NTSC Video Measurements The Basics - Educypedia

38
NTSC Video Measurements Copyright © 1997, Tektronix, Inc. All rights reserved.

Transcript of NTSC Video Measurements The Basics - Educypedia

Page 1: NTSC Video Measurements The Basics - Educypedia

NTSC VideoMeasurements

Copyright © 1997, Tektronix, Inc. All rights reserved.

Page 2: NTSC Video Measurements The Basics - Educypedia

About this Booklet

Over the years, Tektronix hasmade available many articles,application notes, booklets,video tapes, and other materialsto help you, our customers, bet-ter understand the interaction ofyour daily activities and Tek’sproducts. This booklet is a com-pilation of several shorter pieceswhich are (or were) offered byTektronix’ Television Division.

In this booklet we’ve arranged asequence of topics which leadsfrom relatively basic conceptstoward the more advanced. Theideas involved, however, dointerrelate and support eachother — you may find it helpfulto take the time to look throughthe whole booklet even if your initial interest is for a particular topic.

Page numbers and figure num-bers used here indicate both thesection of the booklet and thesequence within the section. Forexample, page 2-3 is the thirdpage of the second section;Fig. 1-5 is the fifth figure in thefirst section.

As always, Tektronix is veryinterested in your feedback andcomment and may be contactedat any of the offices listed in theback of this booklet.

Credits:

Portions of this document have been reprinted with permission as chapters of this book:

• Government and Military Video, P.S.N. Publications, Inc., New York, NY, 1992

• Video Systems, Intertec Publishing Corp., Overland Park, KS, 1992

Page 3: NTSC Video Measurements The Basics - Educypedia

1. General ConceptsWhy Test Video Signals......................................................................................................................................................... 1-1Test Methodology.................................................................................................................................................................. 1-2Connecting and Terminating Instruments.............................................................................................................................. 1-3

2. Basic Video Testing — Waveform Monitor TechniquesConnecting the Instruments................................................................................................................................................... 2-1Understanding the Waveform Display.................................................................................................................................... 2-2Understanding the 1710J Controls........................................................................................................................................ 2-3Making Measurements.......................................................................................................................................................... 2-4

3. Basic Video Testing — Vectorscope TechniquesUnderstanding the Display..................................................................................................................................................... 3-1Using the Vectorscope Controls............................................................................................................................................ 3-2Checking the Display............................................................................................................................................................. 3-3Checking Chrominance Phase............................................................................................................................................... 3-3Checking Chrominance Gain................................................................................................................................................. 3-4Checking White Balance........................................................................................................................................................ 3-4

4. Setting up a Genlocked StudioGenlock Basics...................................................................................................................................................................... 4-1Adjusting TBCs...................................................................................................................................................................... 4-2Adjusting for Synchronization................................................................................................................................................ 4-2

5. Intermediate NTSC Video TestingSetting Up the Test Equipment............................................................................................................................................... 5-2Insertion Gain........................................................................................................................................................................ 5-2Frequency Response Testing................................................................................................................................................. 5-4Chrominance to Luminance Gain.......................................................................................................................................... 5-6Nonlinear Distortions............................................................................................................................................................. 5-7

Luminance Nonlinearity................................................................................................................................................ 5-7Differential Gain............................................................................................................................................................ 5-8Differential Phase..........................................................................................................................................................5-10

6. Beyond Bars — Other Tests and Test Signals .................................................................................................................... 6-1

7. The Color Bar Signal — Why and How ................................................................................................................................ 7-1

8. Glossary ..................................................................................................................................................................................... 8-1

Table of Contents

Page 4: NTSC Video Measurements The Basics - Educypedia

General Concepts

Why Test Video Signals?Working in video, the quality ofyour final product depends onmany factors. Scripting, casting,directing, and many other ingre-dients add together to make(or break) a production. But evenwhen all the creative efforts areperfectly balanced, poor techni-cal quality — and the resultingpoor picture quality — canobscure all the hard work andskill invested.

This holds true in any facilityfrom the largest productionhouse or network studio, to corporate studios, and to thesmallest freelance productionshop. The diff e rence is, the larg e rfacilities have a technical staf fdevoted to keeping equipmentperformance and picture qualityin peak form. Smaller commer-cial, business, and industrialfacilities have smaller staffs.Sometimes a staff of one!

Whether you’re a video veteranor the newest intern, it behoovesyou to know something moreabout video quality than whatyou see on a picture monitor.The goal of this booklet is tohelp you acquire additionaltechnical knowledge, and bemore comfortable and capablewith video testing.

You must, however, exercisecaution and common sense. Be sure to heed all warnings printed on equipment covers.And never remove any equip-ment casings or panels unlessyou are qualified for internalservicing of that equipment. Theinformation in this booklet isintended only for use in extern a lmonitoring and adjustment ofuser accessible controls onvideo equipment.

A television picture is conveyedby an electrical signal (Figure 1-1 ) .This video signal is carried fromone place to another by cables(coax) or by radio-frequency (RF) waves. Along the way, itmust pass through variouspieces of equipment such asvideo tape machines, switchers,character generators, specialeffects generators, and transmit-ters. Any of this equipment canchange or distort the signal inundesirable ways.

Since picture quality is largelydetermined by signal quality, it’simportant to detect and correctany signal distortions. The signalhas to be right before the picturecan be right.

Many video facilities rely heavilyon picture monitors for qualitychecks at various stages of pro-duction and distribution. A p i c t u re monitor, after all, displaysthe final product, the picture. It’s a quick and easy means ofensuring that there are no obvious picture impairments.

But picture monitors do not tellthe whole story. In fact, relyingsolely on picture monitors forvideo quality checks can be anopen invitation to disaster.

First of all, not every pictureimpairment is obvious on a picture monitor. Minor problemsare easily overlooked. Some cannot be seen at all. For exam-ple, a marginal video signal can still produce a seemingly“good” picture on a forgivingmonitor. This can produce afalse sense of security as signaldegradations accumulate thro u g hvarious production stages. Theend result can be a nasty surprisethat can lead to costly remakesand missed deadlines.

1-1

Figure 1-1. A television picture is created by an electrical video signal. In this case, the picture of color bars is created by a color bars test signal.

To avoid such surprises, you needto look at more than pictures.You need to look at the videosignals that convey the pictureinformation through the variousdevices and interconnectingcables of a video system.Specialized instruments havebeen developed to process anddisplay these signals for analysis.

Figure 1-2. Waveform display of a color bars signal.

Page 5: NTSC Video Measurements The Basics - Educypedia

ensure the overall gain of arecord/playback system is cor-rect, i.e., a recording made witha standard amplitude video signalat the machine’s input will pro-duce standard amplitude videoat its output during playback.

In the first case (specific signal)we’ll need test equipment thatenables observation of the signaland knowledge of what theappropriate limits or characteris-tics are. The second case (equip-ment or system testing) is moregeneral and is assumed in thefollowing discussion.

The usual method of evaluatingvideo equipment is with a well-defined, highly stable test signalhaving known characteristics,such as a color bars signal.

Video testing is based on thissimple principle of applying aknown test signal to the videosystem or equipment input andobserving the signal at the out-put. Any distortion or impair-ment caused by the system isobserved and measured on theoutput signal. If there are distor-tions, the equipment is adjustedto eliminate or minimize them.The point is, if the system canpass the test signal from input tooutput with little or no distor-tion, it can cleanly pass picturesignals as well.

The signals necessary for suchtesting are obtained from a testsignal generator. This instrumentproduces a set of precise videosignals with carefully definedand controlled characteristics.Each signal is ideal for verifyingone or more specific attributes

of the video system under test.For all practical purposes, thesetest signals are “perfect” signals.

Other instruments such as wave-form monitors, vectorscopes,combinations of waveform andvector monitors, or specializedvideo measurement sets are usedto evaluate the test signal at theoutput of the path under test. Asan example, Figure 1-2 shows awaveform monitor display of acolor bars signal. This display isalso called a waveform — it isactually a graph of the changingvoltage of the signal (plotted vertically) and time (plotted horizontally). Calibrated scaleson the waveform monitor’sscreen allow the various ampli -tude (voltage) and time parame-ters of the waveform to be mea-sured. Other test signals andtheir related instrumentationand displays are discussed inthe following sections.

Test signal generators and signalevaluation instruments areavailable in a wide variety ofmodels. These can range fromsimple production-orientedinstruments to highly sophisti-cated engineering instruments.Wa v e f o rm monitors, vectorscopes,video test sets, and other spe-cialized equipment to displayand/or evaluate the signal arealso available in a wide varietyof configurations. The followingsections will acquaint you withsome of these tools and withmethods of using them toenhance your effectiveness inmaintaining video quality.

1-2

A waveform monitor is aninstrument used to measureluminance or picture brightnessas well as a high frequency colorsignal called chrominance. Aninstrument called a vectorscopeis required for quality control ofvideo chrominance, especially in more complex systems.

When properly used, these toolsallow you to spot signal pro b l e m sbefore they become pictureproblems. You can make certainthat the signals are optimal,rather than marginal. Plus, withregular video system testing, youcan catch minor degradationsand performance trends thatindicate a need for system main-tenance. This allows you to avoidcostly failures, both in equipmentand in program production ordistribution.

Test MethodologyT h e re are two somewhat diff e r i n gideas used in testing video. Oura p p roach is to test characteristicsof a specific signal to ensure thatit meets certain artistic or tech-nical requirements. Or, we maytest the characteristics of indi-vidual pieces of equipment orseveral pieces in the video signalpath to determine that signaldistortions introduced by thisequipment are acceptable — or,at least, minimized.

An example of the first casemight be monitoring the outputof a video source (camera, char-acter generator, etc.) to ensure itis not producing signals thatexceed black or peak white signallimits. As an example of the second case, we may wish to

Page 6: NTSC Video Measurements The Basics - Educypedia

Connecting and TerminatingInstrumentsThe Tektronix instruments dis-cussed in this booklet, and mostothers, have rear panel BNC connectors. For many videotests, you only need to use oneconnector on each instrument.This is the connector markedTEST SIGNAL on the signal generator and the connectormarked CH A INPUT on thew a v e f o rm monitor or vectorscope.

Notice there are actually two“loop-through” CH A connectorson both the waveform monitorand vectorscope (Figure 1-1). Ifyou feed the signal in one sideof these inputs and out the other,the signal will pass through theinstrument unaffected. This typeof loop-through input lets youconnect the same signal to morethan one instrument.

For example, after running thetest signal from the signal gener-ator through the system undertest and from there to a waveformmonitor, you can then loop thesignal through the waveformmonitor into a vectorscope. Usingthe same method, you can alsoloop through the vectorscope toa picture monitor. This connec-tion method allows you to lookat the same test signal on allthree instrument displays(Figure 1-3). The order in whichthe instruments are connectedd o e s n ’t matter — if the connectingcables are short.

Coaxial cable does have signalloss — the signal’s amplitudedecreases as it progresses downthe cable. For runs of a few feet,this decrease in amplitude istypically 1% or less and is usuallyi g n o red. For longer cable lengths,or when precision measurementsare being made, the loss must betaken into account or correctedwith a compensating distributionamplifier. Not only is coax lossy,the loss is a function of signalfrequency — higher video frequencies are attenuated morethan the low frequencies.

Also, not all cables are wellshielded and signals may cross-talk from outside the cable intothe video path inside. In otherwords, the interconnectingcables themselves may be introducing signal distortions.

While small, flexible cables areconvenient to handle, they arenot without technical cost.Consideration should be given tousing larger, double-shieldedcables in long or critical runs.The techniques in this bookletcan be used to evaluate the dis-tortions introduced by variouslengths and/or quality of cable.

While on the subject of cablesand interconnections, alwaysremember to properly terminateeach signal path. If a signal pathis left “open” at the end — suchas a high impedance loop-thro u g hwith nothing connected on one side — several problems can result.

Most obvious will be a change inamplitude of the signal on thatpath. With an open terminationthe amplitude will be higherthan expected (usually abouttwice amplitude, but actuallydepending on the signal sourceimpedance). With a double ter-mination, such as will result ifan internally terminated loop-through and an external termina-tor are used on the same path,the amplitude will be decreased.A double termination will oftenresult in a two-thirds amplitudesignal, again depending onsource impedances.

Amplitude is not the only effectof misterminations. Don’t betempted to make up for impro p e rtermination by adjusting the signal amplitude. Misterm i n a t i o n salso introduce problems withfrequency response (amplitudebecomes a function of signal frequency) and with differingresponse depending on locationalong the signal path. Use thecorrect terminating resistor onthe end of each coaxial path.

1-3

F i g u re 1-3. Loop-through inputs on instruments allow the same video sourc eto be viewed simultaneously in waveform, vector and picture forms.

There are three ways a signalpath can be properly terminated:

• Some instruments have a single-connector input with abuilt-in terminator. You canconnect such an instrument“as is” at the end of a signalpath, but you can’t loop thesignal through it.

• Some instruments have a two-connector input with a built-interminator and a switch forselecting loop-through or terminating connection. Youcan connect such an instr u-ment anywhere in the signalpath, but you must set theswitch correctly for theintended use (either Hi-Z forloop-through or 75 ohm forend-of-line termination).

• Some instruments, such as the 1720 and 1730, have two-connector inputs with nobuilt-in terminators. You canconnect such an instrumentanywhere in the signal path.But, if you connect it at theend of the signal path, you mustattach a 75 ohm terminator tothe unused connector.

Page 7: NTSC Video Measurements The Basics - Educypedia

The focus of this section is ontests in production facilities usinga waveform monitor. It will behelpful if you have already readthe first section of this booklet,General Concepts. The followingassumes you are already familiarwith that material.

To provide examples of tests in aproduction environment, theTektronix TSG 100 NTSCTelevision Generator (Figure 2-1)and Tektronix 1710J NTSCWaveform Monitor (Figure 2-2)are used since they are modelsoften found in the productionenvironment. However, all waveform monitors and test signal generators, no matter howsophisticated, include the samebasic functions described here.A vectorscope (Figure 2-3) isanother useful instrument forvideo signal evaluation; its useis covered in the next section ofthis booklet.

