A White Background

Consider if the subject in the foreground was holding a piece of the white background seamless paper next to their face with only the key and fill lights on. When the exposure for the face was correct the paper would be reproduced as white wouldn't it? Not 255.255.255 blown out white, but something in the range of 240s which makes it look like what it is, a piece of paper not the blazing sun.

Let's say you measure the key and fill for exposure by pointing the meter at the camera dome-up and get a reading of f/8. Now leave the key and fill on and put the meter on the background near where the head appears in the viewfinder and take a reading. Raise the power of your background lights until you get the same f/8 reading which correctly exposed the background material when the subject in the foreground was holding it.

The area behind the head on the center of background where you took the measurement should white. Not blown out 255.255.255 white, but something in the 240 range. Stop and take the time to look at the files in Photoshop and measure with the eyedropper on various parts of the background.

If you have not evenly lit the background you will see that the edges are darker. Pounding more light into the center behind the subject to make the edges dark is not a good idea because you will grossly overexpose the center and create contrast robbing lens flare.

If you can't evenly illuminate the seamless so it is uniformly white w. some detail (i.e. RGB = 240.240.240) you will have identified the source of your problem. If using two lights you will need to aim each about 1/3 of the way in from the outer edge to get the lighting even. Adjust so the two lights overlapping in the middle equals the intensity of each one individually on the sides. The further you place the background lights from the seamless the more even the distribution will be. Lots of space is needed. If despite your best efforts you can't get the lighting even then resign yourself to doing some lightening of the edges in post processing with a masked screen layer.

Once you get the "off white" background as even as possible raise the lights on the background in 1/3 stop increments and evaluate the results. The first 1/3 stop over the foreground reading should be enough to get the area you measured in the center at or near white. I think its better to err on the side of slight underexposure and keep it around 250 - 253 rather that blowing it out completely in the camera capture. Try increasing the exposure for the background by 2/3 and a full stop over the foreground reading and evaluate the results. Pay close attention to changes in the contrast of the subject (loss of shadow density) which is how lens flare will manifest itself.

By doing this simple test one time you will be able to determine for yourself how best to expose the background relative to the foreground reading. You want to meter the background after setting the foreground and base the background reading on the overall exposure reading, not just what the key light reads, then balance the background relative to that reading. That after all is the way the camera sees it: subject lit by key and fill, and background lit by background light plus falloff / spill from key and fill. The size of the shooting space can make a difference still spill off walls and ceilings in a small space wlll affect the amount of light that winds up being reflected off the background into the camera.

The overexposure warning in the playback is very helpful in the evalution. If you start with the background lights low and raise them to the point where parts of the background start to get overexposed you can see where the hottest spot is and how even the lighting is. When you get it right per the playback you can then meter it to discover the numerical corrolation with the foreground exposure.

Once you do this test you'll know how better to set your lights for an evenly illuminated background and how much over key/fill foreground exposure you need. You will be able to set the background lights with the meter and then know what to expect in the playback.

Contact: Chuck Gardner

Tutorial Table of Contents ©Charles E. Gardner - all rights reserved