Measuring Screen Reflectance
High quality projection screens are very good at dispersing light. And, as anyone who has priced professional quality screens knows, they are also very good at dispersing money. This inspires many to a DIY approach. It may be buying screen fabric through Ebay, or building a screen using specialty screen paint, or experimenting with standard wall paint.
Regardless of the approach, we want performance value for our efforts and money. An issue with a DIY approach is that we’re generally unsure how well the finished screen will perform until we’ve built it.
The key performance factors for a screen, DIY or otherwise, is how much light is reflected and in what direction. This is documented in a gain curve for professional screens, but not so for one-off DIY screens.
A gain curve allows us to predict brightness uniformity across possible screen configurations without having to build them. (See the February 2011 Mike’s Tips article for more about uniformity and image quality.)
Data for producing a gain curve is collected using a test stand and a spot radiometer or light meter. The test stand holds a sample of the screen material and shines light on or through it. The light meter measures the intensity of the light coming from a small area of the screen at a number of angles.
This gear tends to be on the expensive side for hobbyists. I wondered if it was possible to collect similar data using something more affordable.
A quick browse of Ebay turned up several low cost luxmeters. Most resolve down to 1 lux. I chose a slightly more expensive model (about $40) capable of resolving 0.1 lux. I very much doubt that any of these models have outstanding absolute accuracy. However, what I’m really interested in is relative brightness. I want to know how screen brightness changes with viewing angle.
These luxmeters measure illumination across too wide an angle to make good screen brightness measurements. To measure how much light comes off the screen at various angles, the luxmeter sensitivity should be restricted to a narrow input cone. I made a light baffle consisting of a black paper tube with three internal optical stops. The stops prevent shallow angle internal reflections off the tube wall and increase the sensor's directivity.
I cut down a cardboard box to hold the test samples vertically. On the bottom of the box, just in front of the sample, I used a protractor to drawn angle guides for positioning the luxmeter light detector.
I recorded the brightness of the same small area on the test sample when viewed from several different angles. I kept the open end of the baffle the same distance from the sample spot for each measurement.
I was surprised how little light leaves the screen surface across a restricted angle. I started these experiments expecting to use an LED flashlight as a light source. To get good readings on the luxmeter, I had to use a 300 watt work light.
I used two test samples. The first was flat white paper with a brightness rating of 92. The second was a sheet of glossy photo paper. The results are charted below. I couldn’t measure the brightness at 0 degrees (straight out from the sample) because the light detector creates a shadow.
These samples aren’t screen material but they do show that this approach works. Had they been actual screen samples, it would be clear that the second would be unsuitable for a flat, wide field of view screen. There would be a very distracting hot spot in the center and greatly diminished relative brightness toward the edges.
While these measurements allow you to determine relative brightness uniformity across a screen, they don’t directly provide absolute values of screen gain. If you want to determine gain, you could buy a sample of commercial unity-gain screen and use it as a reference.
The cardboard test stand was sufficient for demonstrating proof of concept, but could be improved upon by adding a robust support for the baffle and detector. The stand can be modified for use in testing rear projection screen materials by cutting a hole in the vertical sample support and positioning the light source behind the sample. Cardboard is not a bad material if you’re only planning on testing several screen material options, and it’s an especially good material in that you won’t feel bad throwing it away when you’re done.
Mike Powell, author of
Building Recreational Flight Simulators ,
Building Simulated Aircraft Instrumentation, and
Building Simulator Displays Systems. (A work in progress)