"Roy Lewallen" wrote about broadcast television waveforms:
... No DC value is transmitted....
and
...The actual TV waveform is AC coupled, as it must be, and its
DC values are established in the receiver by the DC restorer...
___________________________

Please think about this. Standard TV color bars are produced by mixing the
outputs of three video pulse generators. The pulses are square waves, with
amplitude limits of 0 and 0.7VDC (~0.25µs transitions). These three video
waveforms are shown in Figure 7-1 in the paper hyperlinked at the bottom of
this post.

The DC present during the "on" time of these video pulses is
indistinguishable from the DC supplied by a continuous 0.7VDC source, over
the same, steady-state time interval. The DC and near-DC values of this
video must be available to the TV display device in order to accurately show
the color bar signal (and any other video waveform).

Low frequency content is lost when video is AC-coupled through a TV
transmitter (or any other circuit). A TV set DC restorer sets and
maintains the DC axis on which the pulse rides, but that does not correct
the distortion of the pulse waveform resulting from loss of its low
frequency content near zero hertz (DC).

Also recall that even the shortest pulse* that can pass undistorted through
the ~4MHz video bandwidth of the US broadcast TV standard contains most of
its energy at, and near zero hertz, e.g., DC.

*a sin² pulse with 0.25µs transitions

So DC _is_ , and must be conveyed by analog broadcast (and other) television
systems, because television wouldn't work well otherwise. It isn't DC as
you may think of it coming out of a battery, but it is DC, nevertheless, and
identical to battery DC of the same amplitude, when compared over equal,
steady-state intervals.

http://www.tek.com/Measurement/cgi-b...Set=television

RF