Recent tune-up work I performed on a Toshiba 50HDX82 rear projection television (RPTV)
has prompted me to point out a weak point inherent in CRT-based RPTVs.
The core of the problem is that the optical path length from the CRT/lens assemblies to
the screen is quite short in comparison to the width of the (flat) screen. This short path
means that the 3 light beams ( Red, Green, Blue) have to spread out from the lenses at a
very wide angle to cover the width of the screen. Check out the simplified illustration
included in the version of this essay posted at my Home Theater Tune-Up site:
www.sonic.net/~dgibbons
Here are the resulting difficulties that occur:
Inability to achieve accurate focus across the entire screen.
Since the distance from the lenses to different points on the (flat) screen varies a lot
due to the wide spread of the light beam for each color, you can never get the entire
screen in focus. This could be corrected if a spherical (dome-shaped) screen was used, but
you can imagine trying to make such a screen work. (Think of the glare problem!) Someone
focusing a set of RPTV lenses has to decide if they want accurate focus at the center of
the screen, with the edges being well out of focus, or if they want so-so focus across the
bulk of the screen area.
When doing a mechanical focus (a physical adjustment of the lenses' focus,) one
position of the focus adjustment will get a good focus at the center of the screen, and
another position will produce good focus at the edges of the screen. Setting the focus
midway between these two points will result in an "average focus." The person
making the adjustment has to decide whether to go for an average focus or for best center
focus...
Color imbalances across the screen.
Again, the differences in angle (most likely) or distance from the three lenses to various
parts of the screen will result in differing brightness ratios between the colors on the
screen. Thus the color temperature (simply, the color of gray) on your TV will vary at
each side of the screen, and the "lens striping" technique (which you can read
about here on the Keohi HDTV web site) comes into play. However, blocking part of a lens'
light output will perhaps lead to a slight dimming at the edges of the screen as a cost of
correcting color imbalances. No free lunch.
Varying line widths across the screen.
The further each light beam travels from the lens assembly, the wider it gets, and
therefore the picture elements in that beam get wider with those longer distances, even if
the image is correctly focused on the screen. Light from the gun on the right side of the
cabinet has to travel less distance to get to the right side of the screen, while the
light from the gun on the left side has to travel further. Therefore, lines of equal width
in the video source signal will not be of the same width as they hit various parts of the
screen. This can help make convergence interesting at the edges of the screen, as lines
may not be the same width. Edges of supposedly white lines at the edges of the screen may
show some color no matter how careful the convergence.
(Problems 2 and 3 are interrelated, as brightness of the light beams falls as the beams
spread. This also can affect color balance.)
CRT-based Front Projectors which sit back a long ways from the projection screen will
suffer much less from these problems as the difference in distances and angle from the 3
projector lenses to the various parts of the screen is much smaller. This is a definite
advantage of front-projection systems.
Some of the digital-device-based RPTV's may do a little better, as they (I believe)
have all the light coming from one lens in the bottom of the cabinet. Still, problems 1
and (to a lesser degree) 3 may still be present in the larger screen sizes. (Physics will
get you every time...)
Mind you, even with these problems, many people use CRT-based RPTVs instead of
CRT-based front projector systems as the RPTV is convenient and more affordable. Pick your
Poison!
