Analog Avenue

Tech Notes

Television Standards - Part I: How Our TVs Work
Mark Sauerwald,
National Semiconductor Corporation
Interface Products Group

Last night, I sat down with two of my kids and watched television with them. Dylan, my 10 year old, had command of the remote control and, to be honest, I haven't a clue as to what we were watching: But it was clearly aimed at an audience with different tastes from my own. As I sat there with my arm around Natalie, my mind began to wonder what sort of monster we engineers had wrought with video technology. I found it incongruous that the hard work of thousands of engineers, to develop a system of true beauty, be used to pour mindless drivel into the heads of innocent (my kids are angels!) children. I used to be proud to tell people that I worked on the chips that made TV better, but in watching this program, I wondered if I should admit to that any more. With this article, I will describe how our television systems work, and some of the history of why they are the way that they are. In particular I will be examining the analog video standard that is in common use today, in the U.S. In my next column, I will look at some of the standards in use in other parts of the world. The third installment will look at the digital standards that are coming on line so that we might better understand where we are headed. I can't help but believe that if they understood all that was behind the screen, the creators of some of today's programs would reach higher, and produce more top-notch programs like Bay Watch!

The video systems that we have now have evolved into being, rather than coming fully formed, into existence. When the first color TV system was developed in the U.S., it had several handicaps placed upon it that were legacies of the existing black and white system. The black and white system used a 0 - 4.2 MHz band, and it was decreed that color signals would occupy this same bandwidth, so that stations would not have to move. Another requirement was that the existing black and white receivers had to be able to display a picture when receiving a color signal. Given these constraints, the NTSC color system was developed. In describing the various standards below, I am painting with a very broad brush, ignoring many details that are vital to having the system work well. What my intention is, is to give you a feeling of what these standards are. If you are interested in the details, I highly recommend the book "Video Demystified" by Keith Jack and published by Harris Semiconductor. This book goes into much detail on all of the various video standards.

Black and White TV

The black and white (monochrome) system in existence split the picture up into horizontal lines, which were interleaved, or interlaced (first all of the even numbered lines are painted on the screen, then the odd numbered lines are painted in-between them.) See Fig. 1.

Figure 1 Interleaved Video: The red lines are painted on the screen first, then the beam goes back to the top of the screen and fills in the green lines.

Each line was broadcast with a signal which consisted of a pulse signifying the start of the line (known as a synchronization or sync pulse) followed by a signal in which a higher voltage indicated a whiter image, and lower voltages indicated darker parts of the image. As an example, if the image on the screen was a grey background with one white and one black vertical stripe, then the signal for a line would look like:

Some time is given between one video line and the next to allow the beam to sweep back to the left hand side of the cathode ray tube (CRT.) During this time, the voltage is held at a level below black, so that the retracing beam does not interfere with the image on the screen.

If we look at this signal in the frequency domain, since it is a repetitive signal, it's spectrum consists of a bunch of discrete lines, with a spacing of the line frequency (about 16 kHz), and tailing off so that there was no energy above 4.2MHz.

NTSC Color TV

Starting with the black and white system outlined above, the National Television Standards Committee came up with a color TV standard which occupies the same bandwidth as B&W, and when received by a B&W receiver, shows a good picture.

A color picture can be broken down into three monochrome pictures to represent the Red, Green and Blue (RGB) components of the image. Simply broadcasting these three pictures would have been the easy way to solve the problem, except that it would have cut the available channel space by a factor of 3, and when viewed on a B&W receiver, would have not resulted in a good picture. What the engineers on the NTSC (National Television Standards Committee) did was to define a new color space, Y, B-Y, R-Y, which takes advantage of the characteristics of the human eye. Some of these characteristics include:

The eye is more sensitive to how bright or dark an image is than its color. This is true both spatially and temporally.

