Standard and procedures for video level measurement

STANDARD AND PROCEDURES FOR VIDEO LEVEL MEASUREMENT

VIDEO LEVEL MEASUREMENT

Setting video levels would be easy if only the cable operator could call up a test pattern when he was ready to set levels. In that sense the broadcast engineer has it much easier. The cable-TV headend technician must make-do with whatever TV signal happens to be on at the time he is ready to set levels, A video waveform can be subjected to a wide variety of measurement that are outside the scope of this paper. We will limit our discussion to measurement of the voltage of the video waveform.

Video level measurement trouble can exist because the peak voltage, as seen on a wideband oscilloscope, and maximum brightness are not necessarily the same thing. Only the luminance component of the signal contributes to the picture brightness. Thus, if there is considerable color saturation (chrominance) in the picture at the time that peak-to-peak video voltage is measured, the scope reading could be 10 percent to 20 percent higher than the actual luminance component of the video signal.

The chrominance component of a color television signal can be found clustered about the color burst and occupies the frequency bank from about 3 Mhz to 4 Mhz. The luminance component of the color television signal is located primarily in the 2 kHz to 600 kHz frequency band. Fine luminance picture detail does of course overlap the band occupied by chrominance, but since the energy of the luminance component in the 3 Mhz to 4 Mhz band is usually quite low, it may be filtered out to make luminance measurements without significantly impairing test accuracy.

Since only the luminance component of the video waveform contributes to picture brightness, a filter must be provided to separate the chrominance signal from the luminance signal when the brightness produced by a given video waveform is to be measured.

A talented video professional can closely estimate the actual luminance amplitude on a test pattern, even when buried by chrominance signals.

However, even the experts disagree when the picture content continually changes. Technicians often interpret the waveform they observe differently according to their own experience and the setting of the oscilloscope controls. This is particularly true if a regular oscilloscope is used. Even with a video waveform monitor a significant reading error will occur if the wrong video low-pass filter is switched in.

New technology

An entirely new type of test instrument has recently become available, called the VVM Video Volt Meter, that measures sync, white and composite video amplitude on a digital scale instead of an oscilloscope. This video volt meter is the size of an ordinary had-held digital voltmeter, and is battery-operated for portability. Since the VVM read-out is digital, video level setting becomes more consistent.

This measurement technique recognized that high frequency chroma information can cause error in TV luminance measurement, so a special filter was proposed that retains the luminance part of the TV signal but filters out the chroma component. This filter is generically known as the “IRE” filter (see Figure 1). Television waveform monitors are equipped with this filter (as well as others) which must be switched in when making video voltage measurements.

(Unfortunately, ordinary scopes do not have such a filter. If you do not have a proper television waveform monitor with the IRE filter, do not despair, you can obtain an IRE filter to connect to your scope. With this filter connected to the broad band scope you can at least get the same volts peak-to-peak reading as with a standard video waveform monitor.)

The IRE (now IEEE) standard measures video I IRE units instead of volts peak-to-peak. Since most cable companies and equipment manufacturers consider 1.0 Vp-p as representing the maximum brightness of a video picture, then this translates to 140 IRE units. Of that, 40 IRE units constitute the amplitude of the sync pulse as measured from the “back porch,” and 100 IRE units constitute maximum brightness, also as measured from back porch (see Figure 2). For 1.0 Vp-p video signal, each IRE unit represents 7.14 millivolts peak-to-peak. So when the video signal is 1.00 volt peak-to-peak, then the sync pulse should be 0.2857 volts or 28.6 percent and the picture component should be 71.4 percent of the composite video signal.

Waveform monitors are also equipped with a special scale (see Figure 3) calibrated in IRE units. However, it is not necessary to use this unit of measurement when the main interest is to set video to 1 Vp-p everywhere that the television signal appears at baseband in the headend. Thus, an ordinary scope, without the special IRE graticule can be used , as long as an external IRE filter is used with it to remove the chrominance component.

The maximum video carrier power occurs during the tip of the sync pulse, therefore -40 IRE units corresponds to 100% carrier power (see Figure 2). Also note that 100 IRE units equals 12.5% video carrier power, not zero carrier power. Since 12.5 carrier power corresponds to 87.5% depth of modulation (100% – 12.5% = 87.5%), then this is the maximum modulation permitted for the luminance component of the video signal. The extra 12.5% remaining of the video carrier is reserved for the chrominance components, otherwise color would be “wiped out” on bright scenes. These relationships reveal why “scope” readings (without the IRE filters) can easily result in serious misadjustments of the TV modulator.

Measurement of the video signal amplitude can be made without interrupting service by including a “BNC-T” connector in the cabling between the baseband video source (i.e. the satellite receiver video output connector) and the input to the television modulator. Since the video volt meter has a high impedance input, the video signal can be measured without changing the level when the meter is connected.

The video volt meter or waveform monitor, or scope with the IRE filter, should be directly connected to the BNC-T because a long cable attached at this point could create reflections that may impair the picture by causing ghosts or ringing to appear. It is particularly useful to provide a panel-mounted BNC-T so that all of the TV modulators in one rack of equipment can be measured from the front of the rack. A short 10-inch to 12-inch cable can be used to connect the video voltmeter to this panel-mounted connector without impairing the video signal.

Accuracy

Even with an IRE filter, scopes do get out of adjustment so that calibration for 1.0 Vp-p becomes “iffy.” Also, the waveform amplitude can be read differently according to the interpretation of each operator.

A digital video volt meter ensures consistent video voltage readings independent of operator “eye.” It reads sync amplitude, white luminance amplitude and peak-to-peak composite video amplitude. The scale is calibrated in IRE units as well as volts peak-to-peak. Since the meter has a digital read-out, there is a minimum of interpretation required. The IRE filter is built-in, insuring measurement according to IRE standards.

The video volt meter has a basic accuracy of 1 percent ±1 IRE unit or 1 percent ±0.01 volts, so readings taken with this meter will be many times more accurate than even a recently calibrated scope.

Conclusion

This paper emphasizes that video peak-to-peak readings must be made with a filter that blocks out the chrominance component and that measurements made with a standard broadband oscilloscope can lead to substantial errors in headend level setting. Video waveform monitors are available that enable video level measurements according to IRE standards, but these monitors are quite expensive. Ordinary scopes can be used, provided that an IRE roll-off filter is used in conjunction with them. Finally, it is proposed that peak-to-peak voltage measurements of video can be made much more accurately and conveniently by a digital video volt meter than with a waveform monitor.