A. It is very unlikely that the light from a welding torch will physically damage your camcorder in the same way pointing your camera at the sun would. Welding equipment does throw off an awful lot of light, with some of the most damaging light produced in the IR and UV spectrum, which is invisible to humans. Your camcorder can “see” this light, however, and it may affect the exposure in unpredictable ways.
Pragmatically, what do you want to shoot? Long and medium shots of someone welding will not damage your camera, but the actual welding arc will be totally blown out and over exposed. If you want to shoot closeup details of the spot being welded, you will need to shield your camera to get a proper exposure and we can think of no better or economical solution than a piece of welder’s filter plate glass. It comes in various shades from 4-14 (with 14 being the darkest). Prices start at around $0.70 for a 2×4-inch piece of glass, so you might want to get a couple of shades to experiment with. As an added bonus, #14 glass is dark enough to protect your eyes and your camcorder even from the sun.
Q. Brian Pogue’s July 2003 article on Lux begs the question of why Videomaker doesn’t include a low light subjective rating based on the opinion of three secretaries and an assistant editor. You might give two ratings: the lux level at which picture deterioration is first noticed and the lux rating where the picture quality becomes unusable. A table of such ratings for past and future cameras might help until the EIA-639 standard is adopted. I might note that my Hi8 TR940 cameras are clearly superior to my digital TRV740 in low light levels.
Michel and Kathleen Caldwell
A. Very astute observations. In the past, we have declined to print manufacturer reported lux numbers, since there were no good standards. All of the major manufacturers have stated that they now support the EIA-639 standard (released in July 1996 and available as CEA-639 from for $49) for measuring lux. As a relative measurement of sensitivity, these numbers will be more valuable to the consumer and we will bring these self-reported figures to you when available.
Most simply: a larger lens and a larger CCD mean better low-light performance. The 394-inch primary mirrors of the Keck telescopes obviously collect more light than the ultra-small DVP9 from JVC. Beyond that, the issue gets very complex. For example, pixel density on the CCD is important. It is also fairly simple to boost the electronic sensitivity of the CCD, but this adds electronic noise to the picture. Boost the electronic gain high enough (many cameras have a manual gain control that goes up to +18dB) and you can shoot video in extremely dark situations. Black areas in the shot, however, will be filled with grainy color noise. The low-light performance of a particular camera, therefore, subjectively depends, to some extent, on what you are willing to compromise.
There are also tricks that can allow a camera to shoot in even less light. For example, cameras that can shoot in an infrared (IR) illuminated mode can get-green-and-grainy video even in no-light situations. Other cameras have slow shutter modes (down to 1/4-second) that will produce stuttery (but colorful) images in very dark situations.
You have some good ideas about measuring the light level (which is easy) at which “picture deterioration is first noticed,” but there is rarely a clear and sudden point where this occurs. Indeed, some cameras show just a hint of color noise in black areas in many situations. Likewise, when exactly does the “picture quality become unusable?” We are not trying to weasel out of giving you an answer, but we hope you appreciate our difficulties. We are always improving our testing procedures and may have a standardized benchmark procedure at some future date.
D. Eric Franks is Videomaker‘s Technical Editor.