1

TelevisionMeasurements

NTSC Systems

Copyright © 1998, Tektronix, Inc. All rights reserved.

2

Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Good Measurement Practices . . . . . . . . . . . . . . . 4

EQUIPMENT REQUIREMENTS . . . . . . . . 4

CALIBRATION . . . . . . . . . . . . . . . . . . . . . 6

INSTRUMENT CALIBRATION . . . . . . . . 6

DEMODULATED RF SIGNALS . . . . . . . . 7

TERMINATION . . . . . . . . . . . . . . . . . . . . 7

STANDARDS AND PERFORMANCE GOALS . . . . . . . . . . . . . 7

Waveform Distortions andMeasurement Methods . . . . . . . . . . . . . . . 8

I. VIDEO AMPLITUDE AND TIME MEASUREMENTS . . . . . . . . 8

Amplitude Measurements . . . . . . . . . 9

Time Measurements . . . . . . . . . . . . . .12

SCH Phase . . . . . . . . . . . . . . . . . . . . .15

II. LINEAR DISTORTIONS . . . . . . . . . .18

Chrominance-to-Luminance Gain and Delay . . . . . . . . . . . . . . . .19

Short Time Distortion . . . . . . . . . . . .24

Line Time Distortion . . . . . . . . . . . . .26

Field Time Distortion . . . . . . . . . . . . .28

Long Time Distortion . . . . . . . . . . . . .30

Frequency Response . . . . . . . . . . . . .31

Group Delay . . . . . . . . . . . . . . . . . . . .36

K Factor Ratings . . . . . . . . . . . . . . . . .38

III. NONLINEAR DISTORTIONS . . . . . .41

Differential P h a s e . . . . . . . . . . . . . . . .42

Differential Gain . . . . . . . . . . . . . . . .46

Luminance Nonlinearity . . . . . . . . . .50

Chrominance Nonlinear P h a s e . . . . . . . . 5 2

Chrominance Nonlinear Gain . . . . . . . . 5 3

Chrominance-to-Luminance

Intermodulation . . . . . . . . . . . . . . . .54

Transient Gain Distortion . . . . . . . . . .56

Dynamic Gain Change . . . . . . . . . . . .57

IV. NOISE MEASUREMENT . . . . . . . . .58

Signal-to Noise Ratio . . . . . . . . . . . . .59

V. TRANSMITTER MEASUREMENTS . . 6 1

ICPM . . . . . . . . . . . . . . . . . . . . . . . . .62

Depth of Modulation . . . . . . . . . . . . .64

GLOSSARY OF TELEVISION TERMS . . . . . . . . . . . . 65

APPENDICES

APPENDIX A - NTSC COLOR BARS . . . .68

APPENDIX B - SINE-SQUARED PULSES 70

APPENDIX C - RS-170A . . . . . . . . . . . . . .71

APPENDIX D - FCC 73.699, FIGURE 6 . .73

To characterize television systemperformance, an understandingof signal distortions and mea-surement methods as well asp roper instrumentation is needed.This booklet provides informa-tion on television test and mea-surement practices and serves asa comprehensive reference onmethods of quantifying signaldistortions. This publicationdeals with NTSC composite ana-log signals. Analog component,digital composite and compo-nent, and HDTV measurementsare outside its scope.

New instruments, test signals,and measurement procedures areintroduced as television test and measurement technologyevolves. This booklet encom-passes both traditional measure-ment techniques and newermethods. After a discussion ofgood measurement practices,five general categories of television measurements are addressed:

I. Video Amplitude and Time Measurements

II. Linear Distortions

III.Nonlinear Distortions

IV. Noise

V. Transmitter Measurements

A basic knowledge of video isassumed and a glossary of com-monly used terms is included asa refresher and to introduce newconcepts. This booklet does notprovide detailed instructions onhow to use particular instru m e n t s .The basics of waveform monitorand vectorscope operation areassumed. Consult the instrumentmanuals for specific operatinginstructions.

3

Preface

4

EQUIPMENT REQUIREMENTSTelevision system performanceis evaluated by sending test sig-nals with known characteristicsthrough the signal path. The sig-nals are then observed at theoutput (or at intermediatepoints) to determine whether ornot they are being accuratelytransferred through the system.Two basic types of television testand measurement equipment arerequired to perform these tasks.Test signal generators providethe stimulus and specializedoscilloscopes, known as wave-form monitors and vectorscopes,provide the tools for evaluatingthe response.

Test Signal Generators. Te l e v i s i o nsignal generators provide a widevariety of test and synchroniza-tion signals. Two key criteria inselection of a test signal genera-tor for precision measurementsare signal complement and accu-racy. The generator should provide all of the test signals tosupport the required measure-ments and the signal accuracymust be better than the toler-ances of the measurements to bemade. If possible, the generatoraccuracy should be twice as

good as the measurement toler-ance. For example, differentialgain measurement to 1% accura-cy should be made with a gener-ator having 0.5% or less differ-ential gain distortion.

Television equipment and sys-tem performance is generallyassessed on either an out-of-ser-vice or in-service basis. In broad-cast television applications,measurements must often bemade during regular broadcasthours or on an in-service basis.This requires a generator capableof placing test signals within thevertical blanking interval (VBI)of the television program signal.Out-of-service measurements,those performed on other thanan in-service basis, may be madewith any suitable full field testsignal generator.

For out-of-service measurements,the Tektronix TG2000 SignalGeneration Platform with theAVG1 and AGL1 modules is therecommended product. TheAVG1 Analog Video Generatorprovides comprehensive signalsets and sufficient accuracy forvirtually all measurementrequirements. The AVG1 is alsoa multiformat unit capable of

supporting measurements inother composite and analog com-ponent formats. This eliminatesthe need for additional signalgeneration equipment wherethere is the requirement for mea-surements in multiple formats.For synchronization of theequipment under test, a blackburst re f e rence signal is pro v i d e dby the TG2000 mainframe. Forapplications requiring the testsignal source be synchronouswith existing equipment, theAGL1 Analog Genlock moduleprovides the interface needed tolock the TG2000 to an externalblack burst reference signal.

For in-service measurements, theTektronix VITS200 Generatorand Inserter is the recommendedproduct. The VITS200 providesa full complement of NTSC testsignals and high degree of flexi-bility in placement of these signals within the VBI. Signalaccuracy is adequate for mosttransmission and transmittermeasurement requirements.

Both the TG2000 and VITS200fully support the measurementcapabilities of the 1780R andVM700 Series VideoMeasurement Sets.

Good Measurement Practices

5

Waveform Monitors andVectorscopes. The instrumentsused to evaluate a system'sresponse to test signals make upthe second major category oftelevision test and measurementequipment. Although some mea-surements can be performedwith a general purpose oscillo-scope, a waveform monitor isgenerally preferred in televisionfacilities. Waveform monitorsprovide TV triggering capabili-ties and video filters that allowevaluation the chrominance andluminance portions of the signalindependently. Most models alsohave a line selector for lookingat signals on individual lines.

A vectorscope, which demodu-lates the signal and displays R-Yversus B-Y, is another importanttest and measurement tool. Witha vectorscope, the chrominanceportion of the signal can beaccurately evaluated.

When selecting a waveformmonitor and vectorscope, care-fully evaluate the feature setsand specifications to make surethey will meet present andfuture needs. This is particularlytrue if making accurate measure-ments of all signal parametersand distortions described in this

booklet. Many varieties of wave-form monitors and vectorscopesare available today but themajority of them are not intend-ed for precision measurementapplications. Most vectorscopes,for example, do not have preci-sion differential phase and gainmeasurement capabilities.

The recommended products forprecision measurement applica-tions are the Tektronix 1780Rand the VM700T. Most of themeasurement procedures in thisbooklet are based on theseinstruments.

The 1780R provides waveformmonitor and vectorscope func-tions as well as many special-ized measurement features andmodes that simplify complexmeasurements.

The VM700T is an automatedmeasurement set with resultsavailable in numeric and graphicform. Waveform and vector dis-plays, similar to those of tradi-tional waveform monitors andvectorscopes operating in lineselect mode, are also provided.The VM700T Measure mode pro-vides unique displays of mea-surement results, many of whichare presented in this book.

A waveform monitor display of color bars.

A vectorscope display of color bars.

CALIBRATIONMost instruments are quite stable over time, however, it isgood practice to verify equip-ment calibration prior to everymeasurement session. Manyinstruments have internally gen-erated calibration signals thatfacilitate this process. In theabsence of a calibrator, or as anadditional check, a test signaldirectly out of a high qualitygenerator makes a good substi -tute. Calibration procedures varyfrom instrument to instrumentand the manuals containdetailed instructions.

Analog CRT-based instrumentssuch as the 1780R have a speci-fied warm up time, typically 20 or 30 minutes. Turn theinstrument on and allow it tooperate for at least that longbefore checking the calibrationand performing measurements.This ensures that the measure-ment instrumentation will have little or no effect on themeasurement results.

Computer-based instrumentssuch as the VM700T also specifya warm up time but the operatordoes not need to verify or adjustthe calibration settings. TheVM700T will automatically cali-brate itself when it is turned onand will continue to do so peri-odically during operation. Forbest results, wait 20 or 30 min-utes after initial turn-on beforemaking any measurements.

INSTRUMENT CONFIGURATIONMost of the functions on analogwaveform monitor and vec-torscope front panels are fairlys t r a i g h t f o rw a rd and have obviousapplications in measurementprocedures. A few controls,however, might need a bit more explanation.

DC Restorer. The basic function ofthe DC restorer in a waveformmonitor is to clamp one point ofthe video waveform to a fixedDC level. This ensures that thedisplay will not move verticallywith changes in signal amplitudeor APL (Average Picture Level).

Some instruments offer a choiceof slow and fast DC restorerspeeds. The slow setting is usedto measure hum or other low fre-quency distortions. The fast setting removes hum from thedisplay so it will not interferewith other measurements. Backporch is the most commonlyused clamp point, but sync tipclamping has some applicationsat the transmitter.

6

The 1780R waveform calibrator.

The 1780R vectorscope calibration oscillator.

7

AFC/Direct. This control providesselection of the method of trig-gering the waveform monitorhorizontal sweep. The ramp thatproduces the horizontal sweep isalways synchronous with the Hor V pulses of the referencevideo and is started either by thepulses themselves (Direct) or bytheir average (AFC).

In the direct mode, the videosync pulses directly trigger thewaveform monitor's horizontalsweep. The direct setting shouldbe used to remove the effects oftime base jitter from the displayin order to evaluate other para-meters. Since a new trigger pointis established for each sweep,line-to-line jitter is not visible inthis mode.

In the AFC (Automatic Fre q u e n c yControl) mode, a phase-lockedloop generates pulses that repre-sent the average timing of thesync pulses. These averagedpulses are used to trigger thesweep. The AFC mode is usefulfor making measurements in thepresence of noise as the effectsof noise-induced horizontal jitterare removed from the display.

The AFC mode is also useful forevaluating the amount of timebase jitter in a signal. The leadingedge of sync will appear wide(blurred) if much time base jitteris present. This method is veryuseful for comparing signals orfor indicating the presence of jit-ter but be cautious about actuallytrying to measure it. The band-width of the AFC phase-lockedloop also affects the display.

75%/100% Bars. Some vectorscopeshave a 75%/100% selection onthe front panel. This settingchanges the calibration of thevectorscope chrominance gain toaccommodate two differenttypes of color bars. The 75%/100% distinction refers to ampli -tude, not to saturation or whitebar level. These issues are dis-cussed in detail in Appendix A.

75% bars are most frequentlyused in NTSC systems as thelarge chrominance peaks in

100% bars may overload thetransmitter. However, some testsignal generators produce both.It is important to know whatamplitude color bar is beingused and to select the corre-sponding gain setting on the vectorscope. Otherwise chromi-nance gain can easily be misadjusted.

DEMODULATED RF SIGNALSAll of the baseband measure m e n t sdiscussed in this booklet mayalso be made on demodulatedRF signals. It is important, how-ever, to eliminate the demodula-tor itself as a possible source ofdistortion. Measurement-qualityi n s t ruments such as the Te k t ro n i xTV1350 and 1450 TelevisionDemodulators will eliminate thelikelihood that the demodulatoris introducing distortion.

TERMINATIONImproper termination is a verycommon source of operator errorand frustration. Always makesure the signal under measure-ment is terminated with a 75 Ohmterminator in one location. It isgenerally best to terminate at thefinal piece of equipment in thesignal path.

The quality of the terminator isalso important, particularlywhen measuring very small dis-tortions. Select a terminator withthe tightest practical tolerance asincorrect termination impedancecan cause amplitude errors, fre-quency response problems, andpulse distortions. Terminators in the 1/2% to 1/4% tolerancerange are widely available andare generally adequate for routine testing.

STANDARDS ANDPERFORMANCE GOALSNo one standard defines allamplitude and timing relation-ships for the NTSC signal. Thereare a number of reference docu-ments produced by differentorganizations, several of whicha re in common use today. RS-170Aand FCC 73.699 Figure 6, two of

the most frequently used, arereproduced in Appendices C and D of this booklet. Both docu-ments define blanking and synchronizing signal parameters.RS-170A includes references toSCH phase and is generally usedin studio environments. TheFCC diagram is used to verifythe quality of transmitted sig-nals. Use them as a reference butexercise caution in assumingthat compliance with these stan -dards is mandatory or that com-pliance is sufficient to ensuresignal quality.

Acceptable levels of distortionare usually determined subjec-tively but a number of organiza-tions publish documents thatprovide recommended limits.These standards, which includeEIA-250-C, are frequently editedand revised. Each facility ulti -mately needs to determine itsown perf o rmance goals, however,these documents can providesome good guidelines.

While there is usually agreementabout the nature of each distor-tion, definitions for expressingthe magnitude of the distortionvary considerably from standardto standard. A number of ques-tions should be kept in mind. Isthe measurement absolute or rel -ative? If it is relative, what is there f e rence? Under what conditionsis the reference established? Isthe peak-to-peak variation or thelargest single deviation to bequoted as the distortion?

A misunderstanding of any oneof these issues can seriouslyaffect measurement results so itis important to become familiarwith the definitions in whateverstandards are used. Make surethose involved in measuring sys-tem performance agree on howto express the amount of distor-tion. It is good practice to recordthis information along with themeasurement results.

This section deals with two fun-damental pro p e rties of the signal,amplitude and time. In thesetwo dimensions, problems aremore frequently caused by oper-ator error than by malfunction-ing equipment. Correction ofamplitude and pulse width prob-lems often simply involves pro p e radjustment of the equipment thesignal passes through.

Two kinds of amplitude mea-surements are important in tele-vision systems. Absolute levels,such as peak-to-peak amplitude,need to be properly adjusted.The relationships between theparts of the signal are alsoimportant. The ratio of sync tothe rest of the signal, for example,must be accurately maintained.

Composite NTSC video signalsare nominally 1 volt peak-to-peak. Amplitudes are also some-times described in terms of theIRE scale, which divides thevideo signal into 140 equalparts. Strictly speaking the IREscale is a relative one and can beused to compare parts of the signal regardless of overall

amplitude. In practice, however,the IRE scale is sometimes tre a t e das an absolute scale with a directrelationship to volts. In the1780R, 1 IRE is defined as anabsolute unit equal to 1/140 of 1 volt. With the VM700T, theuser can define IRE to be relativeto zero carrier or the white baror to be an absolute unit (100 IRE = 714 mV).

When setting video amplitudes,it is not sufficient to simplyadjust the output level of thefinal piece of equipment in thesignal path. Every piece ofequipment should be adjusted toappropriately transfer the signalfrom input to output. Televisionequipment is generally notdesigned to handle signals thatdeviate much from the nominalamplitude. Signals that are toolarge may be clipped or distortedand signals that are too smallwill suffer from degraded signal-to-noise performance.

In most television facilities,video amplitudes are monitoredand adjusted on a daily basis.Signal timing parameters areusually checked less frequently,however, it is still important tounderstand the measurementmethods. A periodic verificationthat all timing parameters arewithin limits is recommended.

This booklet does not addresssystem timing issues which dealwith relative time relationshipsbetween the many signals in atelevision facility. Although system timing is critical to pro-duction quality, it is outside thescope of this publication. Onlythose timing measurements that relate to a single signal are addressed.

8

I. VIDEO AMPLITUDE AND TIME MEASUREMENTS

Waveform Distortions And Measurement Methods

DEFINITIONVideo amplitudes are most fre-quently measured in order toverify that they conform to nom-inal values. The gain must beadjusted if signals are too largeor too small. Similar methods ofevaluating the waveform areused for both measurement andadjustment of signal levels.

M e a s u rements of the peak-to-peakamplitude of video signals aresometimes known as insertiongain measurements. For NTSCsystems, the nominal peak-to-peakamplitude is 1 volt (140 IRE).

PICTURE EFFECTSInsertion gain errors cause thepicture to appear too light or toodark. Because of the effects ofambient light, apparent color saturation is also affected.

TEST SIGNALSInsertion gain is most easilymeasured with a test signal thatcontains a 100 IRE white level.Color bars and pulse and bar signals are most frequently used(see Figures 1 and 2).

MEASUREMENT METHODSWaveform Monitor Graticule. Signalamplitude can be measured witha waveform monitor by compar-ing the waveform to the verticalscale on the graticule. With thewaveform monitor vertical gain

in the calibrated setting (1 voltfull scale), the signal should be 1 volt (140 IRE) from sync tip to peak white (see Figure 3). The graticule in the VM700TWAVEFORM mode can be usedin a similar manner.

Added Calibrator Method. Somewaveform monitors have a fea-ture that allows the internal cali-brator signal to be used as a ref-erence for amplitude measure-ments. This feature is known asWFM + CAL in the 1780R. In the1480 it is accessed by depressingboth the CAL and OPER buttons.