Vectorscopes and waveformmonitors complement oneanother and provide a full repre-sentation of all informationabout the video signal. Becauseof this, both instruments areoften mounted side-by-side ininstrument racks. In some cases,both functions may even appearin the same instrument. The restof this section shows you how touse test signal generators andwaveform monitors for basicvideo testing. For further detailson these instruments and theiradditional uses, you may wantto refer to: Basic WaveformMonitoring Vi d e o t a p e ( 2 4 W- 7 0 2 9 ) .This tape is available throughyour local Tektronix Sales Officeor by calling 1-800-TEK-WIDE,extension TV.

Connecting the InstrumentsOnce you’ve identified theinstruments to be used, the nextstep is to connect them to thevideo system to be tested. Oftenthis has already been done foryou, especially when the instru-ments a re mounted in equipmentracks. Still, it pays to know howthe connections should be madeso you can make sure they havebeen done correctly.

The Tektronix instruments dis-cussed here have rear-panel BNCconnectors. These connectors areused for connecting cables to thesystem under test. For basicvideo testing, you only need touse one connector on eachinstrument. Refer to Connectingand Terminating Instruments onpage 1-3 for further information.

Ensure everything is connectedas shown in Figure 2-4 (you maynot be using a vectorscope orpicture monitor — just be sure toterminate the end of the signalpath). With all equipment turnedon, a test signal also needs to beselected. For the examples inthis section, a color bars signal is the only test signal that you’ll need.

The TSG 100 has a series of signal selection push buttons onits front panel (Figure 2-1). Eachpush button is labeled to indicatethe type of signal selected bythat button. Select the color bars signal.

2-1

Basic Video Testing —Waveform MonitorTechniques

Figure 2-1. The Tektronix TSG 100 NTSC Television Generator produces avariety of special signals for testing video equipment and systems. (NTSCstands for National Television Systems Committee, the committee thatdeveloped the television system we use in the United States.)

Figure 2-2. The Tektronix 1710J NTSC Waveform Monitor displays videosignals for amplitude and timing measurements.

Figure 2-3. The Tektronix 1720 Vectorscope complements a waveformmonitor by providing additional information about the video signal’s color(chrominance) content.

Figure 2-4. Loop-through inputs on instruments allow the same video signalto be viewed simultaneously in waveform, vector and picture forms.

Page 8: NTSC Video Measurements The Basics - Educypedia

If you are using a picture moni-tor, you should see the familiarcolor bars display on that moni-tor. You should also see a colorbars signal on the waveformmonitor display. This waveformis shown in Figure 2-5. This colorbars signal is called SMPTE colorbars because it conforms to thesignal specifications set by theSociety of Motion Picture andTelevision Engineers (SMPTE).

Understanding the WaveformDisplayNotice in Figure 2-6 that there arevarious lines and numbers onthe front of the waveform moni-tor’s CRT. These markings arereferred to as the graticule andallow measurement of waveformamplitude and time parameters.

The nominal video signal levelfor television studios and pro-duction facilities is one volt(1 V) peak-to-peak. (The termpeak-to-peak means from thebottom of the signal to the top. Itis often abbreviated as p-p.)Without such an amplitude standard, signals from differentsources might not be compatiblewith each other. Keep in mindthat this is for a nominal signal,one that contains the brightestpossible (peak white) pictureinformation. Many test signals fitthis description. Actual pictureinformation often does not con-tain any peak white levels andwill therefore not measure 1 Vpeak-to-peak.

For making peak-to-peak andother amplitude measurements,the vertical axis of the waveformmonitor graticule is marked inIRE units. An IRE unit (named forthe Institute of Radio Engineers)is a relative unit of measureequaling 1/140th of the peak-to-peak (p-p) video amplitude. Sincea video signal should be 1 V p-p,an IRE unit is about 0.00714 V,or 7.14 mV, in this case.

Notice also that there is a hori-zontal scale on the waveformmonitor graticule at 0 IRE. Thescale is divided by tick marks

into major and minor divisions.These divisions are used formeasuring time intervals on thevideo signal. The value of eachdivision — the time per division— is determined by the wave-form monitor’s horizontal control settings.

To interpret the display inFigure 2-5, you should be awarethat an NTSC television pictureconsists of 525 horizontal lines.These lines are formed by scan-ning from left to right in a rasterp a t t e rn. This pattern is illustratedin Figure 2-7.

Every line of a video signal con-tains a negative-going synchro-nizing pulse. This ‘sync’ pulsestarts the next line of video andis contained in the signal’s hori-zontal blanking interval. Thisblanking interval is a short periodof time between the scan lines.

The 1710J displays what appearsto be two lines of video side byside in what is generally called a2H mode (two horizontal lines).This is actually a display of all525 lines laid on top of eachother, with half of the lines onthe left and half of the lines onthe right. It also puts a completehorizontal blanking interval nearthe center of the screen. This isshown in Figure 2-5. Somewaveform monitors use a 1Hmode which shows one line ofvideo rather than two lines sideby side. This would be like looking at just the left half of thedisplay in Figure 2-5 expandedhorizontally to fill the screen.

A Closer Look at the Color Bars Waveform

The drawing in Figure 2-6 showsone complete line of a SMPTEcolor bars waveform. This wave-form consists primarily of abrightness signal (called lumi-nance) and a high-frequencycolor signal (called chro m i n a n c e ) .The luminance and chrominanceare added together to form theoverall waveform.

2-2

Figure 2-5. Waveform monitor display of SMPTE color bars waveformshows video signal voltage variations with time.

F i g u re 2-6. One line of a SMPTE color bars signal drawn to show key details.

Figure 2-7. A raster pattern is followed in forming a 525-line television picture. The picture lines are scanned from left to right in a sequence fromtop to bottom.

Page 9: NTSC Video Measurements The Basics - Educypedia

The luminance signal is a seriesof voltages or “levels” that deter-mine brightness variationsacross the picture. Each of thecolors in the color bars signalhas a different luminance level,and the bars are arranged bylevel from highest to lowest(white, yellow, cyan, green,magenta, red, blue, and black).

The chrominance signal is a sinewave. Because of this signal’shigh frequency, the sine wavecycles appear to run together inmost displays. Individual cyclescan be seen, however, when thedisplay is expanded horizontallyas in Figure 2-9.

Notice in Figure 2-6 that thepeak-to-peak amplitude of thechrominance signal varies fromone color bar to the next. Thefirst bar, being white, has nochrominance. All of the otherbars have the correct amount ofchrominance amplitude to produce a full intensity (100%saturated) color. The last bar(Black) also has no chrominance.

Color bar test signals fall into twogeneral categories: 100% bars(full amplitude) and 75% bars(reduced amplitude). You shouldalways use 75% bars for basictesting. This is because 100%bars contain signal levels thatmay be too high to pass througha system without distortion.

At some point, however, you willneed a 100% white referencelevel for checking overall signalamplitude. Many 75% color bartest signals do provide a 100%white level for this purpose(Figure 2-8).

The SMPTE color bars signalshown in Figures 2-5 and 2-6 are75% bars with a 75% whitelevel. This signal also incorpo-rates a 100% white level onsome lines. This 100% whitelevel is easiest to see on a wave-form monitor used in either thelow-pass or dual-filter mode(Figure 2-10).

All of the color bars test signalsillustrated so far were fed dire c t l yf rom the test signal generator intoan accurately calibrated wave-form monitor. This means theyare undistorted, their amp l i t u d e sand timing are correct, and theyrepresent the ideal condition.

But suppose the signal had beenpassed through other equipmentfirst, an amplifier or recorder forexample. Since you know whatthe test signal is supposed tolook like, you’ll be able to quickly tell if the signal has been distorted or changed by theequipment — and by how much.That’s why it’s so important touse test signals with knowncharacteristics when evaluatingvideo systems.

Understanding the 1710JControlsBefore attempting to make anymeasurements, you need toknow how to use the 1710JWaveform Monitor controls. It’salso important to realize thatwhen you adjust these controls,you’re only changing the waythe waveform looks on the moni-tor’s display. The waveformmonitor controls do not in anyway affect the video signalitself. The same is true of picturemonitors and vectorscopes. Iftesting reveals something wrongwith the video signal, you mustuse the controls on the equipmentunder test to correct the pro b l e m .

Referring back to Figure 2-2,notice that the 1710J front-panelcontrols are grouped into func-tional blocks. The following discussion describes those c o n t rols block by block. Althoughthis discussion is specific to the1710J, much of it also applies ingeneral to all waveform monitors.

2-3

Figure 2-8. This 75% color bars signal includes a 100% white referencelevel. An eighth black bar is also included.

Figure 2-9. When a chrominance signal is expanded horizontally, individualcycles of the sine wave can be seen.

Figure 2-10. The 1710J dual-filter display is convenient for making luminance and chrominance measurements from the same display. Theluminance display (L-PASS) is shown on the left, and luminance pluschrominance (FLAT) is on the right.

Page 10: NTSC Video Measurements The Basics - Educypedia

INPUT block. The INPUT con-trols select the input signal to bedisplayed, which parts of it aredisplayed, and how it’s synchro-nized. These controls includethe FILTER, REFerence, andINPUT buttons.

FILTER Button —

F L AT displays the entire signal,such as shown in Figure 2-5.

L PASS removes the high-frequency chrominance signalfrom the display, leaving just the luminance signal for observation.

C H R M displays only thec h rominance inform a t i o n .

Because you need luminanceplus chrominance for someamplitude measurements, butonly luminance for others, boththe FLAT and L PASS displaysare useful. Another option is touse the 1710J dual-filter display.This shows both waveforms atonce with L PASS on one side andFLAT on the other (Figure 2-10).

To obtain a dual-filter display,simply press and hold the1710J’s FILTER button. This display allows you to checkluminance and chrominance levels without switching back andforth between FLAT and L PASS.

REF Button —

INT synchronizes the wave-form monitor to the displayedwaveform. For the testsdescribed in this applicationnote, always use INT (i n t e rn a l) .

E X T s y n c h ronizes the waveformmonitor to an external source.

CAL displays a built-in cali-bration signal for adjusting thewaveform monitor’s gain. Thiswill be discussed later in more detail.

INPUT Button —

A displays any signal connectedto the CH A input BNC connectors on the back panel.

B displays any signal connectedto the CH B input BNC connectors on the back panel.

B O T H is a press and held buttonfor a display of both A and Binputs simultaneously.

You need to select A or B,depending on which set of connectors is being used for connection to the equipmentunder test (Figure 2-4).

V E RTICAL block. The VERT I C A Lcontrols adjust the size and position of the displayed wave-form. (Remember, changing thedisplay doesn’t change the videosignal being looped through the monitor.)

GAIN Button —

G A I N magnifies the vertical sizeof the waveform display x5.

POSITION Knob —

POSITION moves the dis-played waveform up anddown on the screen.

HORIZONTAL block. The HORIZONTAL controls adjustthe waveform’s horizontal sizeand position on the display.

SWEEP Button —

SWEEP selects whether themonitor displays two lines(2H) or two fields (2FLD) ofvideo. Press and hold for aone line display.

MAG Button —

MAG expands the display horizontally. When used witha 2H sweep, MAG provides acalibrated sweep speed of onem i c rosecond per major divisionon the horizontal graticulescale. This setting is used formost timing measurements.

POSITION Knob —

P O S I T I O N moves the displayedwaveform left and right on the screen.

DISPLAY block. The DISPLAYcontrols adjust the display forcomfortable viewing.

FOCUS changes the focus ofthe trace.

SCALE varies the graticuleillumination.

INTENS controls waveformtrace brightness (intensity).

Making MeasurementsNow that you understand thebasic concepts of testing, testsignals, and the instrumentsused for measurements, you areready to make measurements.

First, feed the color bars test signal into the system you wantto test. This signal can comestraight from the test signal generator or it can be a colorbars signal previously recordedon video tape.

Second, connect the system output to the waveform monitorCH A input. Remember to prop-erly terminate loop-throughs.Then, set the waveform monitorcontrols as follows:

POWER ON

INPUTFILTER FLATREF INTINPUT A

VERTICAL GAIN x1 POSITION Blanking level

aligned with thehorizontal axisas shown inFigure 2-5

HORIZONTAL SWEEP 2HMAG 1µs POSITION Sync pulse

approximatelycentered asshown inFigure 2-5

DISPLAY Adjust all con-trols for comfort-able viewing

With all controls set as describedabove, you should see the samedisplay as shown in Figure 2-5or Figure 2-8.

2-4

Page 11: NTSC Video Measurements The Basics - Educypedia

Checking luminance levels

After verifying that the waveformmonitor is calibrated, you canbegin making video signal levelmeasurements with confidence.

First, check the video system’soverall gain. This is done bychecking the amplitude of thecolor bars signal coming fromthe system. If the amplitude istoo high (Figure 2-11), the picture is said to be “hot.” It’slike an overexposed photograph.On the other hand, an amplitudethat is too low is like an under-exposed photo and results in adark picture.

All amplitude measurements aremade relative to the horizontalaxis at 0 IRE. To do this, makesure the blanking level (labeledin Figure 2-6) is aligned with the0 IRE graticule line. Then makesure that the 100% white refer-ence level coincides with the100 IRE graticule line. Also,make sure the black level (sometimes called setup) is at7.5 IRE. Notice there is a specialdashed graticule line at 7.5 IREfor verifying black level. (Be sureyou do not confuse the blacklevel with the blanking level.)

If the video signal’s 100% whitelevel does not correspond to100 IRE, use the gain control onthe system under test to adjustthe top of the signal to 100 IRE.This should also bring the blacklevel to 7.5 IRE. However, theremay be a separate black levelcontrol on the equipment undertest that can be used for finaladjustment of black level. Ifthere is a separate black levelc o n t rol, adjust it first, then adjustthe gain control for 100 IRE.Making adjustments in this ordershould minimize interactionbetween them.

Next, check the sync pulse level.The sync pulse is the narrowpulse in mid-screen in Figure 2-8extending down from the blanking(0 IRE) level. The flat bottom of the sync pulse should be at -40 IRE. If it’s more than a fewIRE off, see if the equipmentunder test has a sync amplitudeadjustment. If it does, adjust thesync amplitude to -40 IRE. Theluminance level and sync levelmust both be correct, i.e., the0 IRE reference level luminancemust extend to 100 IRE at peakwhite and sync must extend to -40 IRE. If the signal cannot beadjusted to match the graticuleat all three points (white at100 IRE, blanking at 0 IRE, and sync at -40 IRE) the equip-ment requires servicing by aqualified technician.

A word of caution

If there is more than one piece ofequipment with gain controls inthe path you’re testing, you mustmake sure that each unit isadjusted for standard input andoutput levels. This is done bymoving the test equipment con-nection to the output of the firstunit, then to the second, etc.,and adjusting each unit in turn.Failure to do this may allow a system to have higher- or lower-than-standard levels atsome points in the path, whichwill almost always introduceincreased distortions. To put thisanother way, the effects of non-standard levels at some point inthe system cannot be fully corrected “downstream.”

2-5

Verifying monitor calibration

B e f o re making any measure m e n t s ,it is wise to make sure the wave-form monitor is properly cali -brated. This is an easy step totake with the 1710J’s built-in calibration signal.

To view the calibration signal,press and hold the REF buttonlocated in the INPUT block.When the calibration signal isselected, the front panel CALLED will be illuminated. Thecolor bars signal display willdisappear and the calibrationsignal — a 1 V p-p square wave— will appear on the screen.

To check calibration, positionthe calibration square wave vertically so that its flat bottomsalign with the -40 IRE graticuleline. The flat tops of the signalshould then line up with the100 IRE graticule line. If this is not the case, the waveformmonitor is out of calibration and should be checked by aqualified technician.

Figure 2-11. An example of a “hot” signal — notice how parts of the signalexceed 100 IRE.

Page 12: NTSC Video Measurements The Basics - Educypedia

Checking chrominance gain

After verifying correct luminancegain, you need to check chromi-nance gain. If the chrominanceamplitudes are wrong with re s p e c tto luminance, the picture’scolor intensity (saturation) is affected.

First, make sure the color burstranges from -20 to +20 IRE. Thencheck the maximum chro m i n a n c elevels of the first two color bars(yellow and cyan). Both shouldbe at exactly 100 IRE. If they arenot, adjust the chrominance gainon the equipment under test tocorrect the chrominance levels.

Checking sync width

Sync width should be checkedafter verifying correct luminanceand chrominance gain. Syncwidth is not as critical as signalamplitude, but it still should be checked to make sure the equipment is operating withindefined limits.

To accurately measure sync pulsewidth, you need to magnify thepulse on the waveform monitorscreen. To do this, press theMAG button. If the sync pulse isat center screen when MAG ispressed, the magnified syncpulse should still appear nearcenter screen (Figure 2-12). If itdoesn’t, adjust the waveformmonitor’s horizontal positioncontrol to bring the pulse to center screen.

Next, adjust the vertical positioncontrol to position the pulse’stop (blanking level) at +20 IREas shown in Figure 2-12. Thebottom of the pulse should nowbe at -20 IRE. Now, the 50%level of the pulse corresponds to 0 IRE.

With the pulse displayed in thism a n n e r, you can read pulse widthd i rectly from the horizontal 0 IREaxis. In 2H MAG, each majordivision on the axis equals 1 µs.Using this scaling to measurepulse width at the 50% level,the sync pulse width should be4.7 µs. If the width isn’t close to4.7 µs and there’s no sync widthadjustment on the equipmentunder test, you should have theequipment serviced.

Checking sync pulse width completes the series of basicmeasurements for objectivelyverifying video signal quality.Again, the verification relies onapplying a known ideal test signal to the input of the equip-ment under test. The signalshould pass through the equip-ment without any changes inluminance and chrominanceamplitudes or sync pulse width.If there are changes in any of theparameters, the equipment eitherneeds adjustment or servicing.

2-6

The next step

The instruments and techniquespresented in this section cansolve or avoid many pictureproblems often encountered invideo facilities. Being able toapply these skills enhances yourconfidence and professional value.

You should, however, be awarethat there are many other, morespecialized, tests and measure-ments that can be made on videosystems. The various test signalselection buttons on the TSG100 (Figure 2-1), other than color bars, offer only a hint ofthe range of other tests and measurements possible.

F i g u re 2-12. Sync pulse display horizontally magnified (MAG) and positionedvertically so the pulse’s 50% levels are on the 0 IRE horizontal axis.

Page 13: NTSC Video Measurements The Basics - Educypedia

Basic Video Testing —Vectorscope Techniques

In the section on waveform mon-itor techniques, it was explainedthat a waveform monitor displaysa graph of the video signal —amplitude (voltage) on the vert i c a laxis and time on the horizontalaxis. A vectorscope also graphsportions of the video signal fortest and measurement purposes,but the vectorscope display differs from a waveform display.

While a waveform monitor displays amplitude informationabout all parts of the signal, avectorscope displays inform a t i o nabout only the chrominance(coloring) portion of the videosignal — it does not respond toother parts of the video signal.

There are two important parame-ters of the chrominance signalthat may suffer distortions leadingto noticeable picture problems.These are amplitude (gain), andphase (timing). Amplitude is anindependent measurement andcan actually be made with awaveform monitor. Phase is therelationship between two signals— in this case, the relationshipbetween the chrominance signaland the reference burst on thevideo. The processing within avectorscope and the display ofthe processed signals is designedto readily detect and evaluateboth phase and gain distortionsof the chrominance.

Understanding the Display There are two parts to a vec-torscope display — the graticuleand the trace. The graticule is ascale that is used to quantify theparameters of the signal underexamination. Graticules may bescreened either onto the face-plate of the CRT itself (internalgraticule) or onto a piece of glassor plastic that fits in front of theCRT (external graticule). Theycan also be electronically gener-ated. The trace represents thevideo signal itself and is elec-tronically generated by thedemodulated chrominance signals (Figure 3-1).

All vectorscope graticules aredesigned to work with a colorbars signal. Remember, the colorbars signal consists of brightnessinformation (luminance) andhigh-frequency color information(chrominance or chroma).

Each bar of the color bars signalc reates a dot on the vectorscope’sdisplay. The position of thesedots relative to the boxes, or targets, on the graticule and thephase of the burst vector are themajor indicators of the chromi-nance (color) signal’s health.(Burst is a reference packet ofsubcarrier sine waves that is senton every line of video.)

3-1

Figure 3-1a. The vectorscope screen shows a display of a SMPTE color barstest signal with all of the dots in their boxes and the color burst correctlyplaced on the horizontal axis.

Figure 3-1b. The vector dots are rotated with respect to their boxes, indicating a chroma phase error.

Page 14: NTSC Video Measurements The Basics - Educypedia

3-2

The graticule

The graticule is usually a fullcircle, with markings in 2- and10-degree increments. The crosspoint in the center is the re f e re n c emark for centering the trace.Within the circle you also seesix target shapes each containingsmaller, sectioned shapes. Thesmaller shapes are where eachdot of the color bars signalshould fall if the chroma gainand phase relationships arecorrect. (In practice a dot thatwill fall entirely within thesmaller target is only created bya very low noise signal, such asdirectly from a color bars genera-tor. Expect the dots to be muchlarger and “fuzzier” on signalsfrom tape players or off-airreceivers. Note also the camerasignal in Figure 3-2.)

The horizontal line bisecting thecircle at zero and 180 degrees(9 o’clock to 3 o’clock positions)is used as a re f e rence for corre c t l ypositioning the burst display.(Burst is indicated by the port i o nof the trace that lies from thecenter part way toward the zerodegree position.) The markingstoward the left of the burst display indicate correct burstamplitude for 75% or 100%color bars. Note that burst doesnot change electrical amplitude— the gain in the vectorscope’sprocessing is increased when itis set to display 75% amplitudebars in the targets — and thati n c reased gain causes the displayof the fixed amplitude burst signal to be longer.

At the left edge and near theouter circumference of thegraticule is a grid used to measure differential gain andphase. These measurements arediscussed in the later section onIntermediate Video Testing.

Some vectorscopes also havesquare boxes outside the gratic-ule’s circle in the lower-left

and upper-right corners of the screen. These boxes, alongwith the vertical line that bisectsthe circle, are used for makingstereo audio gain and phasemeasurements with a Lissajousdisplay. While this capability isuseful in certain applications,we will not discuss it in thisbooklet, but rather concentrateon video measurements.

The trace

The vectorscope’s display is created by decoding the chromi-nance portion of the video signalinto its two components, B-Y(blue minus luminance) and R-Y(red minus luminance). Thesetwo signals are then plottedagainst each other in an X-Yfashion, with B-Y on the hori-zontal axis and R-Y on the vertical axis. While waveformmonitors lock to sync pulses inthe video signal, vectorscopeslock to the 3.58 MHz color burstand use it as the phase referencefor the display.

Three color bars signals are com-monly used with vectorscopes— 75% and 100% full fieldcolor bars and the SMPTE colorbars. The 100% and 75% labelsrefer to the amplitude of the signal, not the saturation of thecolors; both are 100% saturated.In this application note, we willuse the SMPTE color bars signalexclusively, because it is anindustry standard. SMPTE barsare 75% amplitude color bars.

You should always use 75% barsfor basic testing. This is because100% bars contain signal levelstoo high to pass through sometypes of equipment undistorted,even when the equipment isoperating properly. For moreinformation on the various colorbars signals, refer to pages 2-2and 2-3 in the Basic VideoTesting — Waveform MonitorTechniques section.

Using the VectorscopeControls The first rule-of-thumb toremember when you begin tooperate the vectorscope is thatthe controls do not in any wayaffect the video signal itself. Theonly way to adjust the signal,rather than just its vectorscopedisplay, is with the controls ofthe equipment providing ortransferring the video signal.

All the basic controls for poweringup the instrument and adjustingthe display for comfortable viewing (power and display:focus, scale illumination, andintensity) are self-explanatory.The gain controls provide a wayto calibrate the display’s gain foreither 75% or 100% color bars.A variable control expands thetrace to compensate for low signal levels, or so you can analyze the signal in greaterdetail. For white balancing cameras, use of this control alsoallows more visibility of thedetail in the white and blackinformation near the center ofthe vectorscope display.

Figure 3-2. Proper white balance of a video camera is indicated by a fuzzyspot centered on the vectorscope display when the camera is viewing awhite object.

Page 15: NTSC Video Measurements The Basics - Educypedia

3-3

The input controls typically consist of the input channel, reference, and mode selections.The input for channel A, forexample, is likely to be videofrom a switcher’s output (with aVTR, camera, signal generator orother equipment providing theinput to the switcher). Referencemay be internal, where the vec-torscope locks to the color burstof the selected input channel, ore x t e rnal, where lock is to a secondsignal. In this section we areonly using internal reference.Externally referencing a vec-torscope is imperative for thepurpose of matching the chromaphase of various sources toeliminate color shifts at editpoints. This application is discussed in the Setting Up aGenlocked Studio section.

Internally referenced displaysonly show the phase re l a t i o n s h i pbetween the burst and color barssignal of the selected input signalitself — not phase relationshipsbetween different signals.

If the vectorscope has an extern a lX,Y input for displaying audio, a mode control allows you toenable this input for displaywith or without the regular vector display.

The phase control is one youwill use often. In the internallyreferenced mode it must be usedto align the burst vector on thehorizontal axis (pointing towardzero degrees, or 9 o’clock). If thecolor burst is not in this correctposition, the signal is not prop-erly aligned with the graticuleand provides no really usefulinformation. As you rotate thephase control, the whole tracerotates about the center point.

Checking the DisplayWhen you first check the displayof the color bars signal on a vectorscope, you should see:

1.A bright dot in the center ofthe display.

2.Color burst straight out on thehorizontal axis extending tothe 75% mark.

There are also a few things youshould examine on the waveformmonitor. The waveform monitorshould indicate the following:

1. Waveform black level at7.5 IRE. (Black level adjust-ments are made with the setupcontrol on the equipmentunder test while viewing thewaveform monitor.)

2. Waveform luminance (whitebar) at 100 IRE. (Amplitudeadjustments are made with thevideo level, or video gain control on the equipmentunder test while viewing thewaveform monitor.) Refer tothe Waveform MonitorTechniques section, “A Wordof Caution” (page 2-5) if thereis more than one video gain orlevel control in the pathyou’re testing.

Using the test signal generator ’scolor bars as a reference signal,you should check to make certain the signals appear on the waveform monitor and vec-torscope as mentioned above.Again, use the phase control toensure the color burst vector isproperly positioned and the vector color dots are in theirgraticule boxes. With this refer-ence position established, anydifferences in other signals youselect will be obvious.

Checking Chrominance Phase Now that you’ve established areference position on the vec-torscope, the next step for checking the signal from the studio VTR equipment againstthe reference signal is to playback a videotape with the 75%color bars recorded. Select theequipment under test on thevideo switcher (VTR, TBC orproc amp), or by appropriatelyconnecting cables, and look atthe vectorscope display.

If the burst phase vector lines up on the horizontal axis, butthe dot pattern is rotated withrespect to the boxes, there is a chroma phase problem withthe equipment under test(Figure 3-1b).

This rotation of the dot patternmeans that the chrominancephase is incorrect relative to thecolor burst. This phasing errorcauses hues to be wrong — people’s faces appear green orpurple, for example.

Correct chrominance phase iscritical, even if hue errors aren’tobvious on a picture monitor. Tocorrect a chrominance phasee rro r, perf o rm the following steps:

1.Ensure the vectorscope’s burstdisplay is aligned with thehorizontal axis (phase controlon the vectorscope).

2.Adjust the hue control orchroma phase control on theequipment in the path to getthe dots as close to theirrespective graticule boxes aspossible. They may be toonear the center of the displayor too far away, but they willlie in the proper directionfrom the center when phase is correct.

If you are working with a camerain your system, the adjustmentsare no different. You just have tolook at the cameras output, or atthe output of the camera controlunit (CCU) if you are using one,when you make the adjustments.

