All CRT (cathode ray tube) phosphors become less efficient at radiating light with use, direct view or projection. Similarly with plasma-type flat screens.

LCD and DLP projection, as well as LCD direct view flat screens, are not subject to the same problem. With these technologies, the light source may dim or burn out over time and need replacing, somewhere between 1,000 hours and 60,000 hours of use depending on the method of creating the light.

With a CRT, the amount of light generated at a particular point varies by changing the amount of electron bombardment of the phosphor at that point. With LCD and DLP, the light is always being generated behind the LCD or DLP elements, but the amount of light you see at a particular point is regulated by how much is allowed or directed toward the viewing surface by those elements.

On a CRT, all colors (red, green and blue) near their maximum excitation are used to make an approximate white. If I take a CRT direct-view or projection system, feed a signal that shows a large white dot in the center of a black background image area and let it remain that way for too long, the phosphors in the white area become less efficient at showing light compared the relatively unused phosphors in the black area. If I later send a signal that provides an evenly distributed dull gray everywhere, I would see that the former dot area on the screen was slightly less bright. If the blue phosphors wear more quickly than the red or green while producing bright white, then the formerly white area may seem yellowish or brownish compared to the formerly black area. This type of ‘burn’ can occur in a plasma screen after just a few hours. A CRT projection system, because small CRT’s are driven hard, begin to ‘burn-in’ with a few dozen hours of unchanged image area. Direct view CRT’s usually take many hundreds of hours of unchanging image area to ‘burn-in.’

You might say, “I never show a white dot continuously on a black background.” The usual culprits are a television that always monitors a single channel that has a logo or lettering that stays on most of the time in one area while the rest of the screen gets a varying image or viewing a signal that does not fill the image area.

For television viewing where all portions of the image are constantly changing, the phosphors are wearing out slowly and approximately evenly everywhere. If the television is used the majority of the time for images that only use the center of the screen while the borders are black, when a full screen image is shown, the phosphors in the center will eventually appear not as bright and/or discolored compared to the border areas. The point of having gray borders is a compromise. The engineers’ idea was to wear out the phosphors in the border area at an average rate, hoping that the center area would, with varying bright and dark areas, not wear out more quickly or more slowly than the phosphors continuously being used at half brightness in the gray area. If the average-over-time brightness of the center area does not equal the gray border average brightness, eventually you will be able to see, when viewing a full screen image, the boundary that was between the gray area and the varying image area.

Obviously a black border area will never wear at the same rate as a varying image area. What is being relied upon are normal viewing habits. That is: most viewing will be full screen and only a minority of viewing time will use a limited area of the screen. Full screen use tends to reverse minor burn-in tendencies. So if full screen viewing is generously sprinkled in among partial screen viewing time, the burn-in effect either tends not to accumulate or tends to accumulate very, very slowly. Besides, as you know, consumers generally accept black borders and balk at gray borders.

If a CRT has a very minor burn-in, one ‘cure’ to attempt is to send a full-screen white signal to that CRT for 10 to 300 hours immediately after the burn-in occurs. This is similar in theory to lightly sanding a wood surface after the surface receives a very mild abrasion. Generally you are not “healing the wound” as much as bringing the rest of the surface to the same level so that the imperfection is no longer in contrast to the rest of the area and the abrasion’s edges are smoothed out a bit. The success of this cure varies.

I know I tend to go on for a long time, but I hope an engineering viewpoint from someone that works in television and presentation media has helped.

And ... welcome back SLL!

[This message has been edited by bestbang4thebuck (edited April 09, 2004).]