cable tech facts issue 102

In This Issue

• Understanding The Composite Video Signal Part II
• Producing The C(Color) Signal
• Interleaving The Y And C Signal

Back To Cable Tech Facts main index.

Understanding The Composite Video Signal-Part II

In Issue 101 of "Cable Tech Facts" we discussed the components that make up the composite video signal, including the luminance information, composite video amplitude and percentage of modulation. We left off with the test signals that are inserted in the vertical blanking interval.

In this issue we will pick up with the chrominance (color) information and how to measure its quality.

The chroma signal represents the color portion of the TV's picture. Three parameters need to be delivered to the television receiver in order to properly reproduce color, 1.) hue, 2.) saturation, and 3.) the luminance information. Hue, or tint, is the color itself, such as red, green, or yellow. Saturation refers to how little the color is diluted by white. Red, for example is fully saturated, while pink is a desaturated red. The more desaturated a color is, the closer it gets to white. The third parameter, luminance, is the brightness or shade of gray that the color would be if it were reproduced on a Black & White television.

Since the luminance information is already carried by the luminance portion or the composite video signal, only hue and saturation need to be added. These two parameters are combined at the video source to form the chrominance or "C" signal. The "C" signal is modulated onto a 3.58 MHz subcarrier and is combined with the Y Luminance signal to form the composite video signal. It is important to note that without the luminance information, the color information is virtually useless. By itself, the color information will only produce large, dark colored picture areas on the television screen. Unlike the luminance information which affects all three color guns in a color television equally, the color information controls the red, green or blue gun individually.

[Top]

Producing The C(Color) Signal

The easiest way to understand how the chroma signal is produced is to refer to an artist's color wheel, as shown in Figure 1. As you move around the wheel, the tint or hue of the color changes, and as you move farther out from the center, the color becomes more saturated.

Let's identify the specific color indicated by point "A". To do this, overlay an X-Y grid on top of the color circle shown in Figure 1. Starting in the center of the circle which is the intersection of the X and Y axis, move 5 units to the right on the X axis. Then, move 5 units up the Y axis. This defines the color as 5X and 5Y. In NTSC terms, the "X" axis is called the "B-Y" axis, and the "Y" axis is called the "R-Y" axis.

This type of color identification allows any combination of color tint and saturation to be easily identified. To transmit this information we could simply use two amplitude modulated signals (one representing the B-Y level and one representing the R-Y level). The only problem is that there is only room for one additional signal within the composite video signal.

To overcome this space limitation, the developers of the NTSC signal took the color identification one step further by combining the X and Y amplitude into a phase vector. Refer to the degree markings around the outside of the color wheel, starting with 0 degrees on the right X axis and increasing counter clockwise around the circle. Using basic trigonometry, you can identify point "A" at 45 degrees from the 0 phase starting point, and 3.53 units out from the X-Y intersection. In terms of color, point A has a tint of 45 degrees and a saturation level of 3.53 units.

What happens if we move to 1X and -5Y, the location of point "B"? Of course the phase and length of the vector change. Likewise, all points can be identified by a single vector. As the length or amplitude of the X and Y coordinates (signal levels) change, so does the position of the resulting vector. As you can see, a single vector can define any color on the color wheel. The phase of the vector defines the color's hue. The length of the vector defines the color's saturation.

Now, let's look at how this is done electrically. The block diagram in Figure 2 shows how the chrominance and luminance signals are produced. The incoming light is separated into its red, green and blue components by the pickup device inside the video camera at the broadcast studio. ( The R, G, and B levels may also be generated by a computer.) These are added in the proper proportion to produce the luminance or Y signal. The luminance signal is inverted and added to the red and green signals to produce the B-Y and the R-Y signal. (Most solid state pickup devices produce the B-Y and the R-Y signals directly.)

The B-Y and R-Y signals are combined into the "C" vector using two balanced modulators which are fed "quadrature phase" 3.58 MHz. This means that the 3.58 MHz signal fed to one balanced modulator is 90 degrees phase shifted from the 3.58 MHz signal that is applied to the other balanced modulator. In the balanced modulator the B-Y and the R-Y signals amplitude modulate the respective 3.58 MHz signals, much like an audio or video signal modulates an RF carrier. Because a balanced modulator is used, the 3.58 MHz signal cancels out leaving only the sideband information.

As different colors are applied, the signal amplitude at the balanced modulator output changes. (Think of the balanced modulator output as changing locations on the X (B-Y) and Y (R-Y) axis as in Figure 1 ). Now the two balanced modulator output signals are simply added together to produce a resulting C vector. The length and phase of this vector changes according to the amplitude of the B-Y and the R-Y signals, as seen in Figure 1.

As a camera scans a scene, the length and phase of the chroma vector constantly changes to reflect the instantaneous color. The television receiver undoes the modulation process to recreate separate red, green, and blue signals. A color burst is added to the composite video signal to reference the color demodulation. The burst is located during the back porch time or the horizontal blanking signal.

Most video systems modulate using the I and Q (in-phase and quadrature phase) axis as shown in Figure 1, rather than using the R-Y and B-Y axis. This is done to improve color resolution in the flesh tone region of the color spectrum. From a block diagram understanding, this method is the same as the B-Y and the R-Y modulation procedure we discussed earlier. (Because the demodulation process is easier, most television receivers shift the phase back and demodulate using the B-Y and R-Y axis).

[Top]

Interleaving The Y and C Signals

We now have just one signal that represents color, but we still need to combine the C signal with the luminance signal. We could simply put the chroma signals at a higher frequency than the 4.2 MHz luma signals. This, however, would increase the RF bandwidth needed to transmit the video signal. (The 4.2 MHz luminance limit is already a compromise to limit bandwidth).

Another solution is to limit the luminance signal frequencies and put the color signals in their place. However, this would limit picture resolution because the high frequency luminance signals add picture detail.

The solution that the NTSC developed is rather ingenious, and takes in account one important fact: the luminance information appears in clusters that are at multiples of the horizontal scan rate. This is illustrated in Figure 3. Two things need to be done to the chroma signal to take advantage of this: 1.) The luminance and chroma information must have similar spaced clusters, and 2.) the chroma clusters must be offset so that they will fit in between the luminance clusters. This process is called "interleaving."

In order to make the chroma information clusters have the same spacing as the luminance, the color subcarrier must be a phase-locked, harmonic of the horizontal line frequency. To offset the clusters, the subcarrier frequency must be 1/2 of an odd multiple of the horizontal line frequency. The color subcarrier frequency of 3.579545 and the horizontal line frequency of 15,734.264 Hz meet these requirements. Thus by meeting these objectives, the color signal can be "added" to the luminance signal with a minimum degredation to either the luminance or chrominance signal.

[Top]

[Sencore Home Page]