As an object becomes smaller, the eye has a harder time discerning its color

The NTSC system takes advantage of this by using a color space in which one of the components, Y, (Luminance) is the brightness of the image. Y is derived by summing the R, G, and B signals with appropriate scale factors. The Y signal by itself is essentially the same as the old black and white signal. The B-Y and R-Y signals are generated by taking the B and R signals respectively, and subtracting off Y (with appropriate scaling factors), and together they represent the color content of the image. The trick now was to get the B-Y and R-Y information into the existing Y signal.

The Color Subcarrier:

To get the color information onto the Y (B&W) signal, a new signal, Chrominance (C) was defined:

C = {B-Y} sin(w t) + {R-Y} cos(w t)

where w is the color subcarrier frequency - which is about 3.6 MHz. By superimposing the C signal on the Y signal, a color signal is generated with all of the color information present. The signal looks like a 3.6 MHz sine wave, superimposed on the old B&W signal. The amplitude and phase of the sinusoid convey the color information. A reference burst of the sinusoid is added at the beginning of each line, after the sync pulse and before the active video, to make decoding the color information easier. In the figure below, an example line of a color-bar signal is shown:

This reference signal at the color subcarrier frequency and with a specific phase relationship to the subcarrier, is placed at the beginning of each line. This reference is commonly refered to as the "colorburst."

If a B&W TV receives this signal it usually does not have enough bandwidth to see the 3.6 MHz modulation, in which case successively darker grey bars are seen. If the receiver does have enough bandwith, the result will be that a color such as red will appear as alternating very narrow bands of two different dark greys, which the eye will see as one uniform grey.

One of the rather clever things that the NTSC did was careful selection of the subcarrier frequency. The spectrum of the subcarrier which is, like that for the Y signal, a line spectrum with spectral lines spaced at the 16 kHz line rate, has the lines placed right in-between the lines for the Y spectrum. In addition, whereas the bulk of the Y energy is at relatively low frequencies, the bulk of the color information is at high frequencies. This can be seen in the typical spectral plot shown below.

In most receivers, Luminance information is separated from the composite signal with a low pass filter, and the chroma (modulated Chrominance) is separated out with a band pass filter, in some receivers, special filters, known as 'comb' filters, actually manage to untangle the interleaved spectra to obtain a better quality picture.

Summary

In these few words, I hope that I have been able to give you a very broad brush description of how our TV systems work. I have skipped over many details that are vital if you are to fully understand the NTSC system. If you are interested in this topic and want to read more, then I would again suggest the book "Video Demystified" by Keith Jack. There are many books on this topic, but I have found Keith Jack's book to be a good balance between detail and clarity.

The NTSC standard, with which almost all of the United States watching it daily, has proven its ability to provide good quality programming to the masses. The fact that it is still going strong after four decades shows how well the standard was designed. In the next column, I will be looking at some of the problems that NTSC has, and how the Europeans attempted to mitigate these problems through slight differences that they made to the way that they display color TV information.

Editor's Note: The original National Television Standards Committee (NTSC) was set up to standardize the introduction of monochrome television and was reformed when the industry was ready, in 1952, to go ahead with a color standard. There were many debacles during the process including the approval of a color standard by the FCC which was not compatible with monochrome transmissions. The industry refused to accept the standard and instead pushed for an RCA-developed standard (the 'Red Book') to be the basis of color transmissions, what we now call NTSC. Apart from not understanding that there would be problems in recording and editing NTSC without defining one parameter (the phase between sync and burst, SC-H) the standard has worked well; however, what is not well understood outside broadcasting circles is that the use of the terminology "to RS-170A standards" to describe a "broadcast" quality signal from a product refers only to one drawing on a line waveform with timing, voltage and risetime information, the only thing that the reformed committee in that area got around to doing to update the RS-170 monochrome 525-line television standard. NTSC, it might be noted, is the composite TV color standard in the U.S., Mexico, Canada, Japan, Taiwan and S. Korea.

Analog Main | Product of the Week | Columns | Editorial | Tech Notes


		Click here to get your listing up.