Solar Eclipses & Video

2001 Diamond Ring Effect
The "diamong ring" effect occurs seconds before and after a total eclipse begins and ends.
Sony TR310.
(click to see larger image)

Introduction

Total solar eclipses, so beautiful to the eye and long captured on photographic film, are now being recorded on video. Our collective experiences from videotaping the partial phases, diamond rings, and corona at the 1990 and earlier eclipses suggests a number of ideas and observations that others may find useful when planning for the 1991 eclipse in Hawaii and Mexico.

Video camera technology has changed dramatically over the last five years. Most cameras now use CCD or MOS devices instead of the older video tube sensors. The new solid state cameras not only have higher sensitivities, but are also virtually inviolate to intense light that could impair or destroy the older cameras that had tubes. You may remember that Apollo 12 TV transmission from the Moon was prematurely terminated when one of the astronauts accidentally pointed the video camera at the Sun and 'fried' its tube. Fortunately, today's CCD/MOS cameras can tolerate exposure to direct sunlight for short periods of time with no harmful effects. Furthermore, great advances in miniaturization have lead to the introduction of the camcorder, a video camera and recorder self-contained in one compact and portable unit. The highest resolution now available for non-professional systems is found in Hi-8 and VHS-S formats. Both show better resolution than ordinary 8 mm systems, which in turn are better than ordinary VHS (or the compact VHS-C). The amazingly small size and weight of many new 8-mm camcorders is a decided advantage for world travelers and eclipse observers. New models are introduced almost monthly and the latest versions weigh under 2 pounds! It is interesting to note that all three of us independently chose compact 8 mm camcorders to videotape the 1990 solar eclipse in Finland aboard Finnair DC-9's chartered by Scientific Expeditions, Inc.. Immediately following the eclipse, convenient patch cords between cameras allowed us to make instant copies of our precious images, for exchange and for safety.

Most camcorders are equipped with a zoom lens having a range of 6:1 or 8:1. Some models even sport 10:1 or 12:1 zooms. In determining the size of the Sun's image, what you really need to know is the maximum focal length of your zoom and the size of the CCD or MOS detector. The diagonal dimension of most camera detectors is either 1/2-inch or 2/3-inch (check your camera manual). As displayed on a 13-inch diagonal (8 x 10 inch) television monitor, the Sun's image size is approximately equal to the focal length of the video lens multiplied by 0.40 (1/2-inch detectors) or 0.34 (2/3-inch detectors). For instance, a 1/2-inch CCD camera with a 66-mm lens gives an image of the Sun about 26 mm (1 inch) in diameter on a 13-inch television. This image size is fairly small but still produces an acceptable record of the diamond ring and corona, especially from a moving platform like a ship or plane.

Although few if any camcorders possess focal lengths long enough to give truly large images of the Sun, there is a solution. Several lens manufacturers now offer converter lenses for video cameras. These lenses simply screw onto the front of your camcorder's zoom and magnify the image by some set factor. Sigma makes 1.8X, 3X, and 5X video converters while Tokina offers 2X, 4X, 5.5X and 12X versions. Keep in mind that these converters are designed to be used at your zoom's longest focal length. While you can zoom back to slightly shorter focal lengths, you will quickly discover a point at which the converter vignettes the field of view around the corners and edges. An effective focal length of 200 to 300 mm produces an image of the Sun between 80 and 120 mm in diameter (1/2-inch detector with 13-inch-diagonal monitor) and provides good image scale while allowing room to include the corona surrounding the Sun. The accompanying table gives a range of image sizes for various focal lengths. If you're used to thinking in terms of 35-mm photography, you can convert video camera focal lengths to their approximate 35-mm equivalents by multiplying them by 5.2 (1/2-inch detectors) or 4.5 (2/3-inch detectors).

Unfortunately, video converters aren't cheap and they have several other disadvantages. At the larger image scales produced by converters, the eclipsed Sun will slowly drift out of your camcorder's field of view and you must be prepared to track it. Remember that the corona extends one or two solar radii beyond the Sun's limb. If you want to include the entire corona in your video, be sure to choose an image scale that allows two of three solar disks to fit within the camcorder frame. Furthermore, video converters introduce more glass elements into your camcorder's optical system, thereby increasing the possibility of internal reflections. However, video shot with a camcorder looking through a small spotting scope produced terrific images of the 1988 eclipse in spite of the additional optics. Are video converters for you? You must weigh the pro's and con's for yourself because even we can't come to an agreement (JMP doesn't care for them but FE thinks they're great!).

While high-magnification images are most interesting for viewing the form of the corona, wide-angle images are also fascinating. They can show the approach of the Moon's shadow and the shape of the shadow cone during totality. Outside the cone, one sees regions that are not experiencing totality. These regions appear much brighter and are also reddish, giving a sunset effect surrounding the horizon. Although shadow bands could conceivably be detected, they are of such low contrast that we are not aware of any shadow bands recorded on video. The sound track recorded with the video images captures the excitement of totality including the spontaneous exclamations and reactions of the crowd around you. You can also recite details of camera settings as you change them and record any verbal descriptions, impressions, or observations.