The WFM + CAL display consistsof two video traces vertically off-set by the calibrator amplitude.This display is obtained by addingthe incoming signal to a calibrateds q u a re wave of known amplitude.Signal amplitude is equal to thecalibrator amplitude when thebottom of the upper trace and thetop of the lower trace coincide.

The WFM + CAL mode is mostcommonly used to set insertiongain which requires a 1-volt cali-brator signal. If using a 1780R,select a calibrator amplitude of1000 mV (140 IRE). In the 1480,the DC RESTORER setting deter-mines which of two calibratoramplitudes is currently selected.The calibrator amplitude is 1 voltwhen SYNC TIP is selected and 714 millivolts when BACKPORCH is selected.

9

Amplitude Measurements

Figure 1. A 75% amplitude color bar signal with a 100% white reference bar.

Figure 2. A composite signal (also known as FCC Composite) that includes a 100 IRE white bar.

Figure 3. A 1-volt signal properly positioned with respect to the 1780R graticule.

I n s e rtion gain is set by extern a l l yadjusting the signal amplitudeuntil sync tip of the upper traceand peak white of the lowertrace coincide. Figure 4 shows aproperly adjusted signal. Sincethe waveform monitor verticalgain need not be calibrated inthis mode, the gain may beincreased for greater resolution.

The 1780R has a variable cali-brator so the WFM + CAL modecan be used to measure signalamplitudes. Measurements aremade by adjusting the calibratoramplitude (with the large knobon the 1780R front panel) untilthe traces meet. At this point thecalibrator amplitude equals thesignal amplitude and can beread from the screen.

Voltage Cursors (1780R). Somewaveform monitors, such as the1780R, are equipped with on-s c reen voltage cursors for makingaccurate amplitude measure-ments. Peak-to-peak amplitudecan be measured by positioningone cursor on sync tip and theother on peak white (see Figure 5).The 1780R vertical gain control

affects the cursors and the wave-form in the same manner so ver-tical gain can be increased to allow for more accurate positioning of the cursors.

When setting insertion gain, itmay be convenient to first setthe cursor separation for 1000mV (140 IRE). The video signalamplitude should then be adjustedto match the cursor amplitude.

Cursors (VM700T). Manual ampli-tude measurements can be madewith the VM700T by selectingCURSORS in the WAVEFORMmode. The horizontal baseline inthe middle of the screen is usedas a reference. To measure inser-tion gain, first position sync tipon the baseline. Touch the RESETDIFFS selection on the screen toreset the voltage difference tozero. Now move the waveformdown until the white bar is onthe baseline and read the voltagedifference from the screen.

VM700T Automatic Measurement.The VM700T provides ampli-tude measurements in the AUTO mode.

10

Figure 4. The WFM + CAL mode in the 1780R indicating that insertion gainis properly adjusted.

Figure 5. 1780R voltage cursors positioned to measure peak-to-peak amplitude.

NOTES1. Sync to Picture Ratio. When thesignal amplitude is wrong, it isimportant to verify that the prob-lem is really a simple gain errorrather than a distortion. This canbe accomplished by verifyingthat the ratio of sync to the pic-ture signal (the part of the signalabove blanking) is 40:100. If theratio is correct, proceed with thegain adjustment. If the ratio isincorrect, there is a problem andfurther investigation is needed.The signal may be suffering fromdistortion or equipment thatreinserts sync and burst could be malfunctioning.

2. Sync and Burst Measurements.Sync and burst should each be40/140 of the composite videosignal (286 millivolts for a 1-voltsignal). Most of the methods discussed in this section can beused to measure sync and burstamplitudes. When using the1780R voltage cursors, theTRACK mode is a convenienttool for comparing sync andburst amplitudes. In this mode,the separation between the twocursors remains fixed and theycan be moved together withrespect to the waveform.

3. Measurement Accuracy. In gen-eral, the added calibrator andvoltage cursor methods are moreaccurate than the graticule tech-nique. However, some cursorimplementations have far moreresolution than accuracy cre a t i n gan impression of measurementsmore precise than they reallyare. Familiarity with the specifi -cations of the waveform monitorand an understanding of theaccuracy and resolution avail-able in the various monitoringmodes will help make an appropriate choice.

4. Using the Luminance Filter. W h e nsetting insertion gain with a livesignal rather than a test signal, it may be useful to enable thewaveform monitor luminancefilter (also called lowpass orIRE). This filter removes thechrominance information so thatpeak white luminance levels canbe used for setting gain.

5. White Levels. When settinginsertion gain, make sure a 100IRE bar is used as the referencepeak white level. As noted inAppendix A, some color bar signals have a 77 IRE white barrather than 100 IRE.

6. Setup. Most NTSC signals use"black level setup" which isoften simply referred to as"setup". In a signal with setup,video black is 7.5 IRE above theblanking level. The peak-to-peakamplitude and sync amplitudedo not change. The peak whitelevel therefore remains at 100IRE so the 7.5 IRE pedestalreduces the amplitude rangeavailable for picture information.Both luminance and chromi-nance levels of the entire signalare scaled down in order to fitinto the 92.5 IRE range whichremains above the pedestal.

Virtually all NTSC programmaterial has setup but many testsignals do not. When measuringamplitudes, it is necessary toknow whether or not the signalhas setup and to understand thedifferences associated with it.When working with signals thathave setup, do not to confuse theblack level (7.5 IRE) with theblanking level.

11

12

DEFINITIONHorizontal and vertical synchro-nization pulse widths are mea-sured in order to verify they fallwithin specified limits. Risetimes, fall times, and the posi-tion and number of cycles inburst are also specified.

Both RS-170A and the FCC provide recommended limits forthese parameters. However, thetwo standards have different definitions for the various timeintervals. For example, the FCCspecifies sync width between the90% (-4 IRE) points of the twotransitions while RS-170A speci-fies sync width between the50% (-20 IRE) points. The defin-ition for each parameter shouldbe clearly understood beforemeasuring it.

Even when differences in defini-tion are taken into account, RS-170A and the FCC give dif-ferent recommended limits forthe various sync pulse parame-ters. The RS-170A requirementsare generally more stringent.Pulse width requirements for the two standards are given inFigures 6 and 7.

Time Measurements

COLORBACKPORCH1.6 µ SECBREEZWAY

.6 µ SEC

ZZ

4 IRE4 IRE7.5 IRE

BLANKINGLEVEL

REF BLACKLEVEL

FRONTPORCH

1.5 µ SEC± .1 µ SEC

BLANKING10.9 µ SEC ± 0.2 µ SEC

SYNC TO BLANKING END9.4 µ SEC = ± .1 µ SEC

SYNC TO BURST END7.8 µ SEC

SYNC4.7 µ SEC ± .1 µ SEC

BURST2.5 µ SEC

40 IRE(286 mv)

20 IRE

40 IREREF BURSTAMPLITUDE

REF SYNCAMPLITUDE

100 IRE(714 mv)

REF WHITE

Figure 6. RS-170A pulse width requirements.

*

*

*

Figure 7. FCC pulse width requirements.

PICTURE EFFECTSSmall errors in pulse widths will not affect picture quality.However, if the errors become solarge that the pulses cannot beproperly processed (by equip-ment), picture breakup may occur.

TEST SIGNALSTiming measurements may bemade on any composite signalcontaining horizontal, verticaland subcarrier (burst) synchro-nization information.

MEASUREMENT METHODSWaveform Monitor Graticule. Timeintervals can be measured bycomparing the waveform to themarks along the horizontal baseline of a waveform monitorgraticule. In order to get adequateresolution, it is usually neces-sary to magnify the waveformdisplay horizontally. Select thesetting that provides as muchmagnification as possible whilestill keeping the interval of inter-est entirely on-screen. The scalefactor, typically microsecondsper major division, changes withhorizontal magnification. The1780R displays the microsec-onds per division setting on thescreen while for the 1480 it isobtained from the switch setting.

To make measurements betweenthe 90% (-4 IRE) points, it isgenerally most convenient to use

the waveform monitor variablegain control to normalize thesync height to 100 IRE. Theblanking level may then be posi-tioned at +10 IRE and a readingobtained from the marks on thebaseline (see Figure 8).

For measurements specified atthe 50% amplitude points, nor-malizing to 100 IRE is probablynot necessary. In this case, placethe top of the pulse at +20 IREand the bottom at -20 IRE. Thepulse width can then be read fro mthe horizontal scale (see Figure 9).

Time Cursors (1780R). Some wave-form monitors are equipped withcursors to facilitate the measure-ment of time intervals. The timecursors in the 1780R appear asbright dots on the waveform. Tofind the 90% points of the synctransitions, it is best to norm a l i z ethe sync pulse to 100 IRE anduse the vertical graticule scale to locate the proper level.Alternatively, the voltage cursorsin the RELATIVE mode can beused to locate the 90% points.

Similar procedures can be usedto find the 50% points of thetransitions (see Figure 10), but inthis case it may not be necessary.Since it is easier to visually locatethe halfway point of a transition,50% point measurements canoften be made without using anamplitude reference.

13

Figure 8. Horizontal sync width measurement at the 90% amplitude (4 IRE) points.

Figure 9. Horizontal sync width measurement at the 50% amplitude points.

Figure 10. The 1780R time cursors positioned to measure horizontal syncwidth at the 50% amplitude points.

Cursors (VM700T). The cursors inthe VM700T WAVEFORM modecan be used to make pulse widthmeasurements. After establishingthe 100% and 0% points ofsync, the cursors can be movedto either the 50% or 90% pointsto obtain a time measurement(see Figure 11). Consult the manual for detailed instructionson how to use the cursors.

VM700T Automatic Measurement.The H TIMING selection in theVM700T MEASURE mode displays all horizontal blankinginterval timing measurements(see Figure 12). Note that eitherFCC or RS-170A measurementscan be selected. Timing measure-ments are also available in theAUTO mode.

NOTES7. Rise and Fall Time Measure m e n t s .Both the FCC requirements andRS-170A include specificationsfor the rise time and fall time ofthe sync pulse. These measure-ments are indicators of how fastthe transitions occur. They aretypically made between the 10%and 90% points of the transition.The methods used for measuringpulse widths can generally beapplied to rise and fall times.

8. Burst Position and Number ofCycles. The position of burstwith respect to sync and thenumber of subcarrier cycles inburst are specified and it may bedesirable to occasionally verifythese parameters. RS-170A callsfor 9 cycles in burst while theFCC allows 8 to 11 cycles.

RS-170A specifies burst positionwith respect to sync in terms ofsubcarrier cycles. There arenominally 19 subcarrier cyclesbetween the 50% point of theleading edge of sync and thestart of burst (defined as "thezero crossing that precedes thefirst half-cycle of subcarrier thatis 50% or greater of the burstamplitude"). The 1780R FSCTIME MARKS mode may beused to check this parameter.Refer to the SCH Phase sectionof this book for the measurementmethodology.

9. Checking the Vertical Interval.The number of pulses in the ver-tical interval, as well as thewidths of the equalizing pulsesand vertical serrations, are alsospecified. Recommended limitscan be found in Appendices Cand D and most of the pulsewidth measurement methodsdiscussed in this section can be applied.

14

Figure 11. Horizontal sync width measurement at the 50% amplitude pointsusing the VM700T cursors.

Figure 12. The H Timing Measurement display in the VM700T MEASURE mode.

DEFINITIONSCH (SubCarrier to Horizontal)Phase refers to the timing rela-tionship between the 50% pointof the leading edge of sync andthe zero crossings of the re f e re n c esubcarrier. Errors are expressedin degrees of subcarrier phase.

RS-170A specifies that SCHphase must be within ±40degrees. Practically speaking,much tighter tolerances are gen-erally maintained. Modern facili-ties often try to ensure that SCHphase errors do not exceed a few degrees.

PICTURE EFFECTSSCH phase becomes importantonly when television signalsfrom two or more sources arecombined or sequentiallyswitched. In order to preventcolor shifts or horizontal jumpsfrom occurring when a switch ismade, the sync edges of the twosignals must be accurately timedand the phase of color burstmatched. Since both sync andsubcarrier are continuous signalswith a fixed relationship to oneanother, it is possible to simulta-neously achieve both timingconditions only if the two signals have the same SCHphase relationship.

Because of the complex relation-ship between the sync and sub-carrier frequencies, the exactSCH phase relationship for agiven line repeats itself onlyonce every four fields. In orderto understand why this four-field sequence exists, first con -sider the fact that there are anodd number of subcarrier half-cycles in a line. This impliesthat SCH phase must be 180degrees apart on adjacent lines.Since there are also an odd number of lines in a frame, theexact phase relationshipbetween sync and burst for agiven line repeats itself onceevery four fields (two frames).

In order to achieve the sync andburst timing conditions requiredfor a clean switch between twosignals, the four-field sequenceof the signals must be properlylined up (i.e., Field 1 of Signal Aand Field 1 of Signal B mustoccur at the same time). Whenthis condition is achieved, thetwo signals are said to be "colorframed." It is important toremember that color framing isinextricably tied to other systemtiming parameters and is by nomeans an independent variable.Only if two signals have thesame SCH phase relationshipand are properly color framedcan the sync timing and burstphase matching requirements be achieved.

15

SCH Phase

Figure 13. The SCH phase error of this signal is 0 degrees. Note that the 50% point of the leading edge of sync and a zerocrossing of the extrapolated subcarrier are time coincident.

Since signals must have the sameSCH phase relationship in orderto be cleanly combined, stan-dardization on one value of SCH phase will clearly facilitate transfer of program material. Thisis the reason for trying to main-tain 0 degrees of SCH phase erro r.Another reason for keeping SCHphase within reasonable limits isthat various pieces of equipmentneed to be able to distinguishbetween the color frames inorder to process the signal prop-erly. This cannot be done accu-rately if the SCH phase error isallowed to approach 90 degrees.

TEST SIGNALSSCH phase measurements can bemade on any signal with bothsync and color burst present.

MEASUREMENT METHODSPolar Display. Some instruments,such as the 1780R, are equippedwith a polar SCH display thatconsists of the burst vector and adot representing horizontal sync.The phase relationship betweenthe two can be determined byreading directly from the vectorgraticule or by using the preci-sion phase shifter (see Figure 14).

The instrument must be intern a l l yreferenced to measure the SCHphase of a single signal. Sync andburst of the selected signal arecompared to each other in thismode. In the 1780R, two sync dots(180 degrees apart) are displayedalong with the burst vector wheninternal reference is selected.

When external reference is sele-cted, both burst and sync of theselected signal are displayed relative to burst of the external

reference signal. A single syncdot appears with the burst vectorin this mode. This display isused to determine whether ornot two signals are color framed.A s s u m i n g that both the re f e re n c esignal and the displayed signalhave no SCH phase error, thesync dot will be in phase withthe burst vector if the signals arecolor framed and 180 degreesaway if they are not.

FSC Time Marks (1780R). The1780R has a mode called FSCTIME MARKS which is accessiblethrough the MEASURE menu. Inthis mode, bright dots appear onthe waveform at intervals of pre-cisely one subcarrier cycle. Thedots can be advanced or delayedwith respect to the waveformwith the precision phase shifter.

To make a measurement, use thephase shifter to set one of thedots on the 50% point of theleading edge of sync. If there isno SCH phase error, the dots onthe burst should fall on the zerocrossings at the blanking level(see Figure 15). If there is anerror and a numeric measure-ment result is desired, press REFSET to zero the phase readoutand use the phase shifter to posi-tion the burst dots on the zerocrossings. At this point thephase readout indicates theamount of SCH phase error.

This method is not as precise asthe polar display but has theadvantage of allowing verifica-tion that there are 19 subcarriercycles between sync and burst.Since this mode is independentof instrument calibration, it isalso useful for checking calibra-tion of the polar display.

16

Figure 14. The 1780R polar SCH phase display showing an 8 degree error.Internal reference is used for synchronization.

Figure 15. The 1780R FSC TIME MARKS display indicates that this signalhas no SCH phase error.

VM700T Automatic Measurement.Select SCH PHASE in theVM700T MEASURE menu toobtain a display of SCH phase.Figure 16 shows the VM700Tpolar SCH phase display whichis similar to the SCH phase dis-play of the 1780R. The VM700Tfull field SCH phase display (seeFigure 17) plots the SCH phaseof each line in the field and provides an average result. SCHphase measurements are alsoavailable in the AUTO mode.

NOTES10. For More Information. For acomprehensive discussion ofSCH phase and color framingissues, see Tektronix ApplicationNote 20W-5613-2, “Measuringand Monitoring SCH Phase with the 1750A Waveform/Vector Monitor”.

17

Figure 16. The VM700T polar SCH Phase Measurement display.

Figure 17. The VM700T full field SCH Phase Measurement display.

Waveform distortions that areindependent of signal amplitudeare referred to as linear distor-tions. These distortions occur asa result of a system's inability touniformly transfer amplitudeand phase characteristics at all frequencies.

When fast signal componentssuch as transitions and high-frequency chrominance areaffected differently than slowerline or field-rate information,linear distortions are probablypresent. These distortions aremost commonly caused byimperfect transfer characteristicsin the signal path. However,linear distortions can also beexternally introduced. Signalssuch as power line hum can couple into the video signal andmanifest themselves as distort i o n s .

One method of classifying lineardistortions involves groupingthem according to the durationof the signal components thatare affected by the distortion.Four categories, each corre-sponding to a specific televisiontime interval, have been defined.

These categories are:

SHORT TIME (125 nanosec-onds to 1 microsecond)

LINE TIME (1 microsecond to 64 microseconds)

FIELD TIME (64 microsec-onds to 16 milliseconds)

LONG TIME (greater than 16 milliseconds)

This means of classification isconvenient because it permitseasy correlation of the distor-tions with what is seen in thepicture or in a waveform dis-play. A single measurement foreach category takes into accountboth amplitude and phase dis -tortions within that time range.

