broad and even with minimal maxima in the spectrum. It has worked pretty well. I haven’t had problems with color surprises (metamerism) in different lighting conditions since switching to this set up.
The lamp has settings for 2, 3, and 5 bulbs turning on simultaneously, so when checking color carefully, I will try each of the settings (different combinations of bulbs) just to make sure things look consistent.
I haven’t done any objective measurements of the light to know that it overall is a broad spectrum though.
Hi everyone, The packaging for the currently available Cree bulbs do not display the CRI value. Wave Form Lighting lists CRI data and too much more data at least for me. Anyway I bought a couple and will report back after awhile.
I like the Waveform light-bulbs. My response is subjective though.
Lately I’m working on a painting using only Lamp Black and Titanium White. The light gives excellent detail ( I was surprised by a quality of enhanced precision which seem kind of weird so far there is nothing cloying about it) and a even field (no hot spots) which is useful for a close reading of the graphics or the actual mark.
As for color I can’t say from experience. Yet to say something I just looked through my color cabinet, the drawers of which are underneath a lamp with a WF bulb, at the spill-over around the cap and I can discern differences from one color to the next almost easily? And some of these are very close just being the difference between one manufacturer and another.
To sum then the Waveform light-bulbs seem to expand or at least not compress the range of observable detail and color.
I’m surprised that nobody has mentioned Solux bulbs. Particularly their MR-16 low-voltage halogens, which are available at color temperatures up to 5,000K (though I actually prefer the 4,700K bulbs, which are available in a brighter 50 watt version).
Halogens have some downsides. You need special low-voltage fixtures to use them, which can be a bit pricey. (They do sell “PAR”-type lamps that will work in a regular Edison socket, but the color temperature for those only goes up to 3,500K, which is a bit low for my tastes.) You’ll need a fixture or fixtures that can accommodate several bulbs, because a single 50-watt M-16 halogen doesn’t put out enough lumens to be a good easel light. I’d recommend using at least two, but three is better. Get bulbs with the widest possible beam spread (36 degrees). Halogens get hot, so you probably want to mount them high up, so that you don’t accidentally bump into them while working. I have mine on 24" long arm fixtures mounted to the top of my easel mast. They also don’t last as long as CFLs or LEDs.
On the upside, these lamps will give you some of the best color rendering you’ll find, period. They are used by a lot of museums to light artwork, for good reason. CRI is 98, as because they are blackbody radiators, their spectral curve is a pretty close match for sunlight (unlike the spiky spectral curve that you get from fluorescent lighting).
The thing is, CRI is… kind of complicated. It has some shortcomings as a measure of color rendering accuracy, which is why the industry now has an extended version of the CRI standard. Which a lot of CFL and LED manufacturers don’t use, because their lamps fare poorly on it–especially in R9 rendering, which measures a lamp’s ability to render reds. This is pretty important for artists–especially if you paint people–and red rendering is one area where many fluorescent and LED lamps fall short.
On top of that, reported CRI values are not always accurate. For example, I’ve seen multiple independent tests that place the CRI (regular–not even extended) of Ott Lite’s lamps closer to 80.
Thank you so much Benjamin! I will absolutely look further into the Solux bulbs. I vaguely remember someone mentioning them on another forum in passing, but there was no follow up.
Also—big thanks for the link to the CRI/RA article. It’s very helpful.
Some LEDs are actually getting really good in terms of color rendering, too, but you have to look over the specs pretty carefully. The company hosting those articles about CRI–Waveform Lighting–makes some really high quality LED lamps. They have a line specifically targeted at artists–the “Northlux” series, which are all 6500K lamps with an extended CRI average of 95+. They also have bulbs at 4000K (and 3000K, and 2700K), if you prefer a warmer color temperature. They guarantee R9 values of 80+ (which sounds low, but is actually much higher than you’ll find on most LEDs and CFLs). And the actual test data they provide shows that in most cases, their R9 values are in the 90s. I like that they provide comprehensive test data for their lamps, with individual ratings for R1 through R15.
Yuji is another manufacturer of good quality, high-CRI LED bulbs. They don’t provide as much test data on their individual bulbs as Waveform does, but I’ve seen some independent testing that confirms that their lamps are solid performers, with Ra CRI values matching their claims, and R9 values generally in the 90s.