It is often best to go back andforth between the chrominancegain (discussed next) and phaseadjustments to get the dots intheir graticule boxes. The dotswill rarely be exactly centeredon the cross points in the smallerboxes. It is, however, acceptableif they fall somewhere withinthe small boxes rather than thelarger boxes.

Remember, these tests are onlychecking the bar signal relativeto the color burst. Not all equip-ment requires or makes availablethis adjustment. Phasing (ortiming) several sources or pathsso their signals may be switchedor mixed requires external refer-ence for the vectorscope andsomewhat different techniques.These are discussed in theSetting Up a Genlocked Studiosection beginning on page 4-1.

Page 16: NTSC Video Measurements The Basics - Educypedia

3-4

Checking Chrominance Gain(Amplitude) Adjusting chroma phase alonemay not put the dots in theirboxes because they may fallshort of, or extend too far out-side the boxes (Figure 3-3). Thisindicates a chroma gain error.

For checking chroma gain on aVTR, you will again use thecolor bars signal recorded at thehead of the tape — other equip-ment may require a bars signalas its input. Again, you shouldcheck luminance levels on thewaveform monitor, and correctany errors, before you make anymeasurements with the vec-torscope. (Remember the vec-torscope does not display anyinformation about luminance.)

1.Use the setup control (if pro-vided) to position the blacklevel at 7.5 IRE.

2.Use the video level controlson the equipment under test toadjust the white level on thewaveform monitor to 100 IRE.

While the tape is playing therecorded bars, dots will appearon the vectorscope. If the dotsare beyond the boxes, thechroma amplitude is too high.If the dots fall short of theboxes, chrominance is too low.

3. Adjust the chroma gain contro lof the equipment under testuntil the dots fall into their boxes.

If adjusting the chroma gain control doesn’t get the dots intotheir boxes, you may need to re-adjust hue or have the equipment serviced.

If you happen to run the tapepast the color bars signal andinto active video, you’ll noticethat the vector display looksvery different. This is becausethere are typically no large areasof primary or secondary colorsin the picture to create the bardots. The display looks fuzzy orblurred because the vectorshows the blend of hues withinthe picture.

Checking White Balance Along with checking for chromaphase and gain, there is anotherstudio application that uses a

vectorscope in the internal refer-ence mode — checking thewhite balance of video cameras.

White balance is the process ofbalancing the camera’s red, gre e nand blue channels. When thesechannels are properly balanced,the camera’s output signals willreproduce whites in a scenewithout adding color. In o rd e rfor the camera to re p ro d u c ecolors correctly, it must be ableto reproduce white accurately.The signals from the green,blue, and red sensing elementsin the camera must be properlybalanced to ensure there is noc h rominance signal at the cameraoutput when it is viewing white.

B e f o re you can set white balance,you must first check the camerafor proper black balance:

1.Connect the video camera signal to the vectorscopeinput, or switch to the cameraon the video switcher.

2.Cap the cameras lens.

If the black balance is correct,the camera’s signal will producea fuzzy spot in the center of thevectorscope display, the same as for proper white balance(Figure 3-2). If the spot is dis-tended away from the center inany direction, proceed as follows:

3.Adjust the black balance control on the camera untilthe spot returns to the centerof the display.

To check white balance, uncapthe camera and point it at a purewhite target. If white balance iscorrect, the resulting signal fromthe camera should again pro d u c ea fuzzy spot in the center of thevectorscope display (Figure 3-2).If any color channel in the camerais out of balance, the fuzzy spot on the vectorscope will bedistended or moved toward thecorresponding color’s graticulebox. For example, too much redsignal moves the spot towardthe red box.

If the vectorscope display indi-cates incorrect color balance, itcan be corrected by using thecamera’s white balance controls.Cameras that allow manual whitebalance adjustment usually have two controls (red gain and

blue gain). It may help to remem-ber the red balance control willmove the spot vertically and the blue control will move ithorizontally. To manually setwhite balance, the followingsteps are recommended:

1.Aim the camera at a white reference. (Before attemptingto correct a white balanceproblem, make sure you havethe correct filter selected onthe camera for the type oflighting in use. The wrong filter selection will cause an apparent white balance problem on the vectorscope.)

2.Adjust the red and blue gaincontrols on the camera to getthe fuzzy spot in the center ofthe graticule on the vectorscope.

Some cameras with an automaticwhite balance feature may notoffer manual adjustment andwill have to be referred to a service technician if the auto-matic balancing feature does not achieve the desired results.

Conclusion

This section has dealt with threebasic uses of a vectorscope —evaluation of chrominance toburst phase, chrominance gain,and color balancing of a camera.There are other, more extensive,m e a s u rements that you can makewith a simple vectorscope, andothers that require instrumentswith enhanced capabilities.Several of these more advancedtechniques are addressed in thefollowing sections: Setting Up a Genlocked Studio andIntermediate Video Testing.

Figure 3-3. This display shows the vector dots extending beyond theirboxes, indicating a chroma gain er ror.

Page 17: NTSC Video Measurements The Basics - Educypedia

4-1

Setting Up A GenlockedStudio

generator. Equipment that is synchronized by a master gener-ator is often referred to as beinggenerator locked, or “genlocked”for short.

Genlocked equipment workstogether in synchronization witha master sync signal. It’s muchlike the members of a marchingband “locking” their steps to thebeat of the bass drum. Withoutthe steady beat of the drum —the synchronization pulse — theband members wouldn’t have acommon reference for staying instep with each other.

Similarly, video equipment thatisn’t genlocked has difficultyworking together. Trying to dissolve from the output of oneVTR to another results in picturetearing or roll until synchroniza-tion is reestablished. With gen-locked equipment, the dissolveis smooth and “in step” with themaster sync.

The black burst signal (Figure 4-1 )is often used for genlockingequipment. It is a composite signal with a horizontal syncpulse and a small packet of3.58 MHz color subcarrier(color burst). The term blackburst arises from the fact that the active picture portion of the signal is at 7.5 IRE (black) and itcontains color burst.

Establishing a genlocked studiois simple and need not requireexpensive equipment. This section will show you how togenlock your equipment withthe Tektronix TSG 200 N T S CGenerator (Figure 4-2). This gen-erator is an economical source of multiple black burst signaloutputs for genlocking.

Genlock BasicsThere are three basic require-ments for setting up a genlockedstudio. You need:

• Video equipment that can begenlocked

• A genlock signal source

• Adjustment for synchronizingequipment with the genlocksource.

The first of these requirements— video equipment that can begenlocked — is relatively easy tomeet. Most professional videoequipment is genlockable. To seeif your cameras, VCRs or VTRs,and other equipment meet thisrequirement, check the equip-ment for a genlock loop-throughinput or an external referenceinput (Ext. Ref.). Also, the equip-ment must have some provisionfor adjustment of timing, whetherit is a single control for adjustingboth H-timing and SC-phase, or separate controls for adjusting each.

The genlock or external re f e re n c einput connector is necessary forapplying an external synchro-nization signal — typically blackburst — to the equipment. Thetiming controls are necessary foradjusting the equipment for synchronization with the refer-ence’s horizontal (H) sync pulseand subcarrier (SC) color burst phase.

With a genlock signal source,such as a Tektronix TSG 200, ablack burst signal can be distrib-uted to the studio equipment asshown in Figure 4-3.

Color Burst

0 IRE Blanking Level

7.5 IRE Black Level (Setup)

HorizontalSync Pulse

Figure 4-1. The black burst signal used in genlocking is composed of thehorizontal sync pulse, subcarrier color burst, and a 7.5 IRE black signal.

Figure 4-3. A black burst signal is distributed from the test signal generatorfor genlocking studio equipment. If the signal generator has multiple blackburst outputs, as is the case with the TSG 200, a distribution amplifier maynot be necessary since they have five black burst outputs.

Fades and dissolves are a basicpart of most professional videoproductions. Even the smalleststudios use them. Before dissolvesor other editing effects can beused, though, the studio equip-ment must be synchronized.

Synchronization is done by locking cameras, VTRs, editors,switchers, and other videoequipment to a master sync

Figure 4-2. The Tektronix TSG 200 provide multiple black burst signal out-puts for economically genlocking small studios. Additional features on the TSG 200 can be used for basic video testing and other odd jobs around the studio, including ID generation, setting up picture monitors and blacking tapes.

Page 18: NTSC Video Measurements The Basics - Educypedia

4-2

There are several importantthings to note about this distrib-ution diagram. First, the signalgenerator used for genlockingshould have multiple black burstoutputs. A unique black burstsignal is not needed for eachpiece of equipment to be gen-locked. Looping-through equip-ment in different locations within a facility is possible, butpresents one major problem —too long a cable run will pro d u c edelays beyond the adjustmentrange of the equipment being locked.

Generally speaking, a single blackburst feed for each cluster ofequipment is sufficient, with thatsingle feed being looped throughjust one group of equipment.

If multiple black burst outputsare not available, a multiple out-put signal distribution amplifiershould be used with one outputfor each cluster of equipment tobe genlocked.

Also notice in Figure 4-3 that theblack burst signal is distributedto the Time Base Corrector (TBC)for each VTR, rather than theVTR itself. The reason for this isthat the TBC is actually the finalprogram output of the VTR. TheTBC is where genlocking andtiming adjustments need to occur.

Similarly, if you use camera control units (CCUs), connectthe black burst output to theCCUs, not to the cameras.Camera timing adjustments,unlike TBC/VTR adjustments,are made at the camera, not atthe CCU.

Finally, note that one of theblack burst signal outputs fromthe test signal generator is loop-t h rough connected to the extern a lreference inputs of a waveformmonitor and vectorscope. (Don’tforget the 75 ohm terminator atthe end of each chain of genlockor external reference loop-t h roughs.) The waveform monitorand vectorscope are also loop-t h rough connected to the Pro g r a mOutput of the studio switcher.

These connections to a wave-form monitor and vectorscopeare necessary for the finalrequirement — adjusting theequipment for synchronizationwith the genlock source.

Adjusting TBCsTBCs have several controls, inaddition to the timing controlsjust mentioned, that are used tocompensate for gain deficienciesand color errors on tapes. Setup(or black level), video gain,chrominance gain, and hue arecommonly used adjustments onTBCs. Each of these adjustmentshas already been covered in thisapplication note and will not berepeated here. The purpose ofthis section is to link the nearlyuniversal inclusion of color barson tape leaders to TBC adjust-ments and discuss a few guide-lines to follow while makingthese adjustments.

Video levels and hues recordedon tape sometimes differ fromfacility to facility, and VTR toVTR. This is a fact of life. Whendifferences exist they must bec o rrected. By re c o rding color barson the beginning of every tape,operators can make the necessaryTBC adjustments quickly andsimply with just a waveformmonitor and vectorscope.

The TBC’s horizontal timing andsubcarrier phase controlsshouldn’t require checking oradjustment each time you playback a tape from a differentsource. Since the TBC alwaysreplaces the sync and burst ofthe recorded signal, playing backdifferent tapes should not affectthe TBC’s system timing. Manyexperienced operators make TBCadjustments in the followingorder to minimize interactionbetween the adjustments: 1) setup,2) video gain, 3) hue, and 4) chrominance gain.

Remember that TBC adjustmentsmust be made when the VTRconnected to it is playing back arecorded color bars signal, notwhen the VTR is in the electrical-to-electrical (E-to-E) mode.

Adjusting for SynchronizationAs mentioned before, the hori -zontal sync pulse and subcarrierburst of the black burst signalare like the drum beat to whichmembers of a marching bandsynchronize their steps.Everyone’s footsteps shouldstrike the ground in time withthe drum. Additionally, for thewhole band to be in step, every-one’s left foot should be strikingthe ground at the same time.Occasionally, however, a bandmember will be out of step, and you’ll see them do a little“skip-step” as a timing adjustmentto get in step with the rest of the band.

Similarly, each piece of videoequipment must have some timing adjustments made inorder for it to be “in step” withthe rest of the video system.Adjusting system timing is oneof the most fundamental — and critical — procedures in the studio.

System timing adjustmentsinvolve setting the H-timing andSC-phase controls on each pieceof equipment. The adjustmentsare made so that each piece ofequipment’s horizontal syncpulse and color burst phase lineup with the sync pulse and burstphase of the reference genlocksignal (black burst).

Making sure that each piece ofequipment is “in step” with thegenlock signal prevents horizontaljumps and color shifts of the picture when switching betweenvideo sources. Matching the timing of the sync pulses makessure the scanning of each picturesource is in step (e.g., the imageswill stay in the same position,without roll, tearing or “jumps”during a cut or fade). Matchingthe burst phase of the varioussources maintains the correctcolor of the images during editing transitions.

Page 19: NTSC Video Measurements The Basics - Educypedia

4-3

burst vector aligned with the9 o’clock position on the vec-torscope display? If it isn’t, usethe SC-phase control on theselected piece of equipment toset color burst to the 9 o’clockzero reference.

In the case of a single timingadjustment (where SC-phase and H-timing have a fixed rela-tionship), simply observe thevectorscope and adjust theequipment such that the colorburst vector lines up with the9 o’clock reference line on thevectorscope’s graticule. Someequipment has a wide rangeadjustment which will continueto move the sync timing beyondthe first place where the burstphase is correct — if so, continueto adjust the control in thea p p ropriate direction. This shouldmove the H sync pulse to the zero -reference marker as well. If itdoes not, there is either an intern a lH-timing adjustment that is setincorrectly or the equipmentneeds serv i c i n g . In either case,an experienced technicianshould attend to the problem.

Setting sync and color burst tothe established zero referencescompletes timing alignment forthe selected piece of equipment.That piece of equipment is nowin step with the black burst reference signal.

Now, using the switcher, selectanother piece of equipment andadjust its H-timing and SC-phasecontrols to zero reference itssync and color burst output.Continue this equipment selection and zero-referencingprocess for all other pieces ofvideo equipment in the system.Remember that for VTRs, thetiming adjustments are done atthe TBC.

When these adjustments havebeen completed for all pieces ofequipment in the system, eachpiece of equipment will be insynchronization and properlytimed with all other pieces ofequipment. You now will beable to switch smoothly betweenvideo sources and make cleanedits without picture roll or horizontal jumps.

Figure 4-4. Waveform monitor display of the reference sync pulse edgeadjusted to correspond to a major timing mark. This timing mark can nowserve as the zero reference for individually adjusting the H-phase of eachpiece of video equipment in the system.

B e f o re making timing adjustments,you should check the video gainfrom each piece of equipment.The SMPTE color bars signalfrom the TSG 200 can be usedfor this. If necessary, correct anyvideo gains before going on tosystem timing adjustments.

System timing adjustments aremade with a waveform monitorand vectorscope connected tothe switcher output as shown inFigure 4-3. Make sure that boththe waveform monitor and vec-torscope are set to trigger on theexternal reference signal.

The next step is to set up zerotiming references on both thewaveform monitor and vec-torscope. The first step in this“zeroing” process is to select thetest signal generator’s output as

Figure 4-5. Vectorscope display of the reference color burst vector adjustedto the 9 o’clock position. All other pieces of equipment should be SC-phaseadjusted to correspond to the 9 o’clock reference.

the switcher’s active input. Thisapplies the generator’s test signal to the Channel A inputs of both the waveform monitorand vectorscope.

On the waveform monitor, makethe necessary adjustments toexpand the display on the hori-zontal sync pulse. Be sure thewaveform monitor is in theexternal reference mode, i.e.,displaying the program output ofthe switcher, but referenced tothe black burst. Use the waveformmonitor’s horizontal positioningcontrol to place the leading edgeof sync on one of the display’smajor timing marks. This timingmark is now the zero-time refer-ence for inputs from all othervideo equipment, and the hori -zontal position control shouldnot be moved during the rest ofthe system timing adjustments.Figure 4-4 shows an example ofthe reference sync edge displaywith the sync edge positionedon a major timing mark.

The next step is to zero re f e re n c ethe vectorscope to the referencesignal’s color burst subcarrierphase. This is done by using thevectorscope’s phase control toposition the color burst vector tothe 9 o’clock position. Hereagain, be sure the vectorscope isin the external re f e rence mode —locking to the external black burst.This is shown in Figure 4-5.Once the reference signal’s colorburst is positioned to 9 o’clock,do not touch the vectorscope’sphase control again during therest of the timing adjustments.

With sync and color burst zeroreferencing completed, use theswitcher to select the signal output from the first piece ofvideo equipment to be adjusted.

The waveform monitor will display the horizontal syncpulse from the selected piece ofequipment. If the sync edge doesnot line up on the previouslyestablished zero-reference mark,adjust the H-timing control onthe selected piece of equipmentto place the sync edge on thezero-reference marker.

Look at the color burst displayon the vectorscope. Is the color

Page 20: NTSC Video Measurements The Basics - Educypedia

5-1

Previous sections of this booklethave discussed making basicmeasurements and adjustmentsin video systems. This sectiongoes on to discuss intermediatelevel tests using three pieces oft e s t - a n d - m e a s u rement equipment— a signal generator, waveformmonitor, and vectorscope. Thissection describes how to use that test equipment to make the following tests:

• insertion gain

• frequency response

• c h rominance-to-luminance gain

• luminance nonlinearity

• differential gain

• differential phase

The picture impairments causedby the distortions these tests aredesigned to detect are alsodescribed where appropriate.

Test equipment first

To ensure valid test results, youmust use good test equipment.E s s e n t i a l l y, there are two primaryrequirements for this. First, thetest equipment’s specified performance must exceed theexpected performance of the system being tested. And second,the test equipment must be performing to its specifications.

The latter requirement is really atwo-stage requirement. The testequipment must always be setupusing standard operator calibra-tion procedures. For highestconfidence, the equipmentshould also be fully calibratedperiodically at a qualified service center. (Full and precisecalibration of test equipmentrequires special skills and labo-ratory instruments not normallyfound in a video facility.)

As for test equipment capabilities,all the tests described in thisapplication note can easily beperformed with the Tektronixi n s t ruments shown in Figure 5-1:

• 1720 Vectorscope, used toevaluate the chrominance(color) information in a video signal

• 1730 Waveform Monitor, usedfor video signal level and timing measurements

• TSG 170A NTSC TelevisionGenerator, used to generate thenecessary standard test signals

This test equipment is excellentfor all basic and intermediatetesting requirements. Other simi-lar test equipment may be avail-able to you and can be used inessentially the same manner. Insome instances, the waveformmonitor and vectorscope functions

Intermediate NTSCVideo Testing

1730 Waveform Monitor.1720 Vectorscope.

TSG 170A NTSC Television Generator.

Figure 5-1. Video system testing requires a signal generator, a waveform monitor, and a vectorscope.

Page 21: NTSC Video Measurements The Basics - Educypedia

Generator

SignalPathBeingTested

WaveformMonitor

Vectorscope

complete in-service or accep-tance testing. (In-service testingalso requires a VITS inserter,such as the Tektronix VITS200,VITS100 or 1910, to generateand insert vertical interval test signals.)

Setting Up the Test EquipmentRegardless of the specific testequipment models used, thebasic test concepts and setupremains the same. The diagramin Figure 5-2 shows the setup fortesting any video system, nomatter how simple or complex.

The block labeled Signal PathBeing Tested can be any part ofthe video system. It can be a single piece of equipment suchas a playback deck or a re c o rd i n gdeck. Or the path can include aswitcher along with other equip-ment such as cameras, VTRs,distribution amplifiers, character/title generators, special effectsgenerators, and so forth. In anycase, all testing is done in essen-tially the same manner — injecta signal at the start of the path(the input), and observe or mea-sure any distortions to the signalat the end of the path (the output).For additional details, see theearlier section on Connectingand Terminating Instrumentsstarting on page 1-3, and “A Wo rd of Caution” on page 2 - 5 .

Before actual testing begins,however, you should check thetest equipment against its owninternal calibration signal. Thisis typically referred to as a usercalibration procedure, and it willbe outlined in the instrumentoperator’s manual.

Also, as a matter of reference,you should view an undistortedtest signal on the test instru m e n t ’sdisplay. To do this, connect the test signal generator’s outputdirectly to the vectorscope or waveform monitor input terminals.

With the instruments properlyconnected and terminated, you'llsee the test signal displayed onthe vectorscope or waveform

monitor. Study this display carefully. It’s your reference ofwhat the “perfect” signal shouldlook like.

Ideally, you should see exactlythe same “perfect” display afterthe test signal is passed throughthe video path being tested.However, if there are distortionsin the signal path, you’ll see corresponding changes in thedisplayed signal. Various typesof distortion are described furt h e rin the following discussion oftest procedures.

Insertion GainInsertion gain reflects the videopath’s ability to maintain correctsignal amplitudes from input tooutput. The general test proce-dure is to apply a standardNTSC 1 volt (140 IRE) signal tothe video path or equipmentinput. The test signal is thenmeasured at various points alongthe signal path to verify its correct amplitude.

It is important to verify insertiongain before doing any other tests.If there are insertion gain errorsand they are not corrected, sub-sequent tests will be incorrect aswell. This is because most testsare based on the presumptionthat insertion gain is correct.Also equipment that is pro c e s s i n ga nonstandard level signal willoften introduce other distortions.

Insertion gain errors are usuallyexpressed as a percent variationfrom the nominal value. There-f o re, the goal is zero insertion gain.

Positive insertion gain errorsindicate an increase (gain) in signal amplitude from input tooutput. This can lead to distor-tions from signal overload.Negative insertion gain indicatesa decrease (loss) in signal ampli-tude. This causes dark picturesand also reduces the signal-to-noise ratio (SNR).

will be combined in a singleinstrument. This is the case, for example, with the Tektronix1740A and 1750A SeriesWaveform/Vector Monitors. Withthese instruments, you select thew a v e f o rm monitor or vectorscopemode by pressing the Waveformor Vector button on the front panel.

Similarly, more advanced, multi-function instruments, such asthe Tektronix 1780R or VM700TVideo Measurement Sets, allowpush-button selection of wave-form monitor or vectorscopemodes. Additionally, the 1780Rand VM700T provide numerousother features that simplify themeasurements described in thissection as well as many othermeasurements. Some of thesefeatures include microprocessor-assisted or fully automatic m e a s u rements, user- p ro g r a m m a b l emeasurement sequences, graphicdisplay of measurements, and brighter displays in line-select mode.

The more sophisticated capabili -ties and enhanced precision ofthese instruments makes themideal for all measurementsincluding applications such as

5-2

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

Page 22: NTSC Video Measurements The Basics - Educypedia

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

When there is an insertion gainerror, the equipment under testshould be adjusted to removethe error. To do this, use theappropriate control on theequipment under test to alignthe color bars reference bar withthe waveform monitor’s 100 IREgraticule line. In doing this, besure to keep the blanking levelon 0 IRE by using the waveformmonitor’s position control if necessary. With the reference baron 100 IRE, the sync and colorburst amplitudes should both be 40 IRE peak-to-peak. You may need to make separateadjustments to make sure theselevels are correct.

In making gain adjustments, it’simportant to start first withequipment at the beginning ofthe video signal path. Otherwise,you may simply mask ratherthan correct errors occurring earlier in the signal path.

After verifying that insertiongain is correct, you need to checkchrominance amplitudes. Thiscan be done using either a wave-form monitor or a vectorscope.

On the waveform monitor,chrominance amplitude ischecked by first measuring thecolor burst amplitude. It shouldextend from -20 IRE to +20 IRE(40 IRE peak-to-peak). Thencheck the maximum level of thefirst two color bars (yellow andcyan). Both should be at exactly100 IRE. If they aren’t, adjust thechrominance gain on the equip-ment under test to correct thechrominance levels.

Checking chrominance levels onthe vectorscope is done with thedisplay shown in Figure 5-5.The color burst vector shouldemanate from center screenalong the horizontal axis(9 o'clock position). If it doesn't,adjust the vectorscope’s phasecontrol to put the color burst inthe 9 o'clock position. Thencheck to see if all of the othervector dots fall into their properboxes, as shown in Figure 5-5.

5-3

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

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

I n s e rtion gain errors affect overallpicture brightness and may alsoaffect apparent color saturation.The human eye is very sensitiveto even small brightness varia-tions, particularly when theyoccur rapidly. Because of this,its especially important to makesure that all video switcherinputs are matched when different shots of the same sceneare being combined. It shouldalso be kept in mind that smallgain errors in several systemcomponents can quickly cascadeinto big errors.

Insertion gain testing is doneby applying either a SMPTEcolor bars signal or a full-field,75% color bars test signal havinga 100% white reference level.A waveform monitor is thenused to observe the color barssignal and make insertion gain measurements.

The TSG 170A supplies theSMPTE color bars signal. Sincethe white reference for SMPTEcolor bars is at 75%, the 100 IREY signal that overlays the firsttwo color bars, as shown inFigure 5-3, is more convenientfor insertion gain tests. Figure 5-4shows a full-field, 75% colorbars signal with a 100% whitereference. (Either of these signalscan also be used to checkchrominance amplitude andchrominance-to-burst phase ona vectorscope.)

The waveform monitor shouldbe set for a one- or two-linesweep with the filter selection inflat. (Other filter positions alterthe waveform monitor responseand may mask a gain error.)

With the sweep and filter selected,use the vertical position controlto align the trace’s blanking levelwith 0 IRE on the graticule. Asshown in Figure 5-3, the 100%white reference (Y signal) shouldnow coincide with the 100 IREgraticule line, and the sync tipsshould be at 40 IRE. Insertiongain error, if any, is determinedby noting the position of thewhite bar and subtracting 100%.For example, if the bar is at 93 IRE, insertion gain error is 93 - 100, or -7%.

Page 23: NTSC Video Measurements The Basics - Educypedia

If the color vector dots do notfall within their boxes, you’llneed to make chrominance gainor phase adjustments on theequipment under test. When thedots are beyond the boxes,toward the display edges,chrominance amplitude is toohigh. If the dots fall short,toward the display center,chrominance is too low. If thedots are rotated out of theirboxes, chrominance phase isincorrect relative to color burstand needs to be adjusted on theequipment under test.

If all of the dots cannot beadjusted to fall in their respec-tive boxes, other distortionerrors exist in the system. Thesecould include frequency re s p o n s eerrors or other gain or linearityerrors. Finding and correctingthese errors requires further testing and adjustments by aqualified service technician.

Frequency Response TestingFor optimum performance, thefrequency response of the entirevideo system should be as flat aspossible. Frequency responsetesting determines how flat thesystem actually is. If the systemis not flat, signal amplitudes willbecome distorted as a functionof frequency. For example, thesystem may attenuate higher frequency components morethan lower frequency componentsor vice versa.

For test purposes, television signals are divided roughly at1 MHz into low- and high-frequency ranges. Even thoughsome of the most important signal information falls below1 MHz — namely synchronizingpulses, general brightness level,and some active video — low-frequency testing is oftenneglected. This can be a trouble-some and costly oversight.

You should routinely check system response in the low-frequency range. Low-frequencytesting is further divided into line time and field time

distortions. A line bar, such asthe one found in the pulse andbar signal (Figure 5-6) is used forchecking for line-time distort i o n s .Field-time distortion testing isdone with either a field squarewave or a windowed bar signal.

The pulse and bar signal pro v i d e dby the TSG 170A is a windowedsignal, consisting of 130 lines ofthe pulse and bar signal in thecenter of the field. On a picturemonitor, this creates the windoweffect shown in Figure 5-7. If thevideo path can faithfully conveythis signal, you can assume thatthe system’s low-frequencyresponse is satisfactory.

To check field-time distortion,the waveform monitor should beset for a two-field sweep in theflat-response mode. Also, the dcrestorer should be in eitherslow-restore mode or turned off.

If the video path’s low-frequencyresponse is correct, you shouldsee perfectly flat horizontallines. Any tilt in these lines,such as shown in Figure 5-8,represents a field-rate impair-ment. Such an impairmentwould cause brightness varia-tions between the top and bottom of the picture. Providedthere is no insertion gain error,the field-time distortion can bemeasured as the percent varia-tion from the normal flat level,excluding the first and last 0.2milliseconds of the bar.

Similar observations are madewith the line bar signal for line-rate response problems. Any tilt in the line bar wouldproduce brightness variationsbetween the left and right sidesof the picture. With active video,line-time distortion produceshorizontal “streaking” — usuallyseen as light and/or dark streaksextending to the right of horizontaltransitions in the picture. Inquantifying this measurement,exclude the first and lastmicrosecond of the bar, sincedistortions near the transitionoccur at frequencies above theline rate.

Figure 5-6. The line bar in this pulse and bar signal is used to check forlow-frequency distortions at the line rate.

5-4

Figure 5-7. The windowed pulse and bar signal is used to check for low-frequency distortions at the field rate.

Figure 5-8. Waveform monitor display of a windowed pulse and bar signalshowing about 6% field-time distortion.

Page 24: NTSC Video Measurements The Basics - Educypedia

If your equipment exhibits linetime or field time distortionsbeyond the limits specified bythe manufacturer, it must be s e rviced by a qualified technician.T h e re are no external adjustmentsfor either low frequencyresponse or high frequencyresponse errors.

Response to higher frequencies,those above 1 MHz, is oftenchecked with a multiburst testsignal. Response problems above1 MHz can cause impairments in either or both the chro m i n a n c eand monochrome detail of pictures.

The multiburst signal testsresponse by applying packets ofdiscrete frequencies rangingfrom about 500 kHz to 4.2 MHz.The Tektronix TSG 170A NTSCTelevision Generator providesthe multiburst signal shown inFigure 5-9.

The multiburst signal is com-posed of six frequency packets.The second packet from the righthas a frequency of 3.58 MHz andis used to check color subcarrierresponse characteristics. Noticealso that the multiburst signalstarts with a low-frequency signal (bar, sine wave, or, as isthe case in Figure 5-9, a squarewave). This low-frequency signalis used as an amplitude re f e re n c ein measuring the relative amplitudes of the other packets.

Be aware there are many diff e re n tconfigurations of multiburst signals — this test signal has along history and meets many differing needs. When testing aVTR, you should use a reducedamplitude (60 IRE vs. 100 IRE)multiburst signal. This is toavoid intermodulation betweenthe multiburst frequencies andthe FM recording system in theVTR. Such intermodulation cancause signal distortions evenwhen there actually is nothingwrong with the VTR. Also, whenevaluating multiburst amplitudes,you need to take into considerationthe VTR’s specified bandwidth.

This can be significantly less than4.2 MHz, which means that youshould expect to see attenuationof the high-frequency packets.

For multiburst measurements,the waveform monitor should beset to a one- or two-line sweepand flat response. For a perfectsystem response, the multiburstdisplay would show all packetsas having the same peak-to-peakamplitude. Any significantamplitude variations in packetsindicate a frequency responsevariation — it is an error only ifit is outside the equipment’sspecifications. Some videoequipment, such as distributionamplifiers or switchers, maypass the multiburst packets with very little frequencyresponse distortion.

The response at a specific fre-quency can be expressed as apercent of nominal value or indecibels. Either method ofexpression is based on peak-to-peak amplitude measurements.

Again, the absolute frequencyresponse is often not the issue ofgreatest concern. Instead, theresponse relative to a particularspecification or to earlier mea-surements is used to indicateequipment performance. A difference between past and p resent response measurements isa sign of equipment performancechanging and may indicate aneed for service.

High-frequency rolloff, such asshown in Figure 5-10, is pro b a b l ythe most common type ofresponse distortion. When thisoccurs, luminance fine detail isdegraded. Many VTRs will showa much greater rolloff and stillbe within specifications.

H i g h - f requency peaking is anothertype of distortion. This is shownin Figure 5-11. It is usuallycaused by incorrect equalizeradjustment or misadjustment ofsome other compensatingdevice. This problem causesnoisy pictures — you see sparklesand overly emphasized edges.

5-5

Figure 5-9. To display this multiburst signal, the waveform monitor is set forone-line sweep and flat response. This signal shows a flat response for highfrequencies (500 kHz to 4.2 MHz).

Figure 5-10. High-frequency rolloff is apparent in this multiburst signal display. The maximum error occurs where the signal is about 50 out of60IRE, which is -1.6 dB.

Figure 5-11. This multiburst signal suffers from high-frequency peaking.The maximum error occurs where the signal is 80 IRE instead of the nominal 60 IRE (1.3dB increase).

Page 25: NTSC Video Measurements The Basics - Educypedia

Center-frequency dipping orpeaking can also occur. Whenthis affects a broad range of fre-quencies, it can be detected withthe multiburst signal. On theother hand, peaking or dippingmay only affect a narrow rangeof frequencies. When this is thecase and the affected frequenciesoccur between multiburst packets, the distortion can goundetected by this test technique.

To catch such narrowbandresponse problems, a sweep signal or other continuous-bandresponse measurement techniqueis needed. Fortunately, narrow-band peaking or dipping doesnot occur often. And when itdoes occur, it can be detected asnoticeable ringing on sync pulsesor other sharp signal transitions.Such ringing indicates the needfor more thorough response evaluation using frequencysweep, multipulse, or (sin x)/xpulse techniques. To learn moreabout these techniques, refer toTektronix publications Using theMultipulse Waveform to MeasureGroup Delay and AmplitudeErrors (20W-7076), andFrequency-Response TestingUsing a (Sin x)/x Signal and theVM700A Video Measurement Set (20W-7064).

Chrominance to LuminanceGainWhen a signal passes through avideo system, there should be nochange in the relative amplitudesof chrominance and luminance.In other words, the ratio of thechrominance and luminancegains, which is sometimesreferred to as relative chromalevel, should remain the same.If there are relative chromalevel errors, the pictures colorsaturation will be incorrect.

Relative chroma level is checkedby measuring chrominance-to-luminance gain. Measured gain-ratio errors can be expressed inIRE, percent, or dB. Whenchrominance components arepeaked relative to luminance,the error is a positive number.When chrominance is attenuated,the error is negative.

Measurements are made usingthe special chrominance pulseshown in Figure 5-12. This pulse is the modulated 12.5Tsine-squared pulse which isincluded in many combinationtest signals. For example, it’sincluded in both the pulse andbar and NTC 7 Composite signalsp rovided by the TSG 170A NTSCTelevision Generator. The pulseand bar signal also includes aline bar and a 2T pulse.

The chrominance pulse consistsof a low-frequency, sine-squaredluminance component that’sbeen added to a chrominancepacket having a sine-squaredmodulation envelope. Thesecombined pulse componentshave characteristics that allowgain and phase errors to be seenas distortions of the pulse base-line. In the case of Figure 5-12,there are no errors and the baseline is flat.

Figure 5-13 shows what happenswhen there’s a relative chromalevel distortion. The upwardbowing of the baseline (the negative waveform peaks) indicates that chrominance isreduced relative to luminance. If chrominance were increasedrelative to luminance, the baseline would bow downward.

Chrominance-to-luminance gaine rror can be measured directly onthe waveform monitor graticule.This is done by comparing thepeak-to-peak amplitude of thechrominance component of t h e12.5T pulse to the norm a l i z e dwhite level reference bar. In thecase of Figure 5-13, the chromi-nance amplitude is 80 IRE, indicating a 20% error. Thismeasurement approach is validonly if there is no low-frequencyamplitude distortion and thereis negligible chrominance-to-luminance delay.

A chrominance-to-luminancegain error is easy to correct if theequipment under test has anexternal chroma gain control. Ifit does, simply adjust the chro m again control for a flat chro m i n a n c epulse baseline. If it does not, theequipment must be serviced by aqualified technician.

5-6

Figure 5-12. A 12.5T chrominance pulse, shown in the center of this display, is used to evaluate chrominance-to-luminance gain and delayerrors. For zero gain and delay errors, the negative peaks of this modulated sine-squared pulse should line up on the pulse baseline.

Figure 5-13. A single, symmetric peak in the chrominance pulse baselineindicates a chrominance-to-luminance gain error. This example shows anerror of about 20%.

Page 26: NTSC Video Measurements The Basics - Educypedia

more definitive measurementresults can often be obtained byusing a generator, such as theTektronix TSG 170A, that provides test signals at severaldifferent APLs. When using sucha generator, run the test with atleast two APLs, one low and onehigh, and report the worst result.

The three remaining measure-ments to be discussed in thissection fall into the category ofdetermining the degree of signallinearity (or nonlinearity). Theseare the tests for luminance nonlinearity, differential gain,and differential phase. These canbe conducted with test signalsprovided by the same signal generator used in the previoustests, the Tektronix TSG 170ANTSC Television Generator.

Luminance nonlinearity

The luminance nonlinearity testis used to determine if luminancegain is affected by luminance level.This measurement is also re f e rre dto as differential luminance.

Luminance nonlinearity errorsoccur when the video systemfails to process luminance information consistently overthe entire amplitude range.When the progression from onebrightness level to another isnonlinear, the accuracy withwhich the picture display brightness levels in the nonlinearrange is affected. For example,shades of gray that should bedistinct may appear the same.

Luminance nonlinearity can be evaluated using a staircase signal having five or ten steps,with or without high-frequencymodulation. However, the modulated staircase may give adifferent measure of luminancenonlinearity — the high-fre q u e n c ysignals may affect the lower-frequency processing. TheTSG 170A contains both a five-step luminance staircase and afive-step modulated staircase(contained in the NTC 7Composite signal). The five-stepluminance staircase and theNTC 7 Composite signal areshown in Figures 5-15 and 5-16, respectively.

Chrominance-to-luminancedelay, on the other hand, is amore common error. Its presenceis indicated when the chromi-nance pulse baseline has a sinu-soidal distortion such as shownin Figure 5-14. When there isdelay error only, the sinusoidallobes are symmetric and thepulse amplitude should matchthe white level reference baramplitude (100 IRE). This is thecase shown in Figure 5-14.Asymmetrical lobes along withpeaking or attenuation of thepulse amplitude indicate thepresence of combined gain anddelay errors.

Since there are no user adjust-ments for chrominance-to-luminance delay on compositeNTSC equipment, correcting thisproblem requires a trip to a localservice center.

Measuring chrominance-to-luminance delay is beyond thescope of intermediate video system testing. However, you canlearn about these more advancedmeasurements by referring toTektronix publicationsTelevision Measurements ForNTSC Systems (063-0566-00) orUsing the Multipulse Waveformto Measure Group Delay andAmplitude Errors (20W-7076).

Nonlinear DistortionsThus far, the focus has been ondistortions having equal effectsfor signals of differing amplitudes.These are linear distortionsbecause the amount of signaldistortion varies linearly withsignal amplitude. In otherwords, a linear distortion causesthe same percent error on asmall signal as on a large signal.

Nonlinear distortions, by contrast,are amplitude dependent. Theymay be affected by changes inAverage Picture Level (APL) aswell as instantaneous signallevel. In other words, nonlineard i s t o rtion causes diff e rent perc e n terrors depending on signalamplitude. An overdriven amplifier, for example, causesnonlinear distortion when itc o m p resses or clips signal ampli-t u d e peaks. Since APL changesshould be taken into account,

5-7

Figure 5-14. A sinusoidal distortion of the chrominance pulse baseline indi-cates that chrominance is either advanced or delayed relative to luminance.

Figure 5-15. The five-step luminance staircase signal is used to measureluminance nonlinearity.

Figure 5-16. The five-step staircase of the NTC 7 Composite test signal canalso be used to measure luminance nonlinearity. To make the measurement,though, the waveform monitor’s filter must be set to low-pass to removethe chrominance information.

Page 27: NTSC Video Measurements The Basics - Educypedia

To make the luminance nonlin-earity measurement on thewaveform monitor using a luminance staircase, ensure thewaveform monitor’s filter selec-tion is set to flat. When makingthe measurement on a modulatedstaircase, you must first engagethe low-pass filter (LPASS). Thiseliminates the high-frequencymodulation, as shown inFigure 5-17 (but remember, itdoes not eliminate it from theequipment under test). Then calculate the nonlinearity erroras the diff e rence in height betweenthe largest and smallest steps.The error is then expressed as apercentage of the largest step.

In Figure 5-18, for example, thelargest step covers about 23 IRE,and the smallest step covers about16 IRE. From these two values,the luminance nonlinearity erroris computed to be 30%. This is a fairly large error that was chosen simply for the purpose of illustration.

More precise measurements canbe made by using the waveformmonitor’s X5 magnifier toexpand the steps. With X5 magnification, a 20 IRE step vertically covers 100 IRE on thegraticule scale. By positioningthe waveform vertically, you canview each step individually over the full graticule for high-resolution measurements.

Some waveform monitors — suchas the Te k t ronix 1780R — includea diff e rentiated step filter. Thisfilter can be used for lumin a n c enonlinearity measure m e n t s on anunmodulated staircase signal.

The differentiated step filter converts each step to a spike of aheight proportional to the stepheight (Figure 5-19). Any varia-tions in step height becomeimmediately apparent as changesin spike heights. By using thewaveform monitor’s variablegain control, the largest spikecan be set to 100 IRE and thepercent nonlinearity error can beread directly from the graticule.

If the amount of luminance non-linearity measured exceeds themanufacturer’s specifications,the equipment requires serviceor repair.

Differential gain

Differential gain testing allowsyou to determine when chromi-nance gain is being affected byluminance level. Differentialgain, sometimes referred to asdiff gain or dG, occurs when avideo system doesn’t process thechrominance signal consistentlyat all luminance levels. This cancause unwanted chrominanceamplitude increases anddecreases, all in the same signal.In other words, chrominance canbe too high at one luminancelevel and too low at another.

5-8

Figure 5-17. With the waveform monitor in low-pass filter mode, the undistorted modulated staircase appears as five luminance levels with equal steps.

Figure 5-18. This staircase shows increased gain (taller steps) around themiddle of the luminance range and decreased gain (shorter steps) at thehighest luminance level. The maximum linearity error on this signal is about 30%.

Figure 5-19. This differentiated staircase, which is displayed on the 1780RVideo Measurement Set, shows a luminance nonlinearity of about 10%.

Page 28: NTSC Video Measurements The Basics - Educypedia

Differential gain causes color saturation to have an unwarr a n t e ddependence on luminance level.This is seen in the television p i c t u re as saturation being aff e c t e dby variations in brightness.When a colored object movesfrom sunlight to shade, forexample, the color intensity willincrease or decrease abnormally.

A five-step modulated staircase,found on the NTC 7 Compositesignal, can be used to measuredifferential gain with either a waveform monitor or a vectorscope.

A modulated ramp, available in the TSG 170A, can also beused, as well as a 10-step modulated staircase.

To make measurements on awaveform monitor, set theinstrument for one- or two-linesweep with the chrominance filter selected. This filter passesonly the 3.58 MHz chrominancesignal and eliminates the lumi-nance steps. Also, for convenience,adjust the gain control so themaximum peak-to-peak amplitudeof the signal is 100 IRE.Disregarding the color burst signal, the top and bottom of thechrominance signal should beflat, as shown in Figure 5-20. If differential gain is present,packet amplitudes will not beu n i f o rm, as shown in Figure 5-21.

To determine the amount of differential gain, find the differ -ence in peak-to-peak values forthe largest and smallest packetsof the distorted signal. The difference is then expressed as a percentage of the larger number.For example, the differentialgain in Figure 5-21 is about 20%.This is a fairly large distortion,used for purposes of illustration.

You can also use a vectorscopeto check differential gain. On thevectorscope, the modulatedstaircase will appear as shownin Figure 5-22.

To evaluate differential gain,adjust the vectorscope’s variablegain so that the dot representingthe staircase chrominance touches the graticule circle(Figure 5-22). (If undistorted, the staircase chrominance willoverlay the burst vector — theyare each 40 IRE p-p at referencephase at the generator.) Any horizontal elongation of the dotis due to differential gain.

5-9

Figure 5-20. The waveform monitor ’s chrominance filter eliminates varying luminance levels from the modulated staircase or ramp signal.Theflat top and bottom of this chrominance signal indicate an absence ofdifferential gain.

Figure 5-21. This waveform monitor display of a chrominance-filtered modulated staircase shows top and bottom distortions representing a differential gain of about 20%.

Figure 5-22. This vectorscope display of a modulated staircase test signalshows little or no differential gain. The vectorscope’s variable gain has beenincreased for this measurement.

Page 29: NTSC Video Measurements The Basics - Educypedia

5-10

Figure 5-23. Elongation of the chrominance vector dot indicates about 15%differential gain.

Figure 5-24. Widening of the chrominance vector dot along the graticule circumference indicates 5 degrees of differential phase.

Figure 5-25. This display shows simultaneous differential gain and phase ofabout 10% and 9 degrees.

amounts of distortion. But formore precise measurements,especially on smaller amounts ofdistortion, you’ll need a moresophisticated measurement toolwith a special DIFF GAIN mode.The Tektronix 1780R VideoMeasurement Set, for example,provides a DIFF GAIN mode.

Differential gain, like other nonlinear distortions, cannot becorrected by a simple front paneladjustment. Internal adjustmentsor repairs are necessary to corre c texcessive amounts of diff gain.

Tektronix has produced a video-tape that provides further detailson differential gain and how tomeasure it. You can obtain thisvideotape by ordering TektronixTelevision Division MeasurementSeries #1, “Differential Gain”(068-0330-04, NTSC VHS).

Differential phase

Measuring differential phased e t e rmines whether chro m i n a n c ephase is affected by luminancelevel. Differential phase, alsoreferred to as diff phase or dP,occurs when the video systemfails to process chrominanceconsistently at all luminancelevels. This is similar to differ-ential gain, except luminancelevel changes affect chrominancephase rather than gain.

Differential phase causes thechrominance signal phase toeither lead or lag the desiredphase. Such phase errors causecolors to change hue when picture brightness changes. Forexample, when an object movesfrom sunlight to shadow, itappears to change color. Theincorrect reproduction of coloris most likely to occur in thehigh-luminance portions of the picture.

Differential phase can be testedusing the same five- or ten-stepmodulated staircase signal usedfor differential gain testing. Fordigital systems, however, themodulated ramp signal is generally preferred.

Measurements are made using avectorscope. As with differentialgain, the vectorscope’s variable

gain is adjusted so the end of the modulated staircase vectortouches the graticule circle. Anywidening of the dot along thegraticule circumference repre-sents differential phase. This canbe measured using the specialgraticule marks provided onmost vectorscopes (Figure 5-24).

Higher resolution measurements,however, need to be done with amore sophisticated vectorscopehaving a DIFF PHASE mode. ADIFF PHASE mode is provided,for example, on the Tektronix1780R Video Measurement Set.The VM700T Video Measure m e n tSet also provides a DG DP modefor automatic measurement ofdifferential gain and phase. Thisis particularly convenient sincedifferential gain and phase oftenoccur simultaneously, as shownin Figure 5-25.

C o rrecting diff e rential phase, liked i ff e rential gain, re q u i res serv i c i n gby a qualified technician.

Tektronix has produced a video-tape that provides further detailson differential phase and how itis measured. To obtain thisvideotape, order TektronixTelevision Division MeasurementSeries #1, “Differential Phase”(068-0331-04, NTSC VHS).

Conclusion

To maintain optimal signal q u a l i t y, regular system testing is amust. The tests described in thisbooklet, while not exhaustive,do detect most of the distortionscommonly responsible for picture quality problems. Moreimportantly, these tests can warnyou of impending problems thatmay not yet be visible as actualpicture impairments.

It’s wise to establish standardtest procedures for each systemand document the results fromeach test. Such test historiesallow you to spot system deteri-oration trends even before thevideo signal violates acceptablelimits. The reward for this effortwill be video productions ofconsistently high technical quality while avoiding costlyreshoots and reediting.

Most vectorscopes have specialgraticule marks, such as shownin Figure 5-23, to help measurethe amount of differential gain.These graticule markings allowyou to measure fairly large

Page 30: NTSC Video Measurements The Basics - Educypedia

Most folks consider them obnox-ious. After all, who hasn’t dozedoff watching late-night televisiononly to be startled by the blare ofthe 400 Hz tone — you know,the one after the tranquility of“The Star Spangled Banner” —accompanying the video test signals used to check systemsafter programming finishes? It’shard to expect viewers to appre-ciate, especially at 4 a.m., thatthose annoying pictures andsounds are for their benefit.

Color bars are the most ubiquitousof all test signals. When not following a TV station sign-off,they are most often perceived asa preprogram place holder —something like a blank electronicfilm leader. But, as technicalpros know, color bars provideboth serious at-a-glance confir-mation of signal path completionand confidence of video signal acceptability.

Merely observing bars on a TVmonitor, however, is a subjectiveassessment. The monitor willdisplay a signal problem, but itreveals little about the source,nature or degree of the signalimpairment. To extract detailedinformation about video signalfitness requires careful examina-tion of one or more test signalson a waveform monitor and/or vectorscope.

Color bars are only one of manytest signals. Test signals areprecisely defined electronic signals that serve as performancemeasurement benchmarks. Thepurpose of each signal is to makeone or more types of video dis-tortion easy to see and quantify.

Testing 1, 2, 3

The principles behind video sig -nal testing are quite simple. Theoutput of a test signal generatoris applied to the input of a pieceof video equipment or distribu-tion system. The test signal, afterpassing through the components,is displayed on a waveformmonitor and vectorscope. Anysignificant change in the signalbecomes fairly conspicuous, andthe type of change gives manyclues about the possible sourceof the problem.

Color bars are handy for settingup such equipment as time basecorrectors, or TBCs. However,measuring the performance,such as the bandwidth of a TBC,requires more specific testing.Periodic measurement of overallsystem performance helps spotminor problems before theygrow into noticeable pictureproblems. The result is consis-tently higher picture quality.

What test signals ensure videosignal quality? Color bars aremost obvious. They monitor ormeasure several amplitude, timing and color parameters.However, no one test signaldefines all of the amplitude andtiming relationships of the NTSCsignal. Signal path impairmentsand system performance areoften best detected with severaltypes of test signals. You’ll findthe most common of these signals in the Summary of VideoTest Signals chart.

The types of signals suited for anapplication depend mainly ontwo factors: the environmentand the video format.

Standard values

Most studio, production andpost facilities use a variety ofequipment and systems. Theserequire a broad correspondingselection of signals for setup,maintenance and calibration.

Calibrating a picture monitorrequires SMPTE color bars and acrosshatch (convergence) signal.Color bars are used to set chro m a ,hue and brightness adjustments.The convergence signal helpsalign the red, green and bluebeams. The multiburst signal isused to check the picturemonitor’s horizontal resolution.

O b s e rving technical values withina system on a waveform monitoror vectorscope requires a genera-tor with test signals such aspulse-and-bar, modulated s t a i rcase, multipulse or multiburst.However, evaluation with eachtest signal is time consuming.

Combination signals, such asNTC 7 Composite, FCC Compositeand NTC 7 Combination, containtwo or more test signals as elements on each video line.Each element of a combinationsignal is basically a narrow ver-sion of regular signals, allowingtwo or more to fit side by sideon a single line of video.

Many generators also producematrix test signals that combinetwo or more different signals ina single field of video. Differentmatrix signals are available fortransmission and system testingand picture monitor setup appli -cations. A matrix typically consists of 40 or more consecu-tive video lines of one signal followed by the same number of lines of the next signal,throughout the active video lines of a field.

Beyond Bars — OtherTests and Test Signals

6-1

Page 31: NTSC Video Measurements The Basics - Educypedia

Like the combination signals,matrix signals speed the processof equipment and system evalua-tion and adjustment. With severalsignal types, you can observemultiple problems (or theabsence of multiple problems)without changing the test signal.

Continuous technical monitoringof a source is enhanced withVertical Interval Test Signal(VITS) insertion. By includingone or two test signals on unusedlines in the vertical interval, youcan make objective judgementsabout the extent of the signaldegradation instantaneously, atany point in the program. MostTV stations and networks useVITS to monitor and evaluatetechnical parameters withoutinterrupting programming.

Once you’ve decided which signals and formats you need,t h e re are still more considerationsfor choosing a test signal genera-tor. Space is always important;usually the smaller the generator’spackage size, the better, pro v i d e dthe unit supplies the signals you need. Depending on thecomplexity of the facility, othervideo gear in the system mayneed to be genlocked. Some testsignal generators eliminate theneed for an extra distributionamplifier (DA) by providingmultiple black burst outputs.

System performance checksrequire only a few other signals.The NTC 7 Composite andCombination signals or a systemtest matrix signal can provide all components necessary forquick checks of system linearity,insertion gain, frequencyresponse, differential phase andgain, chrominance-to-luminancegain and delay, and other short- and long-time distortions.

More than test

Other signals and functions p rovided by test signal generatorscan simplify several routine p roduction tasks. A safe-title/safe-action-area signal helpsoperators position and size thecritical parts of a scene so theydon’t appear somewhere off theedge of the presentation screen.Blacking tapes is a task thatsome generators can reduce tothe push of two buttons. (Youneed only press Record on theVTR and select a countdownsequence on the generator.)

Evaluating the performance of 2-wire (Y/C) or 3-wire analogcomponent systems requires special signals in the appropriatecomponent form, in addition tothe composite signals. Bowtie isan essential signal for 3-wirecomponent systems designedspecifically for precise amplitudeand timing adjustments. An economical alternative to multipletest signal generators is a multi-format generator with signals in NTSC, Y/B-Y/R-Y and Y/C formats.

Cameras usually don’t requireexternal test signals, but in thestudio they must be genlocked tothe system. In smaller facilitiesor Electronic Field Production(EFP) applications, a generatorwith multiple black burst outputsmight eliminate the need for aseparate master sync generatoror DA. A black burst input to ap roduction switcher is commonlyused as the source of black for afade-to-black.

Distribution distortions

Distribution paths can be subjectto myriad distortions. A distribu-tion path may be as simple as acamera output looped through amonitor to a switcher and termi-nated, or as complicated as a1,000-machine duplication system. That’s why multipurposecombination signals, such as theFCC and NTC 7 Composite andthe NTC 7 Combination, areso widely used as VITS in transmission testing.

Multiburst (which is part of theNTC 7 Combination signal), multipulse and (sinX)/X are allused extensively for frequency-response testing of various distribution systems. Multipulseand (sinX)/X indicate groupdelay as well, but (sinX)/Xrequires a spectrum analyzer orautomated video measurementset for display. Big ticket items,such as spectrum analyzers orautomated measurement systems,often aren’t a liability, becausemany other standard transmissionand distribution tests requirethese instruments. To optimizeperformance, particularly in thisera of higher-resolution formats,good test equipment is a basicrequirement, not an option.

6-2

Page 32: NTSC Video Measurements The Basics - Educypedia

Maintaining performance

Maintenance areas have somespecial requirements. Flexibilityis the key, as the proliferation ofinterconnect and recording formats continues.

It’s not uncommon to find com-posite NTSC gear in use withcomponent analog or Y/C gear.Although all of this gear hasNTSC inputs and outputs, equipment in each video formatrequires test signals in the sameformat to fully exercise its circuitry. In addition to the luminance and color differencesignals primarily used with com-ponent analog video equipment,the time-compressed versions(CTCM or CTDM, the actualrecording formats used) mustalso be available.

Serial digital video is findingfavor as an interconnect formatwithin many video facilities.With it comes the need for yetanother test signal format and, ofcourse, new testing issues. Serialdigital requires the same analogtest signals — you simply needto convert them to the new digital format and add a numberof very complex digital goodiesto the vertical interval.

Operational equipment in needof routine maintenance or troubleshooting can’t always goto the shop. So, with the propersignals and formats, a small,lightweight package can make ita lot easier to bring the tools tothe problem.

The economics of testing

One thing is certain today —viewers expect high video quality,all the time.

Gone are the days when a single,snowy TV channel was reveredas a miracle of modern technologyand occasional problems fromtechnical difficulties were toler-ated. Today, home viewers fixtechnical difficulties on theirown without leaving their armchairs by switching channelswith remotes. Viewers of corpo-rate and other nonbroadcastvideo have similar expectationsand reactions, except instead oftuning out with a remote contro l ,they tune out their brains.