While the combination of thesefour categories covers the entirevideo spectrum, it is also conve-nient to have methods of simul-taneously evaluating response at all frequencies of interest.Frequency response measure-ments look at the system'samplitude versus frequencycharacteristics while group delaymeasurements examine phase

versus frequency. Unlike mea-surements classified by timeinterval, frequency response andgroup delay measurements per-mit separation of amplitude dis-tortions from delay distortions.

The first two measurements dis-cussed in this section specificallyaddress the relationshipsbetween the chrominance andluminance information in a sig-nal. Chrominance-to-luminancegain and delay measurementsquantify a system's ability toprocess chrominance and lumi-nance in correct proportion andwithout relative time delays.

Sine-squared pulses and risetimes are used extensively in themeasurement of linear waveformdistortions. It may be helpful toreview the information inAppendix B which discusses theuse of sine-squared pulses intelevision testing. For moredetailed information, a compre-hensive technical discussion oflinear waveform distortions ispresented in IEEE Standard 511-1979, “Video SignalTransmission Measurement ofLinear Waveform Distortion”.

18

II. LINEAR DISTORTIONS

DEFINITIONChrominance-to-luminance gaininequality (relative chrominancelevel) is a change in the gainratio of the chrominance andluminance components of a video signal. The amount of distortion can be expressed in IRE, percent, or dB. The number is negative for lowchrominance and positive forhigh chrominance.

C h rominance-to-luminance delayinequality (relative chrominancetime) is a change in the timerelationship of the chrominanceand luminance components of avideo signal. The amount of dis-tortion is expressed in units oftime, typically nanoseconds. Thenumber is positive for delayedchrominance and negative foradvanced chrominance.

PICTURE EFFECTSGain errors most commonlyappear as attenuation or peakingof the chrominance informationwhich shows up in the pictureas incorrect color saturation.

Delay distortion will cause colorsmearing or bleeding, part i c u l a r l yat the edges of objects in the picture. It may also cause poorreproduction of sharp luminancetransitions.

TEST SIGNALSChrominance-to-luminance gainand delay measurements can bemade with any test signal con-taining a 12.5T sine-squaredpulse with 3.58 MHz modula-tion. Many combination signals,such as the composite signalsshown in Figures 18 and 19,contain this pulse.

MEASUREMENT METHODSConventional chrominance-to-luminance gain and delay mea-surements are based on analysisof the baseline of a modulated12.5T pulse (see Appendix B forfurther information). This pulseis made up of a sine-squaredluminance pulse and a chromi-nance packet with a sine-square denvelope (see Figure 20).

Modulated sine-squared pulsesoffer several advantages. First ofall, they allow evaluation of bothgain and delay differences witha single signal. A further advan-tage is that modulated sine-squared pulses eliminate theneed to separately establish alow-frequency amplitude refer-ence with a white bar. Since alow-frequency reference pulse ispresent along with the high-frequency information, theamplitude of the pulse itself can be normalized.

19

Chrominance-to-Luminance Gain and Delay

Figure 18. A composite signal (also known as FCC Composite) containinga modulated sine-squared sine pulse.

Figure 19. The composite signal specified in the EIA 250-C Standard (alsoknown as NTC-7 Composite).

3.12 µs

0.5

00.5

0

– 0.5

CHROMINANCECOMPONENT

LUMINANCECOMPONENT

3.12 µs

A

HAD1.56 µs

MOD 12.5T PULSE

1

0

Figure 20. Chrominance and luminance components of the modulated12.5T pulse.

20

The baseline of the modulated12.5T pulse is flat when chromi-nance-to-luminance gain anddelay distortion is absent.Various types of gain and delaydistortion affect the baseline indifferent ways. A single peak in the baseline indicates the presence of gain errors only.Symmetrical positive and nega-tive peaks indicate the presenceof delay errors only. When bothtypes of errors are present, thepositive and negative peaks willhave different amplitudes andthe zero crossing of the baselinedistortion will not be at the center of the pulse. Figure 21shows the effects of varioustypes of distortion.

The 12.5T modulated sine-squared pulse has a half ampli-tude duration (HAD) of 1.56microseconds, or 12.5 times theNTSC System Nyquist interval(see Appendix B). The frequencyspectrum of this pulse includesenergy at low frequencies andenergy centered around the sub-carrier frequency. The HAD of12.5T was chosen in order tooccupy the chrominance band-width of NTSC as fully as possi-ble and to produce a pulse with sufficient sensitivity todelay distortion.

Waveform Monitor and Nomograph.Chrominance-to-luminanceinequalities are quantified bymeasuring the baseline distor-tion peaks of the 12.5T pulse.The amount of distortion iseither calculated from thesenumbers or obtained from a nomograph.

With a traditional waveformmonitor, a nomograph is mostcommonly used. To make a mea-surement, first normalize the12.5T pulse height to 100 IRE.The baseline distortion can bemeasured either by comparingthe waveform to a graticule or byusing voltage cursors. Using anomograph (see Figure 22), findthe locations on the horizontaland vertical axes that correspondto the two measured distortionpeaks. At the point in the nomo-graph where perpendicular linesdrawn from these two locationswould intersect, the gain anddelay numbers may be read fromthe nomograph.

DELAYED CHROMINANCE ADVANCED CHROMINANCE

YM YM

Y2Y1Y1 Y2

DELAY ERRORS ONLY

YM

Y

YM

Y

GAIN ERRORS ONLY

HIGH CHROMINANCE LOW CHROMINANCE

CHROMINANCEHIGH AND DELAYED

CHROMINANCELOW AND DELAYED

YM

Y1 Y2

Y2 Y1> Y1 Y2>

YM

Y1 Y2

CHROMINANCEHIGH AND ADVANCED

Y1 Y2>

YM

Y1Y2

CHROMINANCELOW AND ADVANCED

YM

Y1Y2

Y2 Y1>

GAIN AND DELAY INEQUALITIES BOTH PRESENT

Figure 21. Effects of gain and delay inequalities on the modulated 12.5 T pulse.

21

Figure 22. Chrominance-to-luminance gain and delay nomograph.

1780R Semi-Automatic Procedure.The CHROMA/LUMA selectionin the 1780R MEASURE menueliminates the need for a nomo-graph. The on screen readoutguides the user through cursormeasurements of the variousparameters required to obtain anumber from a nomograph. After all parameters have beenentered, the instrument calcu-lates the results (see Figure 23).The accuracy and resolution ofthis method are roughly equiva-lent to using the graticule and a nomograph.

Waveform Monitor GraticuleApproximations. When a system isfree of significant nonlinearityand delay distortion is withincertain limits, chrominance-to-luminance gain inequality canbe measured directly by compar-ing the height of the 12.5T pulseto the white bar. This methodand the nomograph will yieldidentical results when there isno delay distortion. It is generallyconsidered a valid approxima-tion for signals with less than150 nanoseconds of delay and isaccurate to within 2% for signalswith up to 300 nanoseconds of delay.

The white bar amplitude mustbe normalized to 100 IRE for thismeasurement. Measure theamplitude difference betweenthe 12.5T pulse top and thewhite bar in IRE. This number,times two, is the amount ofchrominance-to-luminance gaindistortion in percent. Note thatwhen the pulse top is higher orlower than the bar, the bottom ofthe pulse is displaced from thebaseline by the same amount.Thus the peak-to-peak differencebetween the 12.5T pulse and thebar is actually twice the differ-ence between their peak values,hence the factor of two.

The graticules in waveformmonitors such as the 1780R andthe 1480 can be used to estimatechrominance-to-luminance delayerrors. This method yields validresults only if gain errors arenegligible and the baseline distortion is symmetrical.Normalize the 12.5T pulseheight to 100 IRE and then cen-ter the pulse on the two graticulelines which cross in the centerof the baseline. When the wave-form monitor is in the 1 volt fullscale mode, these lines indicate200 nanoseconds of delay (seeFigure 24). With X5 vertical gain(0.2 volts full scale) selected, thelines indicate 40 nanoseconds of delay.

22

Figure 23. Results obtained with the CHROMA/LUMA selection in the 1780RMEASURE mode.

Figure 24. The 1780R graticule indicates that this signal has approximately200 nanoseconds of chrominance-to-luminance delay.

VM700T Automatic Measurement.Chrominance-to-luminance gainand delay distortion can be mea-sured by selecting CHROM/LUMGAIN DELAY in the VM700TMEASURE mode. The graphplots the error with respect tozero with numeric results givenat the top of the display (seeFigure 25). The X axis is thedelay (positive or negative) andthe Y axis is the gain inequality.Chrominance-to-luminance measurements are also availablein the AUTO mode.

Calibrated Delay Fixture. Anothermethod of measuring these dis-tortions involves use of a cali-brated delay fixture. The fixtureallows incremental adjustmentof the chro m i n a n c e - t o - l u m i n a n c edelay until there is only onepeak in the baseline indicatingthat delay error has been nulledout. The delay value can then be

read from the fixture and gainmeasured directly from thegraticule. This method can behighly accurate but requires theuse of specialized equipment.

NOTES11. Harmonic Distortion. If harm o n i cdistortion is present, there maybe multiple aberrations in thebaseline rather than one or twoclearly distinguishable peaks. Inthis case, nomograph measure-ment techniques are indetermi-nate. The VM700T, however, iscapable of removing the effectsof harmonic distortion and willyield valid results in this case.Minor discrepancies between theresults of the two methods maybe attributable to the presence ofsmall amounts of harmonic dis-tortion as well as to the higherinherent resolution of theVM700T method.

23

Figure 25. The Chrominance/Luminance Gain Delay display in the VM700TMEASURE mode.

24

DEFINITIONShort time distortions cause am-plitude changes, ringing, over-s h o o t , and undershoot in fast risetimes and 2T pulses. The aff e c t e dsignal components range in dur-ation from 0.125 microsecond to1.0 microsecond. Errors are expre -ssed in "percent SD", which isdefined in the MEASUREMENTMETHODS section below.

The presence of distortions inthe short time domain can alsobe determined by measuringK2T or Kpulse/bar as describedin the K Factor Ratings sectionof this publication.

PICTURE EFFECTSShort time distortions producefuzzy vertical edges. Ringing cansometimes be interpreted aschrominance information (crosscolor) causing color artifactsnear vertical edges.

TEST SIGNALSShort time distortion may bemeasured with any test signalthat includes a T rise time whitebar. A T rise time bar has a 10%to 90% rise time of nominally125 nanoseconds (see Figure 26).The EIA-250-C composite signalincludes a T rise time bar andsome generators allow selectionof T rise times for other signals.See Appendix B for a discussionof the time interval T.

It is very important a T rise timebar be used for this measure m e n t .Many common test signals have2T rather than 1T rise times andare not suitable for this measure-ment. It should also be noted

that T rise time signals will suff e rsignificant distortion whenpassed through a TV transmitteras they contain spectral compo-nents to 8 MHz which will beremoved by the transmitter 4.2 MHz lowpass filter. Shorttime distortion measurementsmade on transmitted signals will therefore evaluate onlythose signal components in the 240 nanosecond to 1 microsecond range.

MEASUREMENT METHODSShort time distortions are mostnoticeable in sharp transitions andappear as ringing, overshoot, orundershoot at the transition cor-ners. The distortion is quantifiedby measuring these aberrations.

The distortion amplitudes arenot generally quoted directly asa percent of the transition amplitude, but rather in terms ofan amplitude weighting system that yields "percent SD". Thisweighting is necessary becausethe amount of distortiondepends not only on the distor-tion amplitude but also on thetime the distortion occurs withrespect to the transition. Theequation for NTSC systems is:

SD = at0.67

where "a" is the lobe amplitudeand "t" is the time between tran-sition and distortion. In practice,special graticules or conversiontables are used to eliminate the necessity for calculations.An example of a short time dis-tortion graticule is shown inFigure 27.

Short Time Distortion

125 nSEC

10%

90%

100 IRE

F i g u re 26. A T rise time bar has a 10% to 90% rise time of 125 nanoseconds.

B

C

W

OUTER MARKINGS : ——— : SD = 5%

INNER MARKINGS : ——— : SD = 2%

SD =

T STEP OF LINEBAR FITTED

THROUGH B, C & W

MAX GRATICULEREADING

Figure 27. A short time distortion graticule.

Waveform Monitor Graticule.Special external graticules forshort time distortion measure-ments are provided with the1780R and 1480. To make a mea-surement, first set the horizontalmagnification to 200 nanoseconds(0.2 microseconds) per division.Points B (Black), C (Center) andW (White) are provided on thegraticule to assist in positioningthe waveform. Use the horizon-tal and vertical position controlsto make sure the waveform pass-es through points B and C andthe variable gain control to makethe top of the pulse pass throughW. Some adjustment iterationmay be necessary.

Once the waveform is properlypositioned, the amount of distor-tion can be determined by comparison to the graticule.Note where the waveform failswith respect to all parts of thegraticule as the largest aberrationis not necessarily the one whichwill determine the amount of distortion.

Since the graticule only showslimits for 2% and 5% SD, inter-polation may be required. Formeasuring smaller distortions,select X5 vertical gain (0.2 voltsfull scale). At this gain settingthe graticule lines indicate limitsof 0.4% and 1%.

VM700T Automatic Measurement.Short time distortion can bemeasured by selecting SHORTTIME DISTORTION in the

VM700T MEASURE mode. TheVM700T automatically measuresboth short time distortion andbar rise time (see Figure 29).Short time distortion measure-ments are also available in theAUTO mode.

NOTES12. Nonlinearities. If the device orsystem under measurement isfree of nonlinear distortion, therising and falling transitions willexhibit symmetrical distortion.In the presence of nonlinearities,however, the transitions may beaffected differently. It is prudentto measure, or at least inspect,both the positive and negativetransitions.

13. Pulse-to-Bar Ratios. Theamplitude ratio between a 2Tpulse and a line bar is some-times used as an indication ofshort time distortion. To make apulse-to-bar measurement with aw a v e f o rm monitor, first norm a l i z ethe bar amplitude to 100%. Thiscan be done either by using theIRE graticule scale or voltagecursors in the RELATIVE mode.Now measure the pulse ampli -tude to obtain pulse-to-bar ratioreading in percent.

A pulse-to-bar measurement canbe obtained from the VM700T byselecting K FACTOR in theMEASURE mode. The pulse-to-bar ratio is given in the upperright-hand corner (see Figure 30).This measurement is also avail-able in the AUTO mode.

25

Figure 28. This waveform exhibits short time distortion of 5% SD.

Figure 29. VM700T Short Time Distortion display.

Figure 30. Pulse-to-bar ratio is given in the VM700T MEASURE modeK FACTOR selection.

26

DEFINITIONLine time distortion causes tiltin line-rate signal componentssuch as white bars. The affectedcomponents range in durationfrom 1.0 microsecond to 64microseconds. The amount ofdistortion may be expressed inIRE or in percent of the line bar amplitude.

Distortions in the line timedomain can also be quantifiedby measuring Kbar as discussedin the K Factor Ratings sectionof this booklet.

PICTURE EFFECTSLine time distortion producesbrightness variations betweenthe left and right sides of thescreen. Horizontal streaking andsmearing may also be apparent.This distortion is most apparentin large picture detail.

TEST SIGNALSLine time distortion may bemeasured with any test signalthat contains an 18 microsecond,100 IRE bar. The composite signals shown in Figures 31 and 33 both include such a bar.The window signal shown in theField Time Distortion section ofthis booklet is also suitable. Risetime of the bar is not critical forthis measurement.

MEASUREMENT METHODSLine time distortion is quantifiedby measuring the amount of tiltin the top of the line bar. The

peak-to-peak deviation of the tilt is generally quoted as theamount of distortion (see Note14). The first and last microsec-ond of the bar should be ignoredas errors near the transition are in the short time domain.Figure 32 illustrates line timedistortion parameters.

Waveform Monitor Graticule. The1780R and 1480 graticules areequipped with marks for mea-suring line time distortion. Notethe box in the graticule upperleft-hand corner labeled LD =2%, 5%. To make a measure-ment, select a one line sweepand use the two arrows on the50 IRE line to position the barhorizontally (see Figure 33).Make sure that the blankinglevel of the waveform is on thebaseline, and that the bar toppasses through 100 IRE at itsmidpoint. It may be necessary touse the variable gain control onthe waveform monitor to nor-malize the gain. Use the ±2%and ±5% graticule marks toquantify the peak-to-peak deviation of any bar top tilt thatoccurs within the box. The boxexcludes the first and lastmicrosecond of the bar.

The waveform monitor verticalgain can be increased for mea-surement of smaller errors. Thegraticule marks correspond to0.4% and 1% limits when theX5 setting (0.2 volts full scale) is selected.

Line Time Distortion

a

1 µs 1 µs

BAR CENTER

LD = a

a IN % OF RA

RA

LINE BAR

Figure 31. A composite signal (also known as FCC Composite).

Figure 32. Parameters for measurement of line time distortion.

Figure 33. Using the waveform monitor graticule to measure line timedistortion. A 3% distortion is shown here.

Although the special line timedistortion graticule is conve-nient, this measurement can bemade with any waveform monitor.To make a measurement, firstuse the variable gain control tonormalize the center of the barto 100 IRE. Ignoring the first andlast microsecond, measure thepeak-to-peak tilt of the top of thebar. Since the gain has been normalized, the tilt measure-ment in IRE is equal to the linetime distortion in percent.

1780R Voltage Cursors. Waveformmonitor voltage cursors in theRELATIVE mode can be used tomeasure line time distortion.Define the amplitude differencebetween blanking level and thebar center as 100%. Positionboth cursors to measure thepeak-to-peak tilt. This number isthe line time distortion.Remember to ignore the first andlast microsecond of the bar.

The 1780R time cursors are con-venient for locating the appro-priate time interval in the centerof the bar. Set the time separa-tion to 16 microseconds and putthe time cursors in the TRACKmode. Move the two cursorstogether until they are centeredon the bar (see Figure 34).

VM700T Automatic Measurement.The VM700T provides a linetime distortion measurement inthe AUTO mode.

NOTES14. Peak-to-Peak Versus MaximumDeviation. In this booklet, bothline time and field time distor-tions are discussed in terms ofpeak-to-peak measurements.This definition is in keepingwith IEEE Standard 511-1979.Some measurement standards,however, define the distortion asthe maximum deviation from thecenter of the bar. If using thatmeasurement definition, themeasurement techniques can beadapted accordingly.

27

Figure 34. The 1780R voltage and time cursors can facilitate line timedistortion measurements.

DEFINITIONField time distortion causesfield-rate tilt in video signals.The affected signal componentsrange in duration from 64microseconds to 16 millisec-onds. The error is expressed inIRE or as a percentage of a refer-ence amplitude which is gener-ally the amplitude at the centerof the line bar.

K60 Hz measurements, which are discussed in the K FactorRatings section of this booklet,provide another method ofdescribing field time distortions.

PICTURE EFFECTSField time distortion will causetop-to-bottom brightness inaccu-racies in large objects in the picture.

TEST SIGNALSField time distortion can bemeasured either with a windowor field square wave test signal.As the two signals may yield dif-ferent results, it is good practiceto note which was used.

The window signal has approxi-mately 130 lines in the center ofthe field that include an 18microsecond line bar. On a picture monitor, this creates the "window" effect shown inFigure 35. This signal is alsosuitable for measuring line time distortions.

A field square wave is similar, butthe lines in the center of the fieldare at 100 IRE for the entire line.

MEASUREMENT METHODSField time distortions are quanti -fied by measuring the amount oftilt in the top of the bar. Thepeak-to-peak deviation of the tiltis generally quoted as theamount of distortion (see Note14 on page 27). Field time mea-surement parameters are shownin Figure 36. The referenceamplitude is usually the centerof the line bar and the first andlast 0.2 milliseconds (about 3lines) of the field bar are i g n o re d .D i s t o rtions in that re g i o n are notin the field time domain.

28

Field Time Distortion

100

0

–40

18 µs 8.3 ms

LINE BAR(T STEP RISE / FALL)

FIELD BAR(COMPOSED OF LINE BARS)

WAVEFORM MONITORFIELD DISPLAY

WAVEFORM MONITORLINE DISPLAY

PICTURE MONITORDISPLAY

IRE

a

0.2 ms 0.2 ms

TOP OF LINE BAR

REFERENCEAMPLITUDE

(RA)

FIELD BAR

FD = a

a IN % OF RA

Figure 35. The window signal.

Figure 36. Parameters for measurement of field time distortion.

29

Waveform Monitor Graticule. Thefirst step in making a field timedistortion measurement is tonormalize the gain. With thewaveform monitor in a line ratesweep mode, use the variablegain control to set the center ofthe line bar to 100 IRE. This canbe done most accurately withthe waveform monitor FAST DCrestorer selected. The DC restor-er will remove the effects of fieldtime distortion from the wave-form monitor display, and there-fore reduce the vertical blurringseen in the line rate display.

Select a field rate sweep andeither the SLOW or OFF settingfor the DC restorer. Measure thepeak-to-peak tilt of the field bar,excluding the first and last 0.2milliseconds. This IRE reading,expressed as a percentage, is theamount of field time distortion(see Figure 37).

1780R Voltage Cursors. The 1780Rvoltage cursors can be used inthe RELATIVE mode to measurefield time distortion. Select a

one or two line sweep anddefine the center of the line bar(relative to blanking) as 100%.Remember to select the FASTDC restorer setting for this partof the measurement procedure.

Select a field rate sweep and setthe DC restorer to SLOW or OFF.Ignoring the first and last threelines in the bar, place the cursorsat the positive and negativeexcursions of the tilt (see Figure38). The voltage cursor readoutnow indicates the amount offield time distortion.

VM700T Automatic Measurement.The VM700T provides a fieldtime distortion measurement inthe AUTO mode.

NOTES15. Hum. Externally introduceddistortions such as mains humare also considered field ratedistortions. Be sure to turn theDC restorer OFF or select theSLOW clamp speed when measuring hum.

Figure 37. A 2-field waveform monitor display showing about 5% fieldtime distortion.

Figure 38. The 1780R voltage cursors can be used to measure fieldtime distortion.

DEFINITIONLong time distortion is the lowfrequency transient resultingfrom a change in APL. This distortion usually appears as avery low frequency dampedoscillation. The affected signalcomponents range in durationfrom 16 milliseconds to tens of seconds.

The peak overshoot, in IRE, isgenerally quoted as the amountof distortion. Settling time isalso sometimes measured.

PICTURE EFFECTSLong time distortions are slowenough that they are often per-ceived as flicker in the picture.

TEST SIGNALSLong time distortion is measuredwith a flat field test signal withvariable APL. The signal shouldbe "bounced", or switchedbetween high and low APL (usu-ally 90% and 10%), at intervalsno shorter than five times thesettling time (see Figure 39).

MEASUREMENT METHODSLong time distortions are mea-sured by examining the dampedlow-frequency oscillation resulting from a change in APL.

Waveform Monitor. It is usuallynecessary to use a storage oscil-loscope or a waveform monitorin the SLOW SWEEP mode tomeasure long time distortion. A waveform photograph can behelpful in quantifying the distor-tion. When a stable display isobtained (or a photograph taken),measure the peak overshoot andsettling time (see Figure 40).

VM700T Automatic Measurement.Select BOUNCE in the VM700TMEASURE mode to obtain a dis-play of long time distortion (seeFigure 41). Peak deviation andsettling time are given at the bottom of the screen.

30

Long Time Distortion

IRE100

80

60

40

20

0

–4010% APL 90% APL 10% APL

X SECONDS X SECONDS

X SECONDS

PEAKOVERSHOOT

SETTLINGTIME

10% APLTO 90% APLTRANSITION

10% APL

BLANK

SYNC

Figure 39. A flat field bounce signal.

Figure 40. Long time distortion measurement parameters.

Figure 41. The VM700T Bounce display.

DEFINITIONFrequency response measure-ments evaluate a system's abilityto uniformly transfer signal com-ponents of different frequencieswithout affecting their ampli-tude. This parameter, alsoknown as gain/frequency distor-tion or amplitude versus fre-quency response, evaluates thesystem's amplitude responseover the entire video spectrum.

The amplitude variation may beexpressed in dB, percent, or IRE.The reference amplitude (0 dB,100%) is typically the white baror some low fre q u e n c y. Fre q u e n c yresponse numbers are onlymeaningful if they contain threepieces of information: the mea-sured amplitude, the frequencyat which the measurement wasmade, and the re f e rence fre q u e n c y.

PICTURE EFFECTSF requency response problems cancause a wide variety of aberra-tions in the picture, including allof the effects discussed in the sec-tions on short time, line time, fieldtime, and long time distortions.

TEST SIGNALSFrequency response can be mea-sured with a number of differenttest signals. Since there are sig-nificant diff e rences between thesesignals, each one is discussed insome detail in this section.

Some test signals are availableeither as full amplitude orreduced amplitude signals. It isgenerally good practice to makemeasurements with both as thepresence of amplitude nonlin-earities in the system will havegreater effect on measurementsmade with full amplitude signals.

Multiburst. The multiburst signaltypically includes six packets ofdiscrete frequencies that fallwithin the TV passband. Thepacket frequencies usually rangefrom 0.5 to 4.1 or 4.2 MHz withfrequency increasing toward theright side of each line (seeFigure 42). This signal is usefulfor a quick approximation of system frequency response andcan be used on an in-servicebasis as a VIT (Vertical IntervalTest) signal.

Multipulse. The multipulse signalis made up of modulated 25Tand 12.5T sine-squared pulseswith high-frequency componentsat various frequencies of interest,generally from 1.25 to 4.1 MHz(see Figure 43). This signal canalso be used as a VIT signal.

31

Frequency Response

0.5 1.25 2.0 3.0 3.58 4.1

4~ 8~ 10~ 14~ 16~ 18~ (CYCLES)

MHz

2T 25T 12.5T 12.5T12.5T12.5T

1.0 2.0 3.0 3.58 4.2 MHz

Figure 42. A reduced amplitude multiburst signal (also known asFCC Multiburst).

Figure 43. A 70 IRE multipulse signal.

Modulated sine-squared pulses,which are also used to measurechrominance-to-luminance gainand delay errors, are discussedon pages 19 and 24. Althoughdifferent high-frequency compo-nents are used in the multipulse,the same measurement princi-ples apply. Bowing of the base-line indicates an amplitude errorbetween the low frequency andhigh frequency components ofthat pulse. Unlike a multiburst,the multipulse facilitates evalua-tion of group delay errors as wellas amplitude errors.

Sweep. It is sometimes recom-mended that line or field ratesweep signals be used for mea-suring frequency response. In asweep signal, the frequency ofthe sine wave is continuouslyincreased over the interval of aline or field. An example of afield sweep is shown in Figure44. The markers indicate 1 MHzfrequency intervals.

A sweep signal allows examina-tion of frequency response con-tinuously over the interval ofinterest rather than only at thediscrete frequencies of the multi -burst and multipulse signals.This can be useful for detailedcharacterization of a system butdoes not offer any significantadvantages in routine testing.While the other signals discussedhere can be used as VITS andt h e re f o re permit in-service testing,a field-rate sweep can only beused on an out-of-service basis.

(Sin x)/x. The (sin x)/x signal hasequal energy present at all har-monics of the horizontal scanfrequency up to its cutoff fre-quency (see Figures 45 and 50).The (sin x)/x is primarilydesigned for use with a spectru manalyzer or an automatic mea-s u rement set such as the VM700T.Very little information is dis-c e rnible in a time domain display.

MEASUREMENT METHODSSince each signal requires a dif-ferent measurement method,separate discussions for the vari -ous test signals are presented inthis section. The first three sig-nals (multiburst, multipulse, andsweep) can all be measured witha waveform monitor using eitherthe graticule or the voltage cur-sors to quantify any distortion.Measurement results are usuallyexpressed in dB but IRE and per-cent of references are also used.

Waveform Monitor - Multiburst.Frequency response measure-ments are made with the multi-burst signal by measuring thepeak-to-peak amplitude of thepackets. There is very littleagreement among measurementstandards about what to use asthe reference level for multiburstmeasurements. To ensure accu-rate and repeatable measure m e n t s ,it is important to select one defi-nition and use it consistently.

32

Figure 44. A 6 MHz field rate sweep signal with markers (2-field display).

Figure 45. A time domain display of the (sin x)/x signal.

With a full amplitude multi-burst, either the white bar or thefirst packet may be used as thereference. When a reduced-amplitude multiburst is used,some standards recommend normalizing the white bar to 100IRE. The difference, in IRE,between the peak-to-peak ampli-tude of each packet and thenominal level is then taken asthe distortion at that frequency.Alternatively, the amount of dis-tortion may be quoted in dB orpercent relative to the reference.

Figures 46 and 47 show the1780R voltage cursors beingused to determine that frequencyresponse is down 3.25 dB at 4.1MHz. The reference is the 125KHz square wave at the beginningof the horizontal line. The errorin dB is calculated as follows:

20 log10 (61.91/90) = -3.25 dB.

61.91 is the amplitude in IRE ofthe 4.1 MHz packet and 90 is theamplitude in IRE of the 125 KHzsquare wave reference.

Waveform Monitor - Multipulse.Frequency response distortionshows up in the multipulse sig-nal as bowing of the pulse base-line (see Figure 48). Distortionsare quantified by measuring theamount of baseline displacementin the pulse of interest. It isoften easy to see which pulseexhibits the largest gain inequal-ity so an overall result can beobtained by measuring thatpulse only.

This measurement is most com-monly made by using a wave-form monitor graticule to mea-sure the baseline distortion andthen transferring the numbers for each pulse to a nomograph.The nomograph used for chromi-nance-to-luminance gain anddelay measurements (see Figure22) also applies to the multi-pulse. When making this mea-surement, normalize each pulseheight to 100 IRE before measur-ing the baseline bowing.

If group delay distortion is alsopresent, the pulse baseline dis-tortion will be sinusoidal ratherthan a single peak. In this case,measure both distortion peaksand apply the numbers to the nomograph. It will yield correct frequency responseresults as well as a group delay measurement.

The CHROMA/LUMA selectionin the 1780R MEASURE menucan also be used to make fre-quency response measurementswith the multipulse. Repeat thecursor measurement procedurefor the pulse corresponding toeach frequency of interest.

It is also possible to estimate theamplitude error without using anomograph. Normalize the whitebar to 100 IRE, and then measurethe displacement of the pulsetop from the white bar. Thisnumber, times two, is theamount of frequency responsedistortion in percent. Thismethod yields valid results evenin the presence of some delaydistortion. When delay distor-tion of more than about 150nanoseconds is present, however,this method is not re c o m m e n d e d .

33

Figure 46. The square wave is measured as a reference.

F i g u re 47. The peak-to-peak amplitude of the smallest packet is then measure d .

Figure 48. The multipulse signal exhibiting high frequency roll-off.

Waveform Monitor - Sweep.Amplitude variations can bemeasured directly from a time-domain display when a sweepsignal is used. Be sure to select afield-rate display on the wave-form monitor when using a fieldsweep. Establish a reference atsome low frequency and mea-sure the peak-to-peak amplitudeat the frequencies of interest (see Figure 49).

Spectrum Analyzer - (Sin x)/x.Frequency response testing withthe (sin x)/x signal is done witha spectrum analyzer. Attenuationor peaking of the flat portion ofthe spectral display can be readdirectly off the analyzer displayin dB (see Figure 50).

In a time domain display, highfrequency roll off will reduce thepulse amplitude and the ampli -tude of the pulse lobes. It is dif-ficult, however, to quantify theamount of distortion. The pres-ence of amplitude nonlinearityin the system will cause asym-metrical distortion of the posi -tive and negative pulses.

VM700T Automatic Measurement.The VM700T provides ampli-tude versus frequency responsemeasurement for the multiburst(MULTIBURST in the MEASUREmode) and (sin x)/x (GROUPD E L AY (SIN X)/X in the M E A S U R E mode) signals. These measurements areshown in Figures 51 and 52.

Multiburst measurements are alsoavailable in the AUTO mode.

34

Figure 49. A field rate sweep signal showing frequency response distortion.

Figure 50. A spectrum analyzer display of a (sin x)/x signal with a cutofffrequency of 4.75 MHz.

Figure 51. The VM700T Multiburst measurement.

NOTES16. Chrominance FrequencyResponse. Special versions of themultiburst and multipulse sig-nals have been developed toassist in accurately measuringthe frequency response of thechrominance channel. Ratherthan examining the entire videopassband, these signals containfrequencies centered around thenominal subcarrier frequency.Measurement procedures arevery similar to those outlined forfrequency response. In theVM700T, select CHROMA FREQUENCY RESPONSE in theMEASURE mode to evaluate thisparameter (see Figure 53).

17. Multipulse and NonlinearDistortions. The device or systemunder test must be reasonablyfree of nonlinear distortions,such as differential phase andgain, when using the multipulsesignal. Large nonlinear distor-tions can cause erroneous read-ings of both frequency responseand group delay.

18. More Information. TwoTektronix application notes areavailable for more informationon frequency response testing.See “Using the MultipulseWaveform to Measure GroupDelay and Amplitude Errors”(20W-7076-1) and “FrequencyResponse Testing Using a (Sin x)/x Test Signal and theVM700A/T Video MeasurementSet” (25W-11149-0).

35

Figure 52. The VM700T Group Delay & Gain measurement also providesfrequency response information.

Figure 53. The VM700T Chroma Frequency Response measurement.

DEFINITIONGroup delay distortion is presentwhen some frequency compo-nents of a signal are delayedmore than others. Distortion isexpressed in units of time. The largest difference in delaybetween a re f e rence low fre q u e n c yand other frequencies is typicallyquoted as the amount of distort i o n .

PICTURE EFFECTSGroup delay problems can causea lack of vertical line sharpnessdue to luminance pulse ringing,overshoot, or undershoot.

TEST SIGNALSThe multipulse signal (seeFigure 54) is used to measuregroup delay distortion. It is alsopossible to measure group delaywith the (sin x)/x signal but onlywith an automatic measurementset such as the VM700T.

MEASUREMENT METHODSGroup delay is measured by ana-lyzing the baseline distortion ofthe modulated sine-squaredpulses in the multipulse signal(see Figure 55). The measure-ment method is very similar tothat used for chrominance-to-luminance delay differing onlyin the number of frequencies atwhich delay is measured.

Waveform Monitor and Nomograph.The baseline distortion of eachpulse must be individually mea-sured and applied to a nomo-graph (see Figure 22). Normalizeeach pulse height to 100 IRE andmeasure the positive and nega-tive peaks of the baseline distor-tion. Apply the numbers to thenomograph to obtain a delayvalue. The largest delay mea-sured is typically quoted as theamount of group delay distor-tion. In practice, it is often easyto see which pulse exhibits themost delay necessitating onlyone measurement when maxi-mum delay is the value of intere s t .

The same nomograph works forany modulated 12.5T pulse,regardless of the modulation fre-quency. However, the first pulsein a multipulse signal is generallya 25T pulse rather than a 12.5Tpulse. When this is the case,multiply the delay number fromthe nomograph by two to obtainthe actual delay value.

1780R Semi-Automatic Procedure.Group delay can be measuredwith the CHROMA/LUMA selec-tion in the 1780R MEASUREmenu. Repeat the measurementprocedure for each frequency of interest.

36

Group Delay

2T 25T 12.5T 12.5T12.5T12.5T

1.0 2.0 3.0 3.58 4.2 MHz

Figure 54. The multipulse signal.

Figure 55. The multipulse signal exhibiting group delay distortion. Groupdelay differences between the low and high-frequency components of thepulse appear as sinusoidal distortion of the baseline.

Figure 56. The VM700T Group Delay & Gain display.

VM700T Automatic Measurement -(Sin x)/x. The VM700T uses the(sin x)/x signal to make groupdelay measurements. Thismethod offers the advantage ofproviding delay information fora large number of frequencies,rather than just the six discretefrequencies included in multi -pulse. Select GROUP DELAY(SIN X)/X in the VM700T MEASURE mode (see Figure 56).

NOTES19. Group Delay Definition. Inmathematical terms, group delayis defined as the derivative ofphase with respect to frequency(dφ/dω). In a distortion free sys-tem, the phase versus frequencyresponse is a linear slope andthe derivative is therefore a constant (see Figure 57).

If the phase versus frequencyresponse is not linear, then thederivative is not a constant andgroup delay distortion is present.The largest difference in dφ/dωthat occurs over the frequencyinterval of interest is the amountof group delay (see Figure 58).

20. Envelope Delay. The term"envelope delay" is often usedinterchangeably with groupdelay in television applications.Strictly speaking, envelope delayis measured by passing anamplitude modulated signalthrough the system and observ-ing the modulation envelope.Group delay, on the other hand,is measured directly by observ-ing phase shift in the signalitself. Since the two methodsyield very nearly the sameresults in practice, it is safe to assume the two terms are synonymous.

37

FREQUENCY

ω1 ω2

AMPLITUDE

FREQUENCY

ω1 ω2

PHASE

ω1 ω2

TIME

GROUP DELAY = dø dω

FREQUENCY

ω1 ω2

AMPLITUDE

FREQUENCY

ω1 ω2

PHASE

FREQUENCY

ω1 ω2

TIME

FREQUENCY

GROUP DELAYERROR

Figure 57. Response of a distortion free system.

Figure 58. Response of a system with amplitude and phase distortion.

DEFINITIONThe K Factor rating system mapslinear distortions of 2T pulseand line bar signals onto subjec-tively determined scales of pic-ture quality. The various distor-tions are weighted in terms ofimpairment to the picture.

The usual K Factor measure-ments are Kpulse/bar, K2T orKpulse (2T pulse response),Kbar, and sometimes K60 Hz.The overall K Factor rating is thelargest value obtained from all ofthese measurements. Specialgraticules can be used to obtainthe K Factor number or it can becalculated from the appropriatef o rmula. Definitions of the four KFactor parameters are as follows:

K2T. K2T is a weighted functionof the amplitude and time of dis-tortion occurring before and afterthe 2T pulse. In practice, agraticule is almost always usedto quantify this distortion.D i ff e rent countries and standard suse slightly different amplitudeweighting factors. The 1780Rgraticule is shown in Figure 59.

Kpulse/bar. Calculation of thisparameter requires measurementof the pulse and bar amplitudes.Kpulse/bar is equal to:

1/4 [(bar-pulse)/pulse] X 100%.

Kbar. A line bar (18 microsec-onds) is used to measure thisparameter. Locate the center ofthe bar time, normalize thatpoint to 100%, and measure themaximum amplitude deviationfor each half of the bar ignoringthe first and last 2.5% (0.45microsecond). The largest of thetwo tilt measurements is theKbar rating.

K60 Hz. A field square wave isused to measure this parameter.Locate the center of the field bartime, normalize that point to100%, and measure the maxi-mum amplitude deviation foreach half of the bar ignoring thefirst and last 2.5% (about 200microseconds). The largest of thetwo tilt measurements, dividedby two, is the K60 Hz rating.

38

K Factor Ratings

W2

S.D. 5%

2% W1

B1 B2

K2T 5%

C

GRAT. B

Figure 59. The outer dotted lines at the bottom of the 1780R externalgraticule indicate 5% K2T limits.

PICTURE EFFECTSAll types of linear distortionsaffect K Factor rating. Pictureeffects may include any of theaberrations discussed in the sections on short time, line time, field time, and long time distortions.

Since overall K factor rating isthe maximum value obtained in the four measurements, thepicture effects corresponding to a given K Factor rating mayvary widely. However, the sub-jective impairment is assumed to be equivalent.

TEST SIGNALSK parameters (except K60 Hz)can be measured with any testsignal containing a 2T pulse and an 18 microsecond line bar.A field square wave is requiredfor measurement of K60 Hz.The composite signal shown in Figure 60 includes therequired elements.

MEASUREMENT METHODSWaveform Monitor. The short timedistortion external graticule(Graticule B) provided with1780R and 1480 waveform mon-itors also includes a 5% K2Tlimit. To make a measurement,use the variable gain control toset the top of the 2T pulse to thesmall circle on the graticule (seeFigure 61). Set the horizontalmagnification to 200 nanosec-onds (0.2 microseconds) per

division. Under these condi-tions, the outer dotted lines atthe bottom of the graticule repre-sent 5% K. Enabling the X5 vertical gain, in addition to thevariable gain required to normal-ize the pulse height, will changethe graticule indication to 1% Klimit. Other K Factor readingsmay be interpolated.

The 1780R is also equipped withan electronic K2T graticule.Select K FACTOR in the MEA-SURE menu and set the horizon-tal magnification to 200 nanosec-onds (0.2 microseconds) perdivision. Set the pulse ampli -tude to 100 IRE which corre-sponds to the small cross drawnwith the beam. Use the largeknob to adjust the graticule sizeuntil it just touches the wave-form. The K2T distortion, in per-cent, is then shown on the readout (see Figure 62).

The standard internal graticulefor the 1780R and 1480 includesKpulse/bar marks in the centernear the top. To use this gratic-ule, first normalize the baramplitude to 100 IRE. Thencompare the amplitude of the 2T pulse to the Kpb scale, andobtain a K Factor reading in percent (see Figure 63).

The other K factor measurementscan be made either with thegraticule or with the voltage cur-sors. Refer to the definitions onpage 38 for general procedures.

39

Figure 60. A composite signal (also known as the FCC Composite) with theelements required for K Factor measurements.

Figure 61. A 2T pulse properly positioned for a K2T measurement with the1780R external graticule.

Figure 62. The 1780R electronic K Factor graticule.

VM700T Automatic Measurement.Select K FACTOR in theVM700T MEASURE mode toobtain a measurement of K2T.Either an EIA or a CMTT gratic-ule may be used for the measure-ment. The graticule can be set toautomatically track the wave-form or adjusted manually withthe front panel knob. This dis-play also provides numeric K2Tresults and a pulse-to-bar ratioreading (see Figure 64). Thispulse-to-bar ratio is not aKpulse/bar reading (see Note21). These measurements arealso available in the VM700TAUTO mode.

NOTES21. Pulse-to-Bar Definitions. Thereare several different methods ofexpressing the relationshipbetween the pulse and baramplitudes and it is important tounderstand the differencebetween methods and which isbeing specified. Three of themost common definitions aregiven below.

PULSE-TO-BAR RATIO =(pulse/bar) X 100%

PULSE-BAR INEQUALITY =(pulse-bar) X 100%

Kpulse/bar =1/4 [(bar-pulse)/pulse] X 100%

40

Figure 63. The 1780R graticule indicates 2% Kpulse/bar.

Figure 64. The VM700T 2T Pulse K Factor measurement display.

Amplitude dependent waveformdistortions are often referred toas nonlinear distortions. Thisclassification includes distor-tions that are dependent on APL(Average Picture Level) changesand/or instantaneous signal level changes.

Since amplifiers and other elec-tronic circuits are linear overonly a limited range, they tendto compress or clip large signals.The result is nonlinear distortionof one type or another. Nonlineardistortions may also manifestthemselves as crosstalk andintermodulation effects betweenthe luminance and chrominanceportions of the signal.

The first three distortions dis-cussed in this section are differ-ential phase, differential gain,and luminance nonlinearity.These are by far the most famil-iar and most frequently mea-sured nonlinear distortions.These parameters are includedin the performance specifica-tions of most video equipmentand are regularly evaluated intelevision facilities. The otherdistortions are generally notmeasured as frequently, howev-er, they are included in mostmeasurement standards and performance checks.

It is recommended that nonlineardistortions be measured at midAPL and at the APL extremes.Some test signal generators provide signals with differentAPL by combining the test signalwith a variable level pedestal.This is usually accomplished by alternating between one lineof the test signal and a group of four lines of the pedestal with the sequence repeatedthroughout the field. Since in-service measurements cannotbe made with full field test signals, measurements requiringc o n t rol of APL are often excludedfrom routine testing.

41

III. NONLINEAR DISTORTIONS

DEFINITIONDifferential phase distortion,often referred to as "diff phase"or "dP", is present when chromi-nance phase is affected by lumi-nance level. This distortionoccurs when chrominance infor-mation is not uniformlyp rocessed at all luminance levels.

Differential phase distortion ise x p ressed in degrees of subcarr i e rphase. Since both positive andnegative (lead and lag) phaseerrors may occur in the same signal, it is important to specifywhether the peak-to-peak phaseerror or maximum deviationfrom zero is being quoted. Ingeneral, NTSC measurementstandards (and this booklet) referto peak-to-peak measurements.

Differential phase should bemeasured at different averagepicture levels and the worsterror quoted.

PICTURE EFFECTSWhen diff e rential phase distort i o nis present, changes in hue occurwhen picture brightness changes.Colors may not be properlyreproduced, particularly in highbrightness areas of the picture.

TEST SIGNALSDifferential phase is measuredwith a test signal that consists ofuniform phase chrominancesuperimposed on different lumi-nance levels. A modulated stair -case (5 or 10 step) or a modulatedramp is typically used (seeFigures 65 and 66). A ramp isnormally used when performingmeasurements on devices andsystems that convert the signalfrom analog to digital and backto analog.

MEASUREMENT METHODSDifferential phase can be easilymeasured after the chrominancehas been demodulated and presented on a vector display.Although a standard vector display can indicate the pre s e n c eof large amounts of distortion, avectorscope equipped with aspecial DIFF PHASE mode or anautomatic measurement set suchas the VM700T is required forprecision measurements.

42

Differential Phase

Figure 65. A 5-step modulated staircase signal.

Figure 66. A modulated ramp signal.

Vectorscope Display. In a vec-torscope display, elongation ofthe dot in the direction of thegraticule circumference indicatesthe presence of differentialphase. Measurements are madeby using the vectorscope vari -able gain control to bring the signal vector out to the graticulecircle and reading the amount ofdistortion from the graticule.Vectorscope graticules generallyhave marks on the left-hand side to help quantify the error(see Figure 67).

R-Y Sweep. Although errors showup in the vectorscope display,there are some advantages to begained by examining the demod-ulated R-Y signal in a voltageversus time display. (Recall thatthe R-Y signal drives the verticalaxis of a vectorscope.) First ofall, more gain and thereforemore measurement resolution ispossible in waveform displays.Secondly, the sweep displayfacilitates correlation of the R-Ysignal with the original test sig-nal in the time dimension. Thisallows determination of exactlyhow the effects of differentialphase vary with luminance levelor how they vary over a field.

Precise measurements of differ-ential phase are therefore madeby examining a voltage versustime display of the demodulatedR-Y information. Distortionsmanifest themselves as tilt orlevel changes across the line.

Two different types of R-Y dis-plays, known as "single trace"and "double trace" , can be usedto make this measurement. Asdescribed below, different mea-surement techniques are usedwith the two displays. In the1780R, these modes are bothaccessed by selecting DIFFPHASE in the MEASURE menu.The SINGLE/DOUBLE touch-screen selection determineswhich of the two displays will appear.

R-Y Sweep - Single Trace Method. Inthe single trace mode, distort i o n sare quantified by comparing theR-Y waveform to a verticalgraticule scale.

To make a measurement, first setthe signal vector to the reference(9 o'clock) phase position. Usethe vectorscope variable gaincontrol to set the signal vectorout to the edge of the vec-torscope graticule circle. Makesure the 1780R waveform moni-tor gain is in the calibrated (1 volt full scale) setting.

The R-Y display appears on thewaveform (right-hand) screen inthe 1780R. Each major division(10 IRE) on the vertical graticulescale corresponds to one degreewhen the R-Y waveform is beingdisplayed. The amount of differ -ential phase distortion may bedetermined by measuring thelargest vertical deviationbetween two parts of the signal(see Figure 68).

43

F i g u re 67. A vectorscope display showing a 5 degree diff e rential phase distort i o n .

Figure 68. The 1780R single trace DIFF PHASE display indicating 6 degreesof differential phase distortion. The 0.00 D PHASE readout indicates aproperly adjusted display, ready for the measurement result to be readfrom the graticule.

R-Y Sweep - Double Trace Method.The double trace method pro-vides a more accurate way ofmeasuring the tilt in a one-linesweep of the R-Y information.Instead of comparing the wave-form to a graticule, the vec-torscope calibrated phase shifteris used to quantify the error.

The double trace display, whichappears on the waveform screenin the 1780R, is produced bydisplaying the single trace R-Yinformation non-inverted forhalf the lines and inverted forthe other half. As shifting phasemoves the two traces verticallywith respect to each other, mea-surements can be made by intro-ducing calibrated amounts ofphase shift with the vectorscopephase control. The basic tech-nique involves establishing a reference at one extreme of thetilt by bringing the inverted andnon-inverted traces together atthat point. The amount of phaseshift required to bring the twotraces together at the otherextreme of the tilt is the differen-tial phase distortion.

Select DOUBLE in the 1780RDIFF PHASE mode to make thismeasurement. Use the phaseshifter to set the signal vector tothe reference (9 o'clock) phaseposition. Neither vectorscope orwaveform monitor gain is criti -cal in this mode (see Note 22),however, setting the vector tothe graticule circle is a goodstarting point. Use the phaseshifter to bring the largest nega-tive excursion of the upper (non-inverted) waveform to meet itsmirror image. When they justtouch, press REF SET to set thephase readout to 0.00 degrees(see Figure 69). Now use thephase shifter to bring the largestpositive excursion in the uppertrace to meet its mirror image.The readout will indicate theamount of differential phasedistortion (see Figure 70).

A similar DOUBLE MODE tech-nique is used with the 520AVectorscope. Start by setting theCALIBRATED PHASE dial tozero. Use the A or B phase con-trol to null the largest negativeexcursion and then use the cali-brated phase shifter to null thelargest positive excursion. Thenumber above the calibratedphase dial will indicate theamount of differential phase.

44

Figure 69. The 1780R double trace DIFF PHASE display with the phasereadout zeroed.

Figure 70. The double trace DIFF PHASE display with the measurementresults indicated on the readout.

VM700T Automatic Measurement.To make an automatic measure-ment of differential phase withthe VM700T, select DG DP in theMEASURE mode. Both differen-tial phase and differential gainare shown on the same display(see Figure 71). The lower graphis differential phase. These mea-surements are also available inthe AUTO mode.

NOTES22. 1780R Waveform and VectorGains. In the single trace mode,the vector gain must be set sothe signal vector extends to thegraticule circle. The waveformgain must be in the calibratedposition. The graticule is cali -brated to one degree per divisiononly under these conditions.

With the double mode display,however, more gain may be usedfor greater resolution. Additionalvectorscope gain and/or wave-form vertical gain can be select-ed without affecting the mea-surement results.

23. Signal Vector. The test signalsused for measuring differentialphase and gain may have either20 IRE or 40 IRE chrominance atthe same phase as color burst.With 40 IRE chrominance theburst and signal vectors coincideon the vectorscope display. Witha 20 IRE signal, the burst andsignal vectors will have the samephase but different amplitudes.In this case, be sure to set the 20IRE signal rather than the 40 IREburst out to the vector graticulecircle. The signal gain is whatmust be normalized.

24. Noise Reduction Filter. A digitalrecursive filter is available in the1780R to facilitate differentialphase and gain measurements inthe presence of noise. Select theNOISE REDUCTION ON touch-screen selection in the DIFFPHASE or DIFF GAIN menu toenable the filter. The filterremoves approximately 15 dB ofnoise from the signal withoutany loss of bandwidth or hori-zontal resolution. This mode isparticularly useful for VTR andtransmitter measurements.

45

Figure 71. The VM700T DG DP display.

DEFINITIOND i ff e rential gain, often re f e rred toas "diff gain" or "dG", is presentwhen chrominance gain is aff e c t e dby luminance level. This distort i o noccurs when chrominance infor-mation is not uniformly pro c e s s e dat all luminance levels.

The amount of differential gaindistortion is expressed in per-cent. Since both attenuation andpeaking of chrominance canoccur in the same signal, it isimportant to specify whether themaximum overall amplitude dif-ference or the maximum devia-tion from the blanking levelamplitude is being quoted. Ingeneral, NTSC measurementstandards (and this booklet)specify the largest amplitudedeviation between any two lev-els expressed as percent of thelargest chrominance amplitude.

Differential gain should be mea-sured at different average picturelevels and the worst error quoted.

PICTURE EFFECTSWhen differential gain distortionis present, changes in color saturation occur when picturebrightness changes. Colors maynot be properly reproduced, particularly in high brightnessareas of the picture.

TEST SIGNALSDifferential gain is measuredwith a test signal that consists ofuniform amplitude chrominancesuperimposed on different luminance levels. A modulated staircase (5 or 10 step) or a modulated ramp is typicallyused (see Figures 72 and 73).

MEASUREMENT METHODSDifferential gain distortions canbe quantified in a number of ways.Chrominance amplitudes can bemeasured directly with a wave-form monitor and large distor-tions can be seen on a vectorscopedisplay. For precision measure-ments, however, a vectorscopewith a special DIFF GAIN modeor an automatic measurement setsuch as the VM700T is required.

Vectorscope Display. Elongation ofthe dot in the radial directionindicates the presence of differ -ential gain in a vectorscope display. Measurements can bemade by using the vectorscopevariable gain control to bring thesignal vector out to the graticulecircle and reading the amount ofd i s t o rtion from the graticule. Mostvectorscope graticules have spe-cial marks on the left side to helpquantify the error (see Figure 74).

46

Differential Gain

Figure 72. A 5-step modulated staircase signal.

Figure 73. A modulated ramp signal.

Figure 74. A vectorscope display indicating 15% differential gain.

Waveform Monitor/ChrominanceFilter. Differential gain measure-ments can also be made with awaveform monitor. This processis facilitated by enabling thechrominance filter which passesonly the chrominance portion ofthe signal. Peak-to-peak chromi-nance amplitude can be easilymeasured in the resulting dis-play. To make a measurement,first normalize the peak-to-peakamplitude of the highest chromi-nance level to 100 IRE. Thenmeasure the peak-to-peak ampli-tude of the lowest chrominancelevel. The amplitude difference,expressed in percent, is theamount of differential gain distortion (see Figure 75).

This measurement can also bemade by using the 1780R voltagecursors in the RELATIVE mode.Define the peak-to-peak ampli-tude of the highest chrominancelevel as 100%. Then move thecursors to measure peak-to-peakamplitude of the lowest chromi-nance level. The amplitude difference, expressed in percent, is the amount of differential gain distortion.

In either case, remember that itis the difference between thehighest and lowest chrominanceamplitudes that represents theamount of differential gain distortion. If the lowest chromi-nance level is 90% of the highest,a differential gain error of 10%should be quoted.

B-Y Sweep. Some vectorscopesare equipped with a specialmode for making accurate

differential gain measurements.A line sweep of demodulated B-Y information is displayed witherrors manifested as tilt or levelchanges across the line. Like theR-Y display used to measuredifferential phase, this displayprovides greater resolution andan indication of how distortionvaries over a line or field. In the1780R, both "single trace" and"double trace" versions of thisdisplay are available. Both areaccessed by selecting DIFF GAINin the MEASURE menu.

B-Y Sweep - Single Trace Method.The single trace differential gaindisplay is familiar to users of the520A Vectorscope and is alsoavailable in the 1780R by selectingSINGLE in the DIFF GAIN menu.E rrors are quantified by comparingthe demodulated waveform to avertical graticule scale.

Use the phase shifter to set thesignal vector to the reference (9 o'clock) position prior to making this measurement.Adjust the vectorscope variablegain control so the signal vectorextends to the edge of the gratic-ule circle. Make sure the 1780Rw a v e f o rm gain is in the calibrated(1 volt full scale) setting.

In the 1780R, the differentialgain display appears on thewaveform screen. Compare thewaveform to the vertical scale onthe graticule and measure thelargest deviation between anytwo parts of the signal. Onemajor graticule division (10 IRE)is equal to one percent (seeFigure 76).

47

Figure 75. A chrominance filter display indicating about 7% differential gain.

Figure 76. The 1780R single trace DIFF GAIN display indicating a distortionof 6%. As with diff phase, the single trace diff gain measurement result isread off the graticule, not the D GAIN readout.

B-Y Sweep - Double Trace Method.The double trace method pro-vides a highly accurate way ofmeasuring the amount of tilt in a one-line sweep of the B-Yinformation. This method is verysimilar to the differential phasedouble trace method, however, acalibrated gain control ratherthan a calibrated phase controlis used to null the traces.

Select DOUBLE in the 1780RDIFF GAIN menu to make thismeasurement. Use the phaseshifter to set the vector phase tothe reference (9 o'clock) posi-tion. The vectorscope variablegain must be adjusted so the sig-nal vector reaches the graticulecircle. The 1780R waveformmonitor gain setting is not criticalin this mode (see Note 26).

Start the measurement proce-dure by using the large knob tobring the largest negative excur-sion of the upper (non-inverted)waveform to meet its mirrorimage. Press REF SET to set thereadout to 0.00 percent (seeFigure 77). Use the large knob tobring the largest positive excur-sion of the upper trace to meetits mirror image. The readoutwill indicate the amount of differential gain distortion (see Figure 78).

VM700T Automatic Measurement.To make an automatic measure-ment of differential gain withthe VM700T, select DG DP in theMEASURE mode. Both differen-tial phase and differential gainare shown on the same display.The upper graph is differentialgain. These measurements arealso available in the AUTOmode (see Figure 79).

NOTES25. Demodulated "B-Y" Signal. Itshould be noted that in instru-ments such as the 520A and the1780R, the displayed signal isnot simply the B-Y demodulatoroutput of the vectorscope.Rather, an envelope (square law)detector scheme is used. Thedemodulated signal is derivedby multiplying the signal byitself rather than by a constant-phase CW subcarrier as in a synchronous demodulator.The primary advantage of thismethod is that in the presence ofboth differential phase and dif -ferential gain, synchronousdetection yields a phase-depen-dent term while square lawdetection does not. Thus thepresence of differential phasedoes not affect the differentialgain result.

48

Figure 77. The 1780R double trace DIFF GAIN display with the gainreadout zeroed.

Figure 78. The 1780R double trace DIFF GAIN display with the measurementresults indicated on the readout.

Figure 79. The VM700T DG DP display.

26. 1780R Waveform and VectorGains. When using the singletrace mode, the vector gain mustbe set to the graticule circle andthe waveform gain must be inthe calibrated position. Thegraticule is calibrated to one percent per division only underthese conditions.

In the double mode display,more gain may be introduced inthe waveform vertical (X5 orVAR) for greater resolution.However, it is critical that thevectorscope gain be set to thegraticule circle to obtain correctmeasurement results.

27. Simultaneous Display of DP and DG. It is sometimes useful tohave a display that shows bothdifferential phase and differen-tial gain, particularly whenadjusting equipment for mini-mum distortion. A displaywhich shows a one-line sweepof differential phase on the leftand a one-line sweep of differen-tial gain on the right can beaccessed by selecting DP & DGin the 1780R MEASURE menu(see Figure 80). The VM700T DGDP display also shows both distortions simultaneously.

49

Figure 80. The 1780R DP & DG display (single trace mode only).

DEFINITIONLuminance nonlinearity, or dif-ferential luminance, is presentwhen luminance gain is affectedby luminance level. In otherwords, there is a nonlinear rela-tionship between the input andoutput signals in the luminancechannel. This amplitude distor-tion occurs when luminanceinformation is not uniformlyprocessed over the entireamplitude range.

The amount of luminance non-linearity distortion is expressedas a percentage. Measurementsare made by comparing theamplitudes of the individualsteps in a staircase test signal.The result is the differencebetween the largest and smalleststeps expressed as a percentageof the largest step.

Luminance nonlinearity shouldbe measured at different averagepicture levels and the worsterror quoted.

PICTURE EFFECTSLuminance nonlinearity is notparticularly noticeable in blackand white pictures. However, iflarge amounts of distortion arepresent, loss of detail in theshadows and highlights may beseen. These effects correspond tocrushing or clipping of the blackand white information.

In color pictures, luminancenonlinearity is often morenoticeable. This is because colorsaturation, to which the eye ismore sensitive, is affected.

TEST SIGNALSLuminance nonlinearity shouldbe measured with a test signal thatconsists of uniform-amplitudeluminance steps. Unmodulated 5 step or 10 step staircase signalsare typically used.

If an unmodulated signal is notavailable, the measurement canalso be made with a modulatedstaircase. This is generally notgood practice, however, sinceboth differential gain and lumi-nance nonlinearity can have thesame net effect on the signal.

MEASUREMENT METHODSLuminance nonlinearities arequantified by comparing the stepamplitudes of the test signal.Since the steps are generated atuniform height, any differencesare a result of this distortion.The waveform in Figure 82exhibits luminance nonlinearity.Note that the top step is shorterthan the others.

50

Luminance Nonlinearity

Figure 81. A 10-step staircase test signal.

Figure 82. An example of luminance nonlinearity distortion.

Waveform Display. This measure-ment can be made with a wave-form monitor by individuallymeasuring each step in the testsignal. It is most convenient touse the variable gain to norm a l i z ethe largest step to 100 IRE sopercentage can be read directlyfrom the graticule. Voltage cur-sors can also be used to measurethe steps. Although this methodcan yield very accurate results, it is time consuming and not frequently used in practice.

Waveform Monitor - DifferentiatedStep Filter. Some waveform moni-tors are equipped with a specialfilter, usually called a "diff step"filter, for measurement of lumi-nance nonlinearity. When thisfilter is selected, each step tran-sition appears as a spike on thedisplay. As the amplitude ofeach spike is proportional to thecorresponding step height, theamount of distortion can bedetermined by comparing thespike amplitudes.

Either the waveform monitorgraticule or the voltage cursorscan be used to measure thespikes. Use the variable gain tonormalize the largest spikeamplitude to 100 IRE whenusing the waveform monitorgraticule. The differencebetween the largest and smallestspikes, expressed as a percentageof the largest, is the amount ofluminance nonlinearity.

The 1780R voltage cursorsshould be in the RELATIVEmode for this measurement.Define the largest spike ampli -tude as 100%. Leave one cursorat the top of the largest spike,and move the other cursor to thetop of the smallest spike. Thereadout will indicate the amountof luminance nonlinearity dis-tortion (see Figure 83).

VM700T Automatic Measurement.Select LUMINANCE NONLIN-EARITY in the VM700T MEA-SURE menu to obtain a displayof this distortion. The VM700Tuses a differentiated step filter in making this measurement (see Figure 84). This measure m e n tis also available in the AUTO mode.

51

Figure 83. The 1780R voltage cursors indicate 9.8% luminance nonlinearity.

Figure 84. The VM700T Luminance Nonlinearity display.

DEFINITIONChrominance nonlinear phase ispresent when chrominancephase is affected by chromi-nance amplitude. This distortionoccurs when all amplitudes ofchrominance information are notuniformly processed.

Chrominance nonlinear phasedistortion is expressed indegrees of subcarrier phase. Thisparameter should be measured atdifferent average picture levelsand the worst error quoted.

PICTURE EFFECTSChrominance nonlinear phased i s t o rtion will cause a shift in hueas color saturation increases. Theeffect is most noticeable in highamplitude chrominance signals.

TEST SIGNALSChrominance nonlinear phase ismeasured with a modulatedpedestal test signal. This signalconsists of a single phase, threelevel chrominance packet super-imposed on a constant lumi-nance level (see Figure 85). Atypical modulated pedestal signal will have a 50 IRE lumi-nance level and 20, 40, and 80 IRE chrominance levels.

MEASUREMENT METHODSChrominance nonlinear phase isquantified by measuring the phasechange of the chrominance levelsfrom their nominal phase value.In some cases, this distortion isdefined as the peak-to-peak phasechange between the three levels.

Vectorscope. Since phase infor-mation is required, a vectorscopeis used to measure chrominancenonlinear phase. Examine thethree dots (which correspond tothe three chrominance levels)and measure the maximumphase difference between thethree signal vectors. This is easiest when the vectorscopevariable gain is adjusted to bringthe largest vector out to thegraticule circle (see Figure 86).When using a 1780R or a 520AVectorscope, the calibratedphase shifter can be used toobtain a precise reading.

VM700T Automatic Measurement.Select CHROMA NONLINEARITYin the VM700T MEASURE modeto obtain a display of this distor-tion. The chrominance nonlinearphase measurement is the middlegraph in the display (see Figure87). These measurements arealso available in the AUTO mode.

52

Figure 85. A modulated pedestal test signal.

Figure 86. The 1780R vectorscope display indicating a chrominance nonlinearphase distortion of 8 degrees.

Figure 87. The VM700T Chrominance Nonlinearity display. Chrominancenonlinear phase is shown in the center graph.

Chrominance Nonlinear Phase

DEFINITIONChrominance nonlinear gain isp resent when chrominance gain isa ffected by chrominance amplitude.This distortion occurs when allamplitudes of chrominanceinformation are not uniformlyprocessed.

Chrominance nonlinear gain distortion is expressed in IRE orpercent. This distortion shouldbe measured at different averagepicture levels and the worsterror quoted.

PICTURE EFFECTSChrominance nonlinear gain dis-tortion will cause a change incolor saturation as chrominanceamplitude increases. The effectis most noticeable in high ampli-tude chrominance signals.

TEST SIGNALSChrominance nonlinear gain ismeasured with a modulated ped-estal test signal. This signal con-sists of a single phase, three levelc h rominance packet superimposedon a constant luminance level(see Figure 88). A typical modu-lated pedestal signal will have a50 IRE luminance level and 20, 40,and 80 IRE chrominance levels.

MEASUREMENT METHODSChrominance nonlinear gain dis-tortion is quantified by measur-ing the amplitude deviation ofthe chrominance levels fromtheir nominal values with the 40IRE level used as the reference.

Waveform Monitor. The waveformmonitor graticule should be usedfor this measurement. First usethe waveform monitor variablegain to normalize the middlesubcarrier packet to its nominalvalue of 40 IRE. The chro m i n a n c enonlinear gain distortion is thelargest deviation from nominalvalue of the other two levels. Theresult may be expressed in IREor as a percentage of the nominalamplitude of the affected packet.The waveform in Figure 89exhibits 14 IRE of chrominancenonlinear gain distortion.

VM700T Automatic Measurement.Select CHROMA NONLINEARITYin the VM700T MEASURE modeto obtain a display of this distort i o n .The chrominance nonlinear gainmeasurement is the top graph inthe display (see Figure 90).These measurements are a l s oavailable in the AUTO mode.

NOTES28. Chroma Filter. It is sometimesrecommended that the chromafilter on the waveform monitorbe enabled when measuringchrominance nonlinear gain.While the chroma filter will makethe display more symmetrical, thesame results should be obtainedeither way since it is the peak-to-peak amplitudes being mea-sured. A possible exception is a case where chrominance har-monic distortion is present. Thechrominance filter can removethe effects of harmonic distor-tion which are likely to be differ-ent for each chrominance level.

53

Chrominance Nonlinear Gain

Figure 88. A modulated pedestal test signal.

Figure 89. A chrominance nonlinear gain distortion of 14 IRE is shownin this display.

Figure 90. The VM700T Chrominance Nonlinearity display. Chrominancenonlinear gain is shown on the upper graph.

DEFINITIONChrominance-to-luminanceintermodulation, also known ascrosstalk or cross-modulation, ispresent when luminance ampli-tude is affected by superimposedchrominance. The luminancechange may be caused by clippingof high-amplitude chrominancepeaks, quadrature distortion, or crosstalk.

The deviation in the pedestallevel may be expressed:

• In IRE with pedestal level normalized to 50 IRE.

• As a percentage of the pedestal level.

• As a percentage of the measured white bar amplitude.

• As a percentage of 714 millivolts.

These definitions will yield diff e r-ent results under some conditionsso it is important to standardizeon a single method of makingintermodulation measurements.

PICTURE EFFECTSChrominance-to-luminanceintermodulation will cause variations in brightness due tocolor saturation changes.

TEST SIGNALSChrominance-to-luminanceintermodulation is measuredwith a modulated pedestal testsignal. This signal consists of asingle phase, three level chromi-nance packet superimposed on aconstant luminance level (seeFigure 91). A typical modulatedpedestal signal will have a 50IRE luminance level and 20, 40,and 80 IRE chrominance levels.

MEASUREMENT METHODSChrominance-to-luminanceintermodulation is quantified bymeasuring the change in lumi-nance level caused by thechrominance information super-imposed on it. This process isfacilitated by removing thechrominance information fromthe display with a waveformmonitor filter.

54

Chrominance-to-Luminance Intermodulation

Figure 91. A modulated pedestal test signal.

Waveform Monitor. The chromi-nance information can be filtered off with either the lumi-nance or lowpass filter in the1780R. The lowpass filter shouldbe selected if using the 1480.The IRE filter is not really suit-able because some chrominanceremains in the display. The Ydisplay of the 520A vectorscopealso works well.

Details of the measurementmethod will depend on how theamount of distortion isexpressed. In general, first usethe variable gain on the wave-form monitor to normalize theportion of the pedestal withoutsuperimposed chrominance tothe measurement reference level.For an absolute value measure-ment, this would be the nominal

pedestal level. For a percentmeasurement, this could be 100IRE (see Figure 92). With thepedestal level normalized, measure the largest level shift at the top of the pedestal.

The 1780R voltage cursors canbe used in RELATIVE mode tomake this measurement. In Figure92, the amplitude deviation is9.2% of the pedestal level.

VM700T Automatic Measurement.Select CHROMA NONLINEARITYin the VM700T MEASURE modeto obtain a display of this distor-tion. The chrominance-to-lumi-nance intermodulation measure-ment is the bottom graph in thedisplay (see Figure 93). Thesemeasurements are also availablein the AUTO mode.

55

Figure 92. A chrominance-to-luminance intermodulation distortion of 9.2%.

Figure 93. The VM700T Chrominance Nonlinearity display. Chrominance-to-luminance intermodulation is shown in the lower graph.

DEFINITIONTransient gain distortion, alsoreferred to as transient nonlin-earity, is present when abruptchanges in APL temporarilyaffect signal amplitude. For thesynchronizing signal, the error isdefined as the maximum transientdeparture in the amplitude ofsync from the amplitude beforethe change in APL. It is generallyexpressed as a percentage of theoriginal amplitude, however, somestandards specify a percentage ofthe largest amplitude. Measure-ment of this distortion requiresan out-of-service test. Both low-to-high and high-to-low APLchanges should be evaluated.

PICTURE EFFECTSIf only the sync portion of thesignal is affected, this distortiondoes not generally cause percep-tible picture effects. However, therest of the signal often suffersfrom the same type of distortion.When this occurs, sudden switchesbetween high APL and low APLpictures can cause transientbrightness effects in the picture.

TEST SIGNALSTransient gain distortion is mea-sured with a flat field test signal(black burst with pedestal). A generator with a "bounce" feature can be used to make the APL transitions if the timeinterval between transitions isconsiderably longer than anytransient effect (see Figure 94).

MEASUREMENT METHODSTransient gain changes are mea-sured by abruptly changing APLand observing the transienteffects on a waveform monitor.

Waveform Monitor. This distortionis easiest to evaluate with thetest signal displayed on a wave-form monitor with the differenti -ated step filter selected. (Recallthat this filter produces spikeswith amplitudes proportional tosignal transitions). The waveformmonitor DC restorer must beturned off for this measurement.

Depending on the nature of thedistortion, it may be possible toobserve it with the waveformmonitor operating in the fieldsweep mode. Otherwise it willbe necessary to use the 1780RSLOW SWEEP mode. (Some1480s are equipped with theSLOW SWEEP option). A wave-form photograph may make themeasurement easier.

Adjust the waveform monitorvariable gain to set the ampli-tude of the positive spike, whichcorresponds to the trailing edgeof sync, equal to 100 IRE. Switchbetween extreme APL levels,typically 10% and 90%. Theresulting envelope of the syncspikes represents the transienterror. Measure the maximumdeparture from 100 IRE andexpress that number as a percentto obtain the amount of transientsync nonlinearity.

The 1780R voltage cursors canalso be used to make this mea-surement. In the RELATIVEmode, define the positive syncspike as 100%. Then use thecursors to measure the largestdeviation from that amplitude.

56

Transient Gain Distortion - Synchronizing Signal

IRE100

80

60

40

20

0

–4010% APL 90% APL 10% APL

X SECONDS X SECONDS

X SECONDS

Figure 94. A flat field bounce test signal.

DEFINITIONDynamic gain distortion of thepicture signal is present whenluminance amplitude is affectedby APL. Dynamic gain distortionof the sync signal is presentwhen sync amplitude is affectedby APL.

The amount of distortion is usu-ally expressed as a percent of theamplitude at 50% APL, althoughsometimes the overall variationin IRE units is quoted. This is anout-of-service test.

PICTURE EFFECTSWhen luminance levels are APLdependent, picture brightnessmay seem to be incorrect orinconsistent as scenes change.

TEST SIGNALSDynamic gain change is mea-sured with a test signal thatextends to 100 IRE and has anAPL that can be varied from10% to 90% (see Figure 95). A staircase signal with variablepedestal is commonly used.

MEASUREMENT METHODSWaveform Monitor. Dynamic pic-ture gain change is evaluated bymeasuring amplitude at variousAPL levels. First select 50% APL

and use the waveform monitorvariable gain to set the top stepof the staircase to 100 IRE. Varythe APL of the signal to 10%and then to 90%. At each APLlevel, record the level of thestaircase top step. The peak-to-peak variation of the top stair-case level, expressed in percent-age, is typically quoted as theamount of dynamic picture gainchange. This measurement canbe made with the 1780R voltagecursors in the RELATIVE mode.

A similar procedure is used fordynamic sync gain. Figures 96and 97 illustrate the measure-ment procedure.

NOTES29. Transient vs. Dynamic GainChanges. The terminology usedto describe the two differenttypes of APL dependent nonlin-ear gain distortions is inconsis-tent from standard to standardand can be quite confusing. It isnecessary to distinguish betweenthe effects of different APL, andeffects of changes in APL. Mostfrequently, the two measure-ments are referred to asDYNAMIC and TRANSIENTrespectively. This bookletadheres to this definition.

57

Dynamic Gain Change

100 IRE PED

50 IRE PED

25 IRE PED

10 IRE PED

BLACK (SETUP) LEVEL

F i g u re 95. A variable pedestal is multiplexed with a 100 IRE test signal forthis measure m e n t .

Figure 96. At 50% APL, the sync pulse amplitude is 40 IRE.

Figure 97. At high APL, the sync pulse amplitude is 36 IRE. This indicatesa dynamic sync gain change of 10%.

The electrical fluctuations whichwe refer to as noise form a verycomplex signal which does notlend itself to straightforwardamplitude measurements. Anumber of special techniqueshave therefore been developedfor measuring noise. A compre-hensive discussion of noise mea-surement is outside the scope ofthis publication. However, someof the methods that apply totelevision systems are presentedin this section.

Special filters are generallyrequired for noise measure-ments. These filters are used to separate the noise into its

various frequency componentsfor analysis. Each measurementstandard typically calls for threeor four measurements made with various combinations of thefilters. Note that specificationsfor the filters vary from standardto standard.

The tangential method of noisemeasurement, useful for makingoperational measurements ofrandom noise, is the onlymethod discussed in detail inthis publication. While not themost accurate technique, the tangential method can provide aquick way of keeping track ofsystem noise performance over

time. Tangential noise measure-ments are made with a speciallyequipped waveform monitor.This feature is standard in the 1780R.

Specialized equipment isrequired to completely charac-terize the noise performance of asystem. Until recently, thesecapabilities were available onlyin dedicated noise measurementinstruments. The VM700T,however, makes highly accuratenoise measurements using filtersimplemented in software. Thenoise measurement features ofthe VM700T are reviewed brieflyin this section.

58

IV. NOISE MEASUREMENT

DEFINITIONNoise refers to the fluctuationsthat are present in any electricalsystem. Noise can be either random or coherent and comesfrom a variety of natural andman-made sources. Althoughthere is always some noise present, an excessive amount isundesirable since it tends todegrade or obscure the signal.

Signal amplitudes do not always remain constant as thevideo signal is processed andtransmitted. An absolute mea-surement of noise is thereforenot particularly relevant - a certain amount of noise willhave very different effects onsignals of different amplitudes.

Since it is the amount of noiserelative to the signal amplitude(rather than the absolute amountof noise) that tends to causeproblems, measurement of signal-to-noise ratio, expressedin dB, is made.

PICTURE EFFECTSNoisy pictures often appeargrainy or snowy and sparkles of color may be noticeable.Extremely noisy signals may bedifficult for equipment to lock tocausing horizontal tearing andvertical roll.

TEST SIGNALSTangential noise measurementmay be made on any portion ofthe video signal with a constantluminance level without chromi-nance. The measurement can bemade on a line in the verticalinterval although full-field mea-surements are more accurate andsomewhat easier to make.

Any line with a constantpedestal level can be used tomake VM700T NOISE SPECTRUMmeasurements. A quiet line inthe vertical interval is typicallyused. The VM700T CHROMANOISE measurement requires ared field test signal (see Figure 98).

MEASUREMENT METHODSTangential Method. Tangentialnoise measurements may bemade with a 1780R with mea-surement results repeatable towithin 1 or 2 dB, down to noiselevels of about 60 dB. Filters canbe inserted in the AUX OUT/AUXIN path to separate noise compo-nents of different frequencies.

Make sure the waveform monitorfilter selection is set to FLAT(unless using the auxiliary filtercapability) and DC restorer toOFF or FAST. Select NOISE inthe 1780R MEASURE menu. Inthe 1480, use the WAVEFORMCOMPARISON mode to split theluminance levels of interest inhalf and overlay the two parts.

59

Signal-to-Noise Ratio

0 = 103.4

Figure 98. A red field test signal.

Figure 99. The 1780R tangential noise measurement mode showingexcessive trace separation.

Figure 100. The 1780R tangential noise measurement mode with trace separation properly adjusted. This signal has a signal-to-noise ratio of 36 dB.

The measurement is made byadjusting the separation betweenthe two traces until the dark areabetween them just disappears.When there is no perceptible dipin brightness between the twotraces, the calibrated offset level(in dB) is the amount of noise(see Figures 99 and 100). In the1780R, the large knob is used tocontrol the offset and the onscreen readout provides the dBreading. In the 1480, the offsetfunction is performed by the twodB NOISE controls in the lowerright hand corn e r. The dB re a d i n gis obtained from the knob settings.

VM700T Automatic Measurement.Select NOISE SPECTRUM in theVM700T MEASURE menu tomake signal-to-noise measure-ments. A spectral display as wellas numeric results is provided inthis mode (see Figure 101).

There are four filters available in this mode: 4.2 MHz lowpass, 5 MHz lowpass, a unifiedweighting filter, and a 3.58 MHztrap. Measurement standardstypically require three or fourmeasurements made with vari-ous combinations of these filters.

The rms signal-to-noise ratio ofthe entire spectrum is alwaysdisplayed in the upper righthand corner of the display.

A cursor can be used to select acertain frequency for a peak-to-peak noise measurement. Thecursors can also be used todefine a narrow range of fre-quencies for S/N measurements.

The CHROMINANCE AM PMselection in the VM700T MEA-SURE menu provides informa-tion about the noise that affectsthe chrominance portion of thesignal. Since the chrominancesignal is sensitive to both theamplitude (AM) and phase (PM)components of noise, separatemeasurements are provided. Aselection of filters is available inthis mode as well. Be sure to usethe red field test signal for thismeasurement (see Figure 102).Noise measurements are alsoavailable in the AUTO mode.

NOTES30. Quiet Lines. "Quiet lines" inthe vertical interval are some-times used to determine theamount of noise introduced in acertain part of the transmissionpath. A line is reinserted (and istherefore relatively noise free) atone end of the transmission pathof interest. This ensures that anynoise measured on that line atthe other end was introduced inthat part of the path.

60

Figure 101. The VM700T Noise Spectrum display.

Figure 102. The VM700T Chrominance AM PM Noise display.

In this section, we discuss twoparameters which should bemonitored and adjusted at thebroadcast television transmitter -depth of modulation and ICPM.These two measurements arecommonly made with timedomain instruments such aswaveform monitors or oscillo-scopes. Most of the other testsfor characterizing transmitterperformance are made with aspectrum analyzer and are notaddressed in this publication.

To make these measurements, ahigh-quality demodulator suchas the Tektronix TV1350 or 1450 is required. These instrumentsprovide envelope and synchro-nous detection demodulation.Unlike envelope detectors, syn-chronous detectors are notaffected by the quadrature dis-tortion inherent in the vestigialsideband transmission system.For measurement purposes, theeffects of quadrature distortionshould be removed so as not toobscure distortions from othersources.

A quadrature output is availablewhen the instrument is operatingin the synchronous detectionmode. Envelope detection ismost similar to the demodula-tion used in most homereceivers and is also available inthe TV1350 and 1450.

The TV1350 and 1450 produce azero carrier reference pulsewhich provides the referencelevel required for depth of modulation measurements. Thispulse is created at the demodu-lator output by briefly reducingthe amplitude of the RF signal tothe zero carrier level prior todemodulation.

61

V. TRANSMITTER MEASUREMENTS

DEFINITIONICPM (Incidental Carrier PhaseModulation) is present whenpicture carrier phase is affectedby video signal level. This dis-tortion occurs in the transmitter.

ICPM errors are expressed indegrees using the following definition:

ICPM = arctan (quadratureamplitude/video amplitude).

PICTURE EFFECTSThe effects of ICPM will dependon the type of demodulationused to recover the baseband signal from the transmitted signal.ICPM shows up in synchronous-ly demodulated signals as differ -ential phase and many othertypes of distortions. With enve-lope demodulation, the demodu-lation typically used in homereceivers, the baseband signal isgenerally not as seriously affect -ed and the effects of ICPM arerarely seen in the picture. Thesound, however, is another matter.

ICPM may manifest itself asaudio buzz in the home receiver.In the intercarrier sound system,the picture carrier is mixed withthe FM sound carrier to form a4.5 MHz sound IF. Audio ratephase modulation in the picturecarrier can therefore be trans-ferred into the audio system andheard as a buzzing noise.

TEST SIGNALSICPM is measured with anunmodulated linearity signal. Astaircase is generally used but aramp signal may also be used.

MEASUREMENT METHODSICPM is measured on an XY plotof VIDEO OUT versus QUADRA-TURE OUT with the demodula-tor operating in the synchronousdetection mode. A phase errorwill produce an output from thequadrature detector. If this phaseerror varies with amplitude, theresult is a tilted display. Thedemodulator zero carrier refer-ence pulse must be turned onand the detection mode set tosynchronous. Select the SLOWtime constant when using the 1450.

Waveform Monitor. To displayICPM on a waveform monitor,connect the demodulator out-puts to the waveform monitorinputs as shown in Figure 103.Select ICPM in the 1780R MEASURE menu or EXT HORIZon the 1480 front panel.

Although it is not strictly neces-sary, it is generally recommend-ed that the signals be lowpassf i l t e red to make the display easierto interpret. With either the1780R or the 1480, this can beaccomplished in the verticalchannel by selecting the LOW-PASS filter. Use an external 250 KHz lowpass filter for thehorizontal (see Figure 103).

62

ICPM

Figure 103. How to set up the 1780R for ICPM measurements.

The display resulting from thisconfiguration, which appears onthe right-hand screen in the1780R, is shown in Figures 104and 105. The amount of tilt(deviation from the vertical) isan indication of ICPM. There isno ICPM in the signal shown inFigure 104 while distortion ispresent in Figure 105. Whenadjusting the transmitter forminimum ICPM, make the lineas nearly vertical as possible.

The 1780R has an electronicgraticule which can be used toquantify the amount of tilt. Thewaveform should be positionedso the small dot correspondingto the zero carrier referencepulse is set on the cross at thetop of the screen. The horizontalmagnification will automaticallybe set to X25 when this mode isselected. If desired, X50 magnifi -cation can be used for greaterresolution. Start with the twograticule lines widely separatedand use the knob to move themtogether to the point where agraticule line first contacts oneof the dots. Disregard the "loops"in the ICPM display. These cor-respond to the staircase steptransitions and are not indicativeof distortion. The amount ofICPM distortion is indicated onthe screen (see Figure 105).

An external ICPM graticule isavailable for the 1480. Positionthe zero carrier reference pulse,which shows up as a small dot,on the cross at the top of thegraticule. The graticule is cali -brated for 2 degrees per divisionwhen the horizontal magnifier isset to X25 or 1 degree per divi-sion with X50 horizontal magni -fication. Read the amount ofICPM from the graticule at thepoint of maximum distortion.

VM700T Automatic Measurement.The VM700T provides an ICPMmeasurement in the AUTOmode. In this case the quadra-ture signal must be connected tothe “C” input.

NOTES31. Configuring the 1480. 1480-series instruments are shippedwith unblanking disabled in theEXTERNAL HORIZONTALmode to prevent damage to theCRT. ICPM measurements can bemade in line select with theinstrument in this mode. Forfull-field ICPM measurements,the unblanking must be enabled.Instructions on how to accom-plish this can be found in theOPERATING CHANGES sectionof the 1480-series manual.

32. Other XY Displays. Any XY display can be used to measureICPM. Connect QUADRATUREOUT to the horizontal andVIDEO OUT to the vertical anduse the formula given on page 62to calculate the amount of dis-tortion. For small errors, someamount of gain will be needed toimprove the measurement reso-lution. Lowpass filters in bothchannels are recommended.

33. Phase Noise. Some demodula-tors have large amounts of phasenoise which make it difficult tomake ICPM measurements. TheTektronix 1450 has sufficientlylow phase noise as do allTV1350 units shipped after July, 1998. Older TV1350 unitscan be retrofitted to improvephase noise performance.Contact your local Tektronix service department for informa-tion on how to update olderinstruments.

63

Figure 104. The 1780R ICPM display with no distortion present.

Figure 105. The 1780R electronic graticule indicating an ICPM distortionof 5 degrees.

ZERO CARRIER REFERENCE 2 /DIV MAGNIFIER X25

Figure 106. The 1480 ICPM graticule.

DEFINITIONDepth of modulation (percentageof modulation) measurementsindicate how video signal levelsare represented in the RF signal.

The NTSC modulation schemeyields an RF signal that reachesits maximum peak-to-peakamplitude at sync tip (100%). Ina properly adjusted signal, blanking level corresponds to 75%and peak white to 12.5%. Thezero carrier reference level corre-sponds to 0% (see Figure 107).

PICTURE EFFECTSOvermodulation often shows upas nonlinear distortions such asd i ff e rential phase and gain. ICPMor white clipping may also re s u l t .U n d e rmodulation often results indegraded signal-to-noise perfor-mance. The picture effects of thesed i s t o rtions were discussed earlier.

TEST SIGNALSDepth of modulation measure-ments can be made with any signal with peak white (100 IRE)and blanking level (0 IRE) refer-ences. This signal is used in conjunction with the zero carrierreference pulse (Figure 108)which typically occurs in thevertical interval on line 20.

MEASUREMENT METHODSModulation depth is measured atthe output of a precision demod-ulator by verifying that the videosignal modulates the vision car-rier as shown in Figure 108. Over-all amplitude is not critical, butit should be adjusted in the systemto be approximately 160 IRE fro msync tip to zero carrier at 100%transmitter power. This willminimize the effects of nonlinear-ities in the measurement system.

Waveform Monitor. Most waveformmonitors, including the 1780Rand 1480, provide a depth ofmodulation scale on the right-hand side of the graticule. Usethe variable gain to position thezero carrier reference pulse onthe 0% mark at the top of thescreen and sync tip on the 100%mark (see Figure 109). Verifythat white level and blankinglevel occur at the prescribedpoints on the graticule scale.The voltage cursors can also beused for this measurement.

NOTES33. Envelope Detection. Depth ofmodulation measurementsshould be made with the demod-ulator in the envelope detectionmode to minimize effects ofICPM. Quadrature distortion willnot affect modulation depth.

64

Depth of Modulation

IRE SCALEDEPTH OF

MODULATIONZERO CARRIER

REFERENCE

REFERENCEWHITE

BLANKING

SYNC TIP

120

100

0

– 40

0%

12.5%

75%

100%

Figure 107. Depth of modulation levels.

120 IRE

ZERO CARRIER REFERENCE PULSE

30 µSEC

Figure 108. The zero carrier reference pulse typically occurs on line 20 butcan be moved to other lines.

Figure 109. Video levels can be compared to the depth of modulation scaleon the right-hand side of the graticule. Note the zero carrier pulse at 120 IRE.

AC-COUPLED — A connectionwhich removes the constant v o l t-age (DC component) on which thesignal (AC component) is riding.Usually implemented by passingthe signal through a capacitor.

AM — Amplitude Modulation(AM) is the process by which theamplitude of a high-frequencycarrier is varied in proportion tothe signal of interest. In the NTSCtelevision system, AM is used toencode the color informationand to transmit the picture.

Several different forms of AMare differentiated by filtering ofthe sidebands and whether ornot the carrier is suppressed.Double sideband suppressed carrier is used to encode theNTSC color information, whilethe signal is transmitted with avestigial sideband scheme.

APL — Average Picture Level.The average signal level (withrespect to blanking) duringactive picture time, expressed asa percentage of the differencebetween the blanking and reference white levels.

BACK PORCH — The portion ofthe video signal which liesbetween the trailing edge of thehorizontal sync pulse and thestart of the active picture time.Burst is located on back porch.

BANDWIDTH — The range offrequencies over which signalamplitude remains constant(within some limit) as it ispassed through a system.

BASEBAND — Refers to thecomposite video signal as itexists before modulating the picture carrier. Composite video distributed throughout astudio and used for recording isat baseband.

BLACK BURST — Also called"color black", black burst is acomposite video signal consistingof all horizontal and verticalsynchronization information,burst, and usually setup.Typically used as the house reference synchronization signalin television facilities.

BLANKING LEVEL — Refers tothe 0 IRE level which existsbefore and after horizontal syncand during the vertical interval.

BREEZEWAY — The portion ofthe video signal which liesbetween the trailing edge of thehorizontal sync pulse and thestart of burst. Breezeway is partof back porch.

BROAD PULSES — Another namefor the vertical synchronizingpulses in the center of the vert i c a linterval. These pulses are longenough to be distinguished fromall others, and are the part of thesignal actually detected by vertical sync separators.

BURST — A small referencepacket of the subcarrier sinewave, typically 8 or 9 cycles,which is sent on every line ofvideo. Since the carrier is sup-p ressed, this phase and fre q u e n c yreference is required for syn-chronous demodulation of thecolor information in the receiver.

B-Y — One of the color differ -ence signals used in the NTSCsystem, obtained by subtractingluminance from the blue camerasignal. This is the signal whichdrives the horizontal axis of a vectorscope.

CHROMINANCE — C h ro m i n a n c erefers to the color information ina television picture. Chro m i n a n c ecan be further broken down intotwo properties of color: hue and saturation.

CHROMINANCE SIGNAL —The high-frequency portion ofthe video signal which isobtained by quadrature ampli-tude modulation of a 3.58 MHzsubcarrier by R-Y and B-Y.

COLOR DIFFERENCE SIGNALS — Signals used bycolor television systems to convey color information insuch a way that the signals go to zero when there is no color in the picture. R-Y, B-Y, I and Qare all color difference signals.

COMPONENT VIDEO — Videowhich exists in the form of threeseparate signals, all of which arerequired in order to completelyspecify the color picture. Forexample: R, G and B or Y, R-Y,and B-Y.

COMPOSITE VIDEO — A singlevideo signal containing all of thenecessary information to repro-duce a color picture. Created byadding quadrature amplitudemodulated R-Y and B-Y to theluminance signal.

CW — Continuous Wave. Refersto a separate subcarrier sinewave used for synchronizationof chrominance information.

dB (DECIBEL) — A decibel is alogarithmic unit used to describesignal ratios. For voltages,

dB = 20 Log10V1 V2

DC-COUPLED — A connectionconfigured so that both the signal (AC component) and theconstant voltage on which it isriding (DC component) arepassed through.

DC RESTORER — A circuit usedin picture monitors and wave-form monitors to clamp onepoint of the waveform to a fixedDC level.

DEMODULATOR — In general,this term refers to any devicewhich recovers the original sig-nal after it has modulated a highfrequency carrier. In television,it may refer to:

(1) An instrument, such as aTe k t ronix TV1350 or 1450, whichtakes video in its transmittedform (modulated picture carrier)and converts it to baseband.

(2) The circuits which recover R-Y and B-Y from the compositesignal.

EQUALIZER — The pulses whichoccur before and after the broadpulses in the vertical interval.

ENVELOPE DETECTION —A-demodulation process inwhich the shape of the RF envelope is sensed. This is theprocess used by a diode detector.

FIELD — In interlaced scan sys-tems, the information for onepicture is divided up into twofields. Each field contains onehalf of the lines required to produce the entire picture.Adjacent lines in the pictur eare in alternate fields.

65

GLOSSARY OF TELEVISION TERMS

FM — Frequency Modulation(FM) is the process by which thefrequency of a carrier signal isvaried in proportion to the signal of interest. In the NTSCtelevision system, audio infor-mation is transmitted using FM.

FRAME — A frame contains allthe information required for acomplete picture. For interlacedscan systems, there are twofields in a frame.

FRONT PORCH — The portionof the video signal between theend of active picture time and theleading edge of horizontal sync.

GAMMA — Since picture moni-tors have a nonlinear relation-ship between the input voltageand brightness, the signal mustbe correspondingly predistorted.Gamma correction is alwaysdone at the source (camera) intelevision systems: the R, G andB signals are converted to R 1/g ,G 1/g and B 1/g. Values of about2.2 are typically used for gamma.

GENLOCK — The process oflocking both sync and burst ofone signal to sync and burst ofanother, making the two signalscompletely synchronous.

GRATICULE — The scale whichis used to quantify the informa-tion on a waveform monitor orvectorscope display. Graticulesmay either be screened onto thefaceplate of the CRT itself (inter-nal graticule), or onto a piece ofglass or plastic which fits infront of the CRT (external gratic-ule). They can also be electroni-cally generated.

HARMONIC DISTORTION —If a sine wave of a single fre-quency is put into a system, andharmonic content at multiples ofthat frequency appears at theoutput, there is harmonic distor-tion present in the system.Harmonic distortion is caused bynonlinearities in the system.

HORIZONTAL BLANKING —Horizontal blanking is the entiretime between the end of theactive picture time of one lineand the beginning of active pic-ture time of the next line. Itextends from the start of frontporch to the end of back porch.

HORIZONTAL SYNC —Horizontal sync is the -40 IREpulse occurring at the beginningof each line. This pulse signalsthe picture monitor to go back tothe left side of the screen andtrace another horizontal line ofpicture information.

HUE — Hue is the property ofcolor which allows us to distin-guish between colors such asred, yellow, purple, etc.

HUM — Undesirable coupling ofthe 60 Hz power sine wave intoother electrical signals.

INTERCARRIER SOUND — Amethod used to recover audioinformation in the NTSC system.Sound is separated from videoby beating the sound carrieragainst the video carrier, produc-ing a 4.5 MHz IF which containsthe sound information.

IRE — A unit equal to 1/140 ofthe peak-to-peak amplitude ofthe video signal, which is typi-cally one volt. The 0 IRE point isat blanking level, with sync tipat -40 IRE and white extendingto +100 IRE. IRE stands forInstitute of Radio Engineers, theorganization which defined the unit.

LINEAR DISTORTION — Refers to distortions which areindependent of signal amplitude.

LUMINANCE — The signalwhich represents brightness, orthe amount of light in the pic-ture. This is the only signalrequired for black and white pictures, and for color systems it is obtained as a weighted sum (Y = 0.3R + 0.59G + 0.11B) of theR, G and B signals.

MODULATED — When referringto television test signals, thisterm implies that chrominanceinformation is present. (Forexample, a modulated staircasehas subcarrier on each step.)

MODULATION — A processwhich allows signal informationto be moved to other frequenciesin order to facilitate transmissionor frequency-domain multiplexing.See AM and FM for details.

NONLINEAR DISTORTION —Refers to distortions which areamplitude-dependent.

NTSC — National TelevisionSystem Committee. The organi-zation which developed the tele-vision standard currently in usein the United States, Canada andJapan. Now generally used torefer to that standard.

PAL — Phase Alternate Line.Refers to the television systemused in Europe and many otherparts of the world. The phase ofthe chrominance signal alter-nates from line to line to helpcancel out phase errors.

QUADRATURE AM —A process which allows two different signals to modulate asingle carrier frequency. The two signals of interest AmplitudeModulate carrier signals whichare the same frequency but differin phase by 90 degrees (hencethe Quadrature notation). Thetwo resultant signals can beadded together, and both signalsrecovered at the other end, ifthey are also demodulated 90degrees apart.

Q U A D R ATURE DISTORTION —Distortion resulting from theasymmetry of sidebands used investigial sideband televisiontransmission. Quadrature distor-tion appears when envelopedetection is used, but can beeliminated by using a synchro-nous demodulator.

RF — Radio Frequency. In tele-vision applications, RF generallyrefers to the television signalafter the picture carrier modula-tion process.

RGB — Red, Green and Blue.The three primary colors used incolor television's additive colorreproduction system. These arethe three color signals generatedby the camera and used by the picture monitor to produce a picture.

R-Y — One of the color differ -ence signals used in the NTSCsystem, obtained by subtractingluminance from the red camerasignal. The R-Y signal drives thevertical axis of a vectorscope.

66

SATURATION — The propertyof color which relates to theamount of white light in thecolor. Highly saturated colors arevivid, while less saturated colorsappear pastel. For example, redis highly saturated, while pink is the same hue but much less saturated.

SETUP — In NTSC systems,video black is typically 7.5 IREabove the blanking level. This7.5 IRE level is referred to as the black setup level, or simplyas setup.

SUBCARRIER — The modula-tion sidebands of the color sub-carrier contain the R-Y and B-Yi n f o rmation. For NTSC, subcarr i e rfrequency is 3.579545 MHz.

SYNCHRONOUS DETECTION —A demodulation process inwhich the original signal isrecovered by multiplying themodulated signal with the out-put of a synchronous oscillatorlocked to the carrier.

TERMINATION — In order toaccurately send a signal througha transmission line, there mustbe an impedance at the endwhich matches the impedance ofthe source and of the line itself.Amplitude errors and reflectionswill otherwise result. Video is a75 Ohm system, so a 75 Ohmterminator must be put at theend of the signal path.

UNMODULATED — When usedto describe television test sig-nals, this term refers to pulsesand pedestals which do not havehigh-frequency chrominanceinformation added to them.

VECTORSCOPE — A specializedoscilloscope which demodulatesthe video signal and presents adisplay of R-Y versus B-Y. Theangle and magnitude of the dis-played vectors are respectivelyrelated to hue and saturation.

VERTICAL INTERVAL — The synchronizing informationwhich appears between fieldsand signals the picture monitorto go back to the top of the scre e nto begin another vertical scan.

WAVEFORM MONITOR —A specialized oscilloscope for evaluating television signals.

Y — A b b reviation for luminance.

ZERO CARRIER REFERENCE —A 120 IRE pulse in the verticalinterval which is produced bythe demodulator to provide areference for evaluating depth of modulation.

67

There are two basic types ofNTSC color bar signals in com-mon use. The terms "75% bars"and "100% bars" are generallyused to distinguish between thetwo types. While this terminolo-gy is widely used, there is oftenconfusion about exactly whichparameters the 75% versus100% notation refers to.

RGB Amplitudes. The 75%/100%nomenclature specifically refersto the maximum amplitudesreached by the Red, Green, andBlue signals when they form the

six primary and secondary col-ors required for color bars. For75% bars, the maximum ampli-tude of the RGB signals is 75%of the peak white level. For100% bars, the RGB signals canextend up to 100% of peakwhite (see Figures 110 and 111).

Saturation. Both 75% and 100%amplitude color bars are 100%saturated. In the RGB format,colors are saturated if at leastone of the primaries is at zero.Note in Figures 110 and 111 thatthe zero signal level is at setup(7.5 IRE) for NTSC.

68

APPENDIX A - NTSC COLOR BARS

100%

0%7.5

77

100%

0%7.5

77

100%

0%7.5

77

REDSIGNAL

GREENSIGNAL

BLUESIGNAL

100%

0%7.5

100%

0%7.5

100%

0%7.5

REDSIGNAL

GREENSIGNAL

BLUESIGNAL

Figure 110. RGB levels decoded from 75% bars with 75% white.

Figure 111. RGB levels decoded from 100% bars with 100% white.

The Composite Signal. In the com-posite signal, both chrominanceand luminance amplitudes varyaccording to the 75%/100% dis-tinction. However, the ratiobetween chrominance and lumi-nance amplitudes remains con-stant in order to maintain 100%saturation (see Figures 112 and 113).

White Bar Levels. Color bar signalscan also have different white barlevels, typically either 75% or100%. This parameter is com-pletely independent of the75%/100% amplitude distinctionand either white level may beassociated with either type of bars.

Effects of Setup. Because of setup,the 75% signal level for NTSC isat 77 IRE. The maximum availablesignal amplitude is 100 - 7.5, or92.5 IRE. 75% of 92.5 IRE is 69.4 IRE, which when added tothe 7.5 IRE pedestal, yields alevel of approximately 77 IRE.

Note in Figure 110 that the 75%white bar and the 75% RGB signals extend to 77 IRE.

69

Figure 112. 75% bars with 100% white.

Figure 113. 100% bars with 100% white.

Testing Bandlimited Systems. Fastrise time square waves cannot beused for testing bandwidth limitedsystems as attenuation andphase shift of out-of-band com-ponents will cause ringing in theoutput pulse. These out-of-banddistortions can obscure the in-band distortions of interest.Sine-squared pulses are them-selves bandwidth limited, andare thus useful for testing band-width limited television systems.

Description of the Pulse. The sine-s q u a red pulse looks like one cycleof a sine wave (see Figure 114).Mathematically, a sine-squaredpulse is obtained by squaring a half-cycle of a sine wave.P h y s i c a l l y, the pulse is generatedby passing an impulse through asine-squared shaping filter.

T Intervals. Sine-squared pulsesare specified in terms of halfamplitude duration (HAD) whichis the pulse width measured at50% of the pulse amplitude.

Pulses with a HAD that is a mul-tiple of the time interval T areused to test bandwidth limitedsystems. T, 2T and 12.5T pulsesare common examples. T is theNyquist interval, or

12fc

where fc is the cutoff frequencyof the system to be measured.For NTSC, fc is taken to be 4MHz and T is therefore 125nanoseconds.

T Steps. The rise times of transi -tions to a constant luminancelevel (such as a white bar) arealso specified in terms of T. A Tstep has a 10% to 90% rise timeof nominally 125 nanosecondswhile a 2T step has a rise time of nominally 250 nanoseconds(see Figure 115). Mathematically,a T step is obtained by integratinga sine-squared pulse. Physically,it is produced by passing a stepthrough a sine-squared shapingfilter.

Energy Distribution. Sine-squaredpulses possess negligible energyat frequencies above f = 1/HAD.The amplitude of the envelopeof the frequency spectrum at1/(2 HAD) is one-half of theamplitude at zero frequency.Energy distributions for a Tpulse, 2T pulse, and T step are shown in Figure 116.

70

APPENDIX B - SINE-SQUARED PULSES

HAD125 ns

HAD250 ns 2T T

RISE & FALL125 ns nom

90%

10%

90%

10%

Figure 114. 2T pulse and 1T pulse for NTSC systems.

Figure 115. T rise time step.

SQUARE WAVE WITHT STEP RISE / FALL

T PULSE

1

0.5

00 1 2 3 4 5 6 7 8 MHz

FREQUENCY

2T PULSE

Figure 116. Frequency spectra of T pulse, 2T pulse, and T step.

71

APPENDIX C - RS-170A

72

APPENDIX C - RS-170ANTSC Standard

73

APPENDIX D - FCC 73.699, Fig. 6

8/98 FL3636/XBS 25W–7049–3

Copyright © 1997, Tektronix, Inc. All rights reserved. Tektronix products are covered by U.S. and foreign patents, issued and pending. Information in this publication supersedes that in all previously published material. Specification and price change privileges reserved. TEKTRONIX and TEK are regis-tered

For further information, contact Tektronix:World Wide Web: http://www.tek.com; ASEAN Countries (65) 356-3900; Australia & New Zealand 61 (2) 888-7066; Austria, Eastern Europe, & Middle East 43 (1) 70177-261; Belgium 32 (2) 725-96-10;Brazil and South America 55 (11) 3741 8360; Canada 1 (800) 661-5625; Denmark 445 (44) 850700; Finland 358 (9) 4783 400; France & North Africa 33 (1) 69 86 81 08; Germany 49 (221) 94 77-400; Hong Kong (852) 2585-6688; India 91 (80) 2275577; Italy 39 (2) 250861; Japan (Sony/Tektronix Corporation) 81 (3) 3448-4611; Mexico, Central America, & Caribbean 52 (5) 666-6333;The Netherlands 31235695555; Norway 47 (22)070700; People’s Republic of China ( 86) 10-62351230; Republic of Korea 82 (2) 528-5299; Spain & Portugal 34 (1) 372 6000; Sweden 46 (8) 629 6500;Switzerland 41 (41) 7119192; Taiwan 886 (2) 765-6362; United Kingdom & Eir e 44 (1628) 403300; USA 1 (800)426-2200