Normally, the solid and steady mounting of camera equipment is a prime requirement in eclipse photography. However, the 1990 airborne eclipse expedition to Finland precluded the use of tripods mounted on terra firma. Aboard the chartered Scientific Expeditions, Inc., Finnair DC-9's, our camcorders were mounted using everything from sophisticated Ken-Lab hand-held gyros (rented by the week from the E. P. Levine Co., Boston) to ordinary tripods taped to the aircraft floor and seat frames (cushioned with foam to dampen vibration) to just plain hand holding. Although there are different trade-offs with each technique, satisfactory images were achieved with all of them. The photos appearing with this article were shot directly off a TV screen using a 35-mm camera loaded with Kodak Gold 100 film and set at 1/15 second (slower than 1/30 second scan time, thus avoiding the diagonal bars that otherwise appear). The proper exposure of f/4 was determined by metering average scenery videotaped before the eclipse. During playback, the "pause" control on the Canon Hi-8 VCR gave a steady image on the screen to be photographed.

Since most eclipse chasers will view the 1991 eclipse from solid ground, a good camera support is mandatory. Even if you plan to view from a ship, a tripod with a smooth pan head or a monopod will allow you to support your camcorder far better than simply hand holding it. Since the Earth rotates 15° per hour, the Sun will appear to move 1° during the 4 minutes of totality in Hawaii and almost 2° in the 7 minutes of totality in Baja. Thus, you must use a low enough magnification to keep the Sun in the field of view or else move your camera during totality. You can choose to move the camcorder abruptly at one or two instants during totality, or else mount the camcorder on a tracking drive (or looking into a tracking mirror). If you are tracking, the sidereal rate is close enough to the solar rate since they differ by only 4 minutes of time per day.

To record the partial phases, solar filters must be used. Commercially available solar filters are of two types: glass and mylar. Glass filters (available from Thousand Oaks Optical or major telescope manufacturers) have thin chromium metal films deposited on their surfaces and produce a pleasing orange cast to the Sun. Mylar filters use a thin aluminum film that gives a decidedly blue cast to the Sun (available from Roger Tuthill, Inc., now in both ordinary and super-dense grade especially suitable for videotaping the solar disk). This blue color can be warmed up by sandwiching a Wratten 21 filter with the mylar filter. You can also produce your own solar filters by exposing a roll of black-and-white film to direct sunlight and overdeveloping it by 50%. Such homemade filters give a neutral or white image of the Sun. (Don't use color film; it contains none of the silver necessary to make the filtering safe to the eye or camera.) Check for internal reflections between the filter and the many lens surfaces; one advantage of videotaping is that you can detect any such reflected images directly in the viewfinder of your camcorder or on an external TV monitor.

The crucial task in videotaping (or photographing) an eclipse is to remove the solar filter at the correct moment for maximum beauty of the diamond ring. Since CCD's are more-or-less impervious to overexposure, you can remove the filter as soon as the first diamond ring begins. At the end of totality, replace the filter about 5 seconds after the second diamond ring appears. Of course, no solar filter is necessary during totality itself. However, the sensitivity of many camcorders is high enough that a small amount of neutral density filtering (~8X or ND=0.9) may be useful, especially if you would like to see more structure in the inner solar corona and prominences.

The more complex camcorders have some manual exposure controls that can be used to emphasize and differentiate between the inner and outer corona and prominences. Gain controls, manual diaphragms and variable shutter speeds can be adjusted during totality while monitoring their effects in the camcorder viewfinder. At the Finland eclipse, Dr. Marjorie Nicol ran through the available speeds of her camcorder shutter, and found that the highest speed gave the least overexposed image or the corona. You might want to imprint the date and perhaps the time on your eclipse video, especially if your camcorder displays the time to a precision of one second (which is rare), though you might prefer to have an unsullied image.

Since the full Moon is comparable in brightness to the solar corona and is about the same size as the solar disk (without the corona), it offers a convenient subject for evaluating the image scale and light level sensitivity of your camcorder and lens. Use the full Moon to test your system and avoid any unpleasant surprises on eclipse day. If your destination is Mexico, be sure your camcorder and tripod can point to the zenith (where the sun will be during totality) and is stable in that configuration. The 23° altitude of the eclipse from Hawaii allows normal tripod accessibility. It's also crucial to become thoroughly familiar with your equipment by setting it up and rehearsing for the eclipse. Remember to replace your battery with a freshly charged one within a half hour of totality and make sure that the camera is recording properly (It may say 'REC' in the viewfinder and you may even be able to see the tape moving or the counter advancing). Just remember to take some time out and actually watch the eclipse!

Additional details on videotaping solar eclipses can be found in Totality - Chapter 12.

Table 1

Video Camera Images of the Sun for Various Focal Lengths

          Focal Length        Size of Sun*     Size of Sun*               (mm)          (1/2-inch CCD)   (2/3-inch CCD)               50                20                17               60                24                20               70                28                24               80                32                27               100               40                34               150               60                51               200               80                68               250              100                85               300              120               102               350              140               119               400              160               136               500              200               170         * Size of solar disk in millimeters as seen on a            13-inch-diagonal television monitor.                          Multiply size by three to include one solar diameter            of corona on either side.

Video of the Total Solar Eclipse of 1994 November 3

Fred Espenak videotaped the total solar eclipse of 1994 Nov 3 from La Lava, Bolivia. The eclipse was captured on a Sony 8mm handicam, and was digitized with a Radius Video Vision Studio 2.0, Apple PowerMacintosh 8100/80, and Adobe Premiere 4.0. The result is available in two forms:

If you have the bandwidth, the patience, and a QuickTime player, we recommend the QuickTime version, not only because the resolution is better than in the MPEG movie, but also because the sound of the whooping eclipse afficionados is half the fun.

Video of the Total Solar Eclipse of 1995 October 24

Fred Espenak videotaped the total solar eclipse of 1995 October 24 from Dundlod, India. The eclipse was captured on a Sony Hi-8 handicam, and was digitized (using Adobe Premiere) and saved in two forms: