Test equipment first

To ensure valid test results, you must use good test equipment. Essentially, there are two primary requirements for this. First, the test equipment's specified performance must exceed the expected performance of the system being tested. And second, the test equipment must be performing to its specifications.

The latter requirement is really a two-stage requirement. The test equipment must always be setup using standard operator calibration procedures. For highest confidence, the equipment should also be fully calibrated periodically at a qualified service center. (Full and precise calibration of test equipment requires special skills and laboratory instruments not normally found in a video facility.)

As for test equipment capabilities, all the tests described in this application note can easily be performed with the Tektronix instruments shown in Figure 5-1:

• 1720 Vectorscope, used to evaluate the chrominance (color) information in a video signal
  
  
• 1730 Waveform Monitor, used for video signal level and timing measurements
  
  
• TSG 170A NTSC Television Generator, used to generate the necessary standard test signals

 

This test equipment is excellent for all basic and intermediate testing requirements. Other similar test equipment may be available to you and can be used in essentially the same manner. In some instances, the waveform monitor and vectorscope functions will be combined in a single instrument. This is the case, for example, with the Tektronix 1740A and 1750A Series Waveform/Vector Monitors. With these instruments, you select the waveform monitor or vectorscope mode by pressing the Waveform or Vector button on the front panel.

Similarly, more advanced, multi-function instruments, such as the Tektronix 1780R or VM700T Video Measurement Sets, allow push-button selection of waveform monitor or vectorscope modes. Additionally, the 1780R and VM700T provide numerous other features that simplify the measurements described in this section as well as many other measurements. Some of these features include microprocessor-assisted or fully automatic measurements, user-programmable measurement sequences, graphic display of measurements, and brighter displays in line-select mode.

The more sophisticated capabilities and enhanced precision of these instruments makes them ideal for all measurements including applications such as complete in-service or acceptance testing. (In-service testing also requires a VITS inserter, such as the Tektronix VITS200, VITS100 or 1910, to generate and insert vertical interval test signals.)

Setting Up the Test Equipment

Regardless of the specific test equipment models used, the basic test concepts and setup remains the same. The diagram in Figure 5-2 shows the setup for testing any video system, no matter how simple or complex.

 

Figure 5-2. The basic test setup for any video system involves applying a standard test signal at the path input and observing or measuring the resulting signal at the path output.

The block labeled Signal Path Being Tested can be any part of the video system. It can be a single piece of equipment such as a playback deck or a recording deck. Or the path can include a switcher along with other equipment such as cameras, VTRs, distribution amplifiers, character/ title generators, special effects generators, and so forth. In any case, all testing is done in essentially the same manner -- inject a signal at the start of the path (the input), and observe or measure any distortions to the signal at the end of the path (the output). For additional details, see the earlier section on Connecting and Terminating Instruments and "A word of caution".

Before actual testing begins, however, you should check the test equipment against its own internal calibration signal. This is typically referred to as a user calibration procedure, and it will be outlined in the instrument operator's manual.

Also, as a matter of reference, you should view an undistorted test signal on the test instrument's display. To do this, connect the test signal generator's output directly to the vectorscope or waveform monitor input terminals.

With the instruments properly connected and terminated, you'll see the test signal displayed on the vectorscope or waveform monitor. Study this display carefully. It's your reference of what the "perfect" signal should look like.

Ideally, you should see exactly the same "perfect" display after the test signal is passed through the video path being tested. However, if there are distortions in the signal path, you'll see corresponding changes in the displayed signal. Various types of distortion are described further in the following discussion of test procedures.

Insertion Gain

Insertion gain reflects the video path's ability to maintain correct signal amplitudes from input to output. The general test procedure is to apply a standard NTSC 1 volt (140 IRE) signal to the video path or equipment input. The test signal is then measured at various points along the signal path to verify its correct amplitude.

It is important to verify insertion gain before doing any other tests. If there are insertion gain errors and they are not corrected, subsequent tests will be incorrect as well. This is because most tests are based on the presumption that insertion gain is correct. Also equipment that is processing a nonstandard level signal will often introduce other distortions.

Insertion gain errors are usually expressed as a percent variation from the nominal value. There- fore, the goal is zero insertion gain.

Positive insertion gain errors indicate an increase (gain) in signal amplitude from input to output. This can lead to distortions from signal overload. Negative insertion gain indicates a decrease (loss) in signal amplitude. This causes dark pictures and also reduces the signal-to-noise ratio (SNR).

Insertion gain errors affect overall picture brightness and may also affect apparent color saturation. The human eye is very sensitive to even small brightness variations, particularly when they occur rapidly. Because of this, its especially important to make sure that all video switcher inputs are matched when different shots of the same scene are being combined. It should also be kept in mind that small gain errors in several system components can quickly cascade into big errors.

Insertion gain testing is done by applying either a SMPTE color bars signal or a full-field, 75% color bars test signal having a 100% white reference level. A waveform monitor is then used to observe the color bars signal and make insertion gain measurements.

The TSG 170A supplies the SMPTE color bars signal. Since the white reference for SMPTE color bars is at 75%, the 100 IRE Y signal that overlays the first two color bars, as shown in Figure 5-3, is more convenient for insertion gain tests. Figure 5-4 shows a full-field, 75% color bars signal with a 100% white reference. (Either of these signals can also be used to check chrominance amplitude and chrominance-to-burst phase on a vectorscope.)

 

Figure 5-3. A waveform monitor, set for one-line sweep and flat response, displays a properly adjusted SMPTE color bars signal. This test signal is used for detecting insertion gain and color burst amplitude errors.

 

Figure 5-4. Full-field, 75% color bars signal with 100% white reference.

The waveform monitor should be set for a one- or two-line sweep with the filter selection in flat. (Other filter positions alter the waveform monitor response and may mask a gain error.)

With the sweep and filter selected, use the vertical position control to align the trace's blanking level with 0 IRE on the graticule. As shown in Figure 5-3, the 100% white reference (Y signal) should now coincide with the 100 IRE graticule line, and the sync tips should be at 40 IRE. Insertion gain error, if any, is determined by noting the position of the white bar and subtracting 100%. For example, if the bar is at 93 IRE, insertion gain error is 93 - 100, or -7%.

When there is an insertion gain error, the equipment under test should be adjusted to remove the error. To do this, use the appropriate control on the equipment under test to align the color bars reference bar with the waveform monitor's 100 IRE graticule line. In doing this, be sure to keep the blanking level on 0 IRE by using the waveform monitor's position control if necessary. With the reference bar on 100 IRE, the sync and color burst amplitudes should both be 40 IRE peak-to-peak. You may need to make separate adjustments to make sure these levels are correct.

In making gain adjustments, it's important to start first with equipment at the beginning of the video signal path. Otherwise, you may simply mask rather than correct errors occurring earlier in the signal path.

After verifying that insertion gain is correct, you need to check chrominance amplitudes. This can be done using either a waveform monitor or a vectorscope.

On the waveform monitor, chrominance amplitude is checked by first measuring the color burst amplitude. It should extend from -20 IRE to +20 IRE (40 IRE peak-to-peak). Then check the maximum level of the first two color bars (yellow and cyan). Both should be at exactly 100 IRE. If they aren't, adjust the chrominance gain on the equipment under test to correct the chrominance levels.

Checking chrominance levels on the vectorscope is done with the display shown in Figure 5-5. The color burst vector should emanate from center screen along the horizontal axis (9 o'clock position). If it doesn't, adjust the vectorscope's phase control to put the color burst in the 9 o'clock position. Then check to see if all of the other vector dots fall into their proper boxes, as shown in Figure 5-5.

 

Figure 5-5. A vectorscope display of a SMPTE color bars test signal. All of the dots should be in their boxes when the color burst is aligned with the horizontal axis.

If the color vector dots do not fall within their boxes, you'll need to make chrominance gain or phase adjustments on the equipment under test. When the dots are beyond the boxes, toward the display edges, chrominance amplitude is too high. If the dots fall short, toward the display center, chrominance is too low. If the dots are rotated out of their boxes, chrominance phase is incorrect relative to color burst and needs to be adjusted on the equipment under test.

If all of the dots cannot be adjusted to fall in their respective boxes, other distortion errors exist in the system. These could include frequency response errors or other gain or linearity errors. Finding and correcting these errors requires further testing and adjustments by a qualified service technician.