Shooting daylight film with daylight corrected tungsten lights - results in blue cast?

[leahandhercamera]

Hi guys

I am currently working on a project which requires me to set up a home studio with limited lighting options available to me. I am shooting Kodak Portra 160 (a daylight balanced film) and using two LED spotlights as my sole lighting source which have a warm temperature of 2700K (linked below). I presumed that because the light temperature was so warm that I would need to correct it with daylight filters so I used 201 Lee lighting gels to correct the tungsten light for daylight film. Despite this, on a test shoot my images came out with a super blue cast (example attached below). I didn’t think to shoot a frame without the gels so I don’t know what the outcome would be with just the unfiltered light. I had presumed they would be yellow/green typical of daylight film shot under tungsten light.

Can someone explain why the photos came out so blue and not properly balanced? Are the lighting gels unnecessary?

LED lights: https://www.amazon.de/Strahler-RGBW...robe-Modus-Wasserschutzart/dp/B08PZC5ZJD?th=1

Thanks!

 

[koraks]

How are you processing, scanning and color balancing the film? Can we see a photo of an actual negative? The reason I'm asking is that 9 times out of 10, color problems in digitized color negatives turn out to be problems with color balancing in the digital domain. This is a quick & dirty correction on your image:

The main problem with the scan is contrast-related; whether or not there's an actual color problem in the lighting-recording film combination is hard to tell. Color fidelity in my edit is of course poor because of a very dramatic correction being done on an 8-bit original, but you probably see where this might be going.

In principle your approach should (sort of) work provided the lights are correctly set to white light only. You can verify whether the color balance is (sort of) correct by using a digital camera or a phone with a manual settings mode, set the camera/phone to 2700K custom white balance and photograph the scene. The color should render (sort of) correctly.

I write 'sort of' because white LED is kind of tricky in terms of its color spectrum, and white balance generally relies on a fairly simplistic ratio of blue vs. red light, without particular regard to the continuity of the spectrum. It so happens that white LEDs are typically blue emitters with a phosphor-coated layer/dome on top which converts blue to the rest of the spectrum. The mix of these phosphors determines the continuity of the spectrum and hence the quality of the light. With cheapish emitters like probably used in your lights, this will not be a very ideal spectrum, or very similar to actual tungsten light, so you can expect some metamerism problems. However, by and large, the rendering should be mostly correct for most subjects and certainly not as extremely off the mark as shown here for the subject matter you've shown.

I had presumed they would be yellow/green typical of daylight film shot under tungsten light.

Shooting something like Portra under tungsten light will give red-orange results, tending towards yellow. Not green. Green you get when shooting under fluorescent lighting, including CFL's. Those have fast gone our of fashion of course. Here's an example of what you can expect if you shoot daylight-balanced film under warm white LED light and color balance the negatives in digital space to daylight (right frame; left frame shot under daylight shown for reference):

Note that the cast tends towards red, predominantly. This is in line with what I point out above w.r.t. white balance boiling down (very simplistically put) to a red/blue balance. For reference, this is the same frames with only a (severe and haphazard) correction to the blue channel:

Note how the background starts to balance out reasonably well; the lighting in the foreground is at an even lower color temperature and there are strong reflections of ocher curtains being projected back into the room, so the foreground would require even more adjustment. Still, you probably get the point that it's mostly a blue/red thing.

Are the lighting gels unnecessary?

What filter did you actually use?

Whether you need the filter depends a bit on how strict you are; if you're going to do the color balancing in digital space, you can get pretty decent results without the gels. Even if you were to print the color negs optically in the darkroom, you generally get a reasonably decent result if you adjust the filter settings appropriately.

 

[BrianShaw]

This is a situation where a color temperature meter might significantly reduce the required experimentation if one desires a fully analog process.

 

[koraks]

Forgive me by adding “D” and making this a “H” thread, but can’t some of the analytic capabilities of a D camera help refine lighting color temperature issues?

That's a very valid point and I'd in fact recommend doing that, since digital cameras and phones are a lot more plentiful than color temperature meters, and will be perfectly adequate in this case:

You can verify whether the color balance is (sort of) correct by using a digital camera or a phone with a manual settings mode, set the camera/phone to 2700K custom white balance and photograph the scene. The color should render (sort of) correctly.

 

[Ian C]

For film use, there is a problem using a color meter with discontinuous light sores, such as LED, florescent, and so forth. The readings are only a sort of average. They don’t report the missing parts of the emitted spectrum.

Filters work well to adjust for various types of continuous-spectrum light: daylight, flash, tungsten filament, quartz-halogen (uses a heated tungsten filament), even flashbulbs.

But no filter can replace missing parts of the spectrum. A good example is the magenta filters used to “correct” florescent light. The filter helps, but never truly compensates for the absent parts of the spectrum. The result is, at best, only “so-so.”