Reliable, high-quality, uninter-rupted video — the type clientspay for — requires extensivetesting and preventative mainte-nance to catch and fix decliningsystem performance beforevisible picture impairmentoccurs. Once quality starts toslip, so will the attention of your viewers, as well as futurebusiness from your clients.

P roducing video without adequatetest and measurement gear is asrisky as driving a car at nightwithout headlights. Today’sdigital-based test signal generatorsprovide many high-precision signals in small, economicalpackages. Waveform monitorsand vectorscopes come in a variety of affordable packages,and sometimes both functionsare combined for even greatereconomy. With compact, easy-to-use gear, the serious videoproducer can’t afford not to usetest equipment every day.

6-3

Page 33: NTSC Video Measurements The Basics - Educypedia

Summary of Video Test SignalsTest Signal Signal Shape Use and Benefits

Color Bars General amplitude and timing measurements. Most widely available test signal. Used in all aspects ofsystem setup and testing from ENG/EFP units to the broadcast transmitter.

Black Burst Commonly used for synchronizing video gear. Also used for noise measurements.

Multiburst Contains packets of six different frequencies. Used for basic frequency-response checks of equipmentand distribution paths in ENG/EFP and studio work.

Modulated Staircase Available in 5- and 10-step forms. Tests differential gain/phase and luminance linearity. Used inENG/EFP, studio and distribution.

Pulse and Bar Used for amplitude, timing, and distortion measurements. Modulated pulse portion tests chrominance-to-luminance gain and delay. Used in ENG/EFP, studio and distribution.

Multipulse Contains pulses modulated at different frequencies for comprehensive measurement of amplitude and group delay errors over the video baseband. Especially important for transmitter testing.

(Sin x)/x Provides frequency-response and group-delay test coverage of all baseband frequencies. Can be used as a VIT signal, making it ideal for in-service transmitter testing.

NTC 7 Combination Combines multiburst and modulated pedestal for frequency-response and distortion tests. Designed for distribution and transmission system testing.

NTC 7 Composite Contains various signal elements allowing amplitude, phase and some distortion measurements. Designed for studio and distribution testing. Rise time too fast for broadcast transmitter use.

FCC Composite Offers the same uses and benefits as the NTC 7 Composite signal. Its slower rise time makes it appropriate for VITS use with broadcast transmitters.

Modulated Ramp Used the same as modulated staircase but provides finer-grained results.

Sweep Provides a continuous sweep of video baseband frequencies, usually with embedded 1 MHz markers. Used for detailed frequency-response testing, but not VITS compatible.

Bowtie Component analog video (CAV) test signal used for high-precision measurement of component channel gain and delay inequalities.

6-4

Page 34: NTSC Video Measurements The Basics - Educypedia

Monochrome Video

The information to reconstruct a black and white image is rela-tively easy to understand —each point in the image isscanned in the familiar line pattern and at each point alongthe line the camera produces avoltage representing the amountof light at that point. The rapidsuccession of points scanned —and the corresponding successionof signal voltages — creates thevideo waveform. This waveformis “one dimensional,” i.e., onlyone piece of information is neededfor each point in the image.

Color Video — History

The situation gets quite a bitmore complex with color video.The problem with color video isit’s not one dimensional, butthree dimensional — threepieces of information are neededfor each point in the image.

There are several ways to conveyinformation about color. Mosttake advantage of a characteristicof human vision known as thetristimulus theory — we can satisfactorily reproduce a widerange of color using only threecolors of light mixed in certainways. Color photography andcolor printing are based on thesesame ideas — three filters, threedyes, three inks, etc. (The four-color printing process usesblack ink in addition to makedark or gray areas in the imageeasier to control.)

While other combinations arepossible, there are some practicalreasons for choosing Red, Greenand Blue (RGB) as the three “ p r i m a ry” colors for a color videosystem. In fact, the earliest colortelevision systems used three

The Color Bars Signal —

cameras, scanning in parallel,each with a Red, Green or Bluefilter over the lens to producecolor video. The display alsoused multiple images, filtersand optics to combine the Red,Green and Blue images into onefull-color picture.

The Color Bars Signal — RGB

Setting up a system with threecameras, three video transmissionpaths, and three displays is achore! Matching the three channels to “track” from black topeak brightness is especiallyi m p o rtant. Test signals have beendeveloped for this sole purpose.In this RGB system, each videopath is normalized so the signalsmatch at black level (zero light)and at peak level (maximumlight). A good test signal consistsof a square wave with black andpeak levels in each of the threechannels. If the square waves aretimed diff e rently in each channel,all combinations of high and low levels in the channels areexercised (Figure 7-1). Sincethere are three channels, eachwith two possible levels, therea re two cubed (eight) possibilities.These eight combinations exerc i s ethe primary colors individually(Red, Green, and Blue), the secondary colors (combinationsof two primaries: Yellow, Cyan,and Magenta), White, and Black.The colors (color bars) arearranged in order of decreasingtotal brightness: White, Yellow,Cyan, Green, Magenta, Red, Blue,Black. The color bars signal isuseful in setting up a RGB-colorvideo system, and it gives apleasing image that also conveyssome indication of the “health”of the color video system.

7-1

100

0

G B R

Figure 7-1. Video waveforms of the three channels for RGB-color bars(100% bars).

Encoded color — NTSC

The RGB system, with parallelfull-bandwidth video in threechannels, makes excellent pictures but has several practicaldisadvantages. Perhaps the most severe is the very widebandwidth needed in broadcastservice or over other long interconnect paths.

There have been severalattempts at “compressing” colorvideo information so it is morepractical to record, interconnect,and broadcast. The winner, byf a r, is the NTSC system developedin the early 1950s. This “encoding” scheme manages toput the essential parts of thethree-dimensional color signalinto the spectrum needed for amonochrome signal — putting itall in one path. (Work is stillcontinuing today with complexdigital filters, advanced modula-tion techniques and “smart”compression algorithms, butthese are beyond the scope ofthis tutorial.)

Page 35: NTSC Video Measurements The Basics - Educypedia

7-2

NTSC must still maintain threeinformation paths, inherent inthe nature of color. The specificset of three signals, however, ischanged from RGB to a moremanageable set. The three NTSCsignals are Luminance, and twoColor Difference components.

The symbol used for luminanceis “Y.” It is closely related tomovement in the “y” directionin a common x vs y graphicalre p resentation of color. The colordifference signals also contain Yand one of the primary colors —usually B-Y and R-Y. For example,B-Y is the Blue video signalminus the luminance signal.

Luminance

The NTSC luminance componentis very much like the signal form o n o c h rome video — containinginformation about the amount oflight in each element of the picture. This is one of the keyreasons the NTSC system is“compatible” with older monochrome equipment —monochrome sets simply ignorethe color information and displayusing luminance only.

Luminance is derived from RGBcolor signals. While the tristimu-lus theory re q u i res three diff e re n tcolored lights, it does not implythat each light contributes thesame amount to our perceptionof brightness of a point in theimage. In fact, for the particularcolors chosen (a certain Green, acertain Blue, and a certain Red),Green contributes about 59% ofour perception of brightness in awhite part of the image. The Redlight contributes about 30%, andBlue only about 11% of thebrightness of white. These numbers are called the luminancec o e fficients of the primary colors.(A certain color of white must bechosen as well, but that and themath involved are beyond ourscope here.)

The luminance signal is pro d u c e dby combining (adding) the videosignals from each primary color

channel, taking into account theluminance coefficients of eachprimary, i.e., Y = 0.59G + 0.30R+ 0.11B. If the input signal forthis process is a RGB color barssignal, the resulting luminancevideo waveform is the familiardecreasing “staircase” shown inFigure 7-2.

Note the steps in this staircaseare not equal amplitude — theydepend on the luminance coeffi -cients. Consider, for example,the Green to Magenta transition(in the center of the bars wave-form). The luminance level forGreen is 59%. Magenta consistsof Blue and Red only, its lumi-nance value is 0.11 + 0.30 = 0.41or 41%. The amplitudes of the other levels may be similarly calculated.

Color difference signals — B-Y and R-Y

Once the luminance signal isavailable, it takes only a littlemore processing to derive thecolor difference signals. Forexample, if luminance is sub-tracted from the B component ofthe RGB video the result is B-Y.This process, and the parallelone that produces R-Y, are fairlyeasy to accomplish with electro n i ccircuits. (Sometimes, another setof color difference signals wereused — I and Q. We’ll discuss Iand Q a little later.)

With the color information nowin the form of luminance (Y) and color difference components(B-Y and R-Y) we still have three signals and the associatedbandwidth and interconnectproblems. A bit more processingis needed to get it all into the“‘one-wire” NTSC signal format.

Encoded color (NTSC)Another characteristic of humanvision is we can’t see fine detailnearly as well for changes in coloring as we can for changes inluminance. In other words, thepicture won’t suffer very much ifwe reduce the bandwidth of thecoloring components, provided

Figure 7-2. Color bars luminance signal (100% bars).

Figure 7-3. Unmodulated and modulated color difference signals and modulated chrominance for color bars.

100

0

Page 36: NTSC Video Measurements The Basics - Educypedia

7-3

we can maintain essentially fullbandwidth of the luminance sig-nal. In fact, this is a good reasonfor developing color differencecomponents in the first place.

Even a full bandwidth luminancesignal doesn’t have very muchenergy in the upper end of itss p e c t rum — the higher fre q u e n c ysignals are quite a bit loweramplitude almost all the time.These two facts (less bandwidthre q u i red for the color inform a t i o nand some “room” available inthe luminance spectrum) allowthe NTSC system to place thecolor components in only the upper portion of the luminance spectrum.

Color subcarriers

The technique for getting theR-Y and B-Y signals moved upin frequency is amplitude modu-lation — a carrier (subcarrier) ismodulated by each of the colordifference signals. Remember,there are two different signals,so two carriers are needed. Bothsignals are at the same frequency(about 3.58 MHz) but are main-tained in phase quadrature(90 degrees apart) so the B-Y and R-Y information can be keptseparated. This process is knownas Quadrature AmplitudeModulation (QAM). Balancedmodulators are also used so thec a rriers themselves are suppre s s e dat the modulator output. Onlythe sideband energy generated in the modulation process issaved. This modulated color difference information is usuallycalled chrominance.

The NTSC composite signal

When luminance and chromi-nance are combined, the result is the NTSC composite color signal. Figure 7-3 shows the B-Yand R-Y signals for color barsand the resultant envelope at theoutput of each modulator, as wellas the combined chrominanceenvelope containing both B-Yand R-Y information.

Some adjustment of amplitudesis needed when the luminanceand chrominance signals arecombined. Figure 7-4 shows the result if the signals for full-amplitude bars are combinedwithout these adjustments. Notethe peak level of the chro m i n a n c eis very much above 100% whiteluminance. This signal will beseriously distorted in passingt h rough a video system, especiallya system originally designed forhandling monochrome signals.In Figure 7-5, the level of B-Y isreduced by a factor of 2.03 andR-Y level is reduced by a factorof 1.14. These specific factorsare chosen to limit the peak ofthe waveform to 100% whitelevel with a 75% amplitude barssignal (Figure 7-6). This 75%bars signal has become the most common for testing NTSCsystems. (Refer to the discussionof various bars signals onpages 2-2 and 6-1.)

Conclusion

There are several aspects of thegeneration of NTSC video signalsthat underlie recommendedmonitoring and test methods.

1.Luminance and Chrominanceare a “matched set” and mustremain balanced in amplitudeand timing in order to accu-rately reconstruct the RGB signal (and the color picture).

2.The color difference compo-nents are only kept separateby differences in subcarrierphase. Small phase distortionscan create large color distor-tions in the end picture.

3. C h rominance information is allat the upper end of the videospectrum and maintaining frequency response flatnesst h roughout the system is critical.

The color bars signal, whether inRGB or NTSC form, offers aknown, accurate signal that exercises the practical limits ofthe color video system.

Figure 7-4. Full amplitude bars without gain adjustments.

Figure 7-5. Full amplitude bars.

Figure 7-6. Corrected 75% bars signal (with 100% white).

Page 37: NTSC Video Measurements The Basics - Educypedia

Glossary

audio signals — XLR connectorsprovide dual-channel audio sig-nals. The left channel can be setto click as a means of easily dis-tinguishing the left channel fromthe right channel in audio tests.

average picture level (APL) —The average signal level withrespect to blanking during theactive picture time. APL isexpressed as a percentage of thedifference between the blankingand reference white levels.

chrominance — The color i n f o r-mation in a television picture .Chrominance affects two proper-ties of color: hue and saturation.Also called chroma.

chrominance-to-burst phase —The difference between theexpected phase and the actualphase of the chrominance portion of the video signal relative to burst phase.

chrominance-to-luminancedelay — The difference in timethat it takes for the chrominanceportion of the video signal topass through a system relative tothe time it takes for the luminanceportion. Also called relativechroma time.

chrominance-to-luminancegain — The difference betweenthe gain of the chrominance portion of the video signal andthe gain of the luminance port i o nas they pass through a system.

color burst — The burst of colorsubcarrier added to the backporch of the composite videosignal. It serves as a frequencyand phase reference for thechrominance signal.

composite video — A singlevideo signal containing all of the necessary information toreproduce a color picture.

c o n v e rgence — Used for adjustingc o n v e rgence of the green, blue andred beams on picture monitors.

decibel — A logarithmic unitthat expresses the ratio betweena signal and a reference signal.For voltages, dB = 20 log(Vmeasured/Vnominal).

differential gain — Variation inthe gain of the chrominance signal as the luminance signalon which it rides is varied fromblanking to white level.

differential phase — Variation inthe phase of the chrominancesubcarrier as the luminance signal on which it rides is variedfrom blanking to white level.

f requency response — A system’sgain characteristic versus fre-quency. Frequency response isoften stated as a range of signalf requencies over which gain variesby less than a specified amount.

graticule — The calibrated scalefor quantifying information on aw a v e f o rm monitor or vectorscopescreen. The graticule can be silk-screened onto the CRT face plate(internal graticule), silk-screenedonto a piece of glass or plasticthat fits in front of the CRT(external graticule), or it can beelectronically generated as partof the display.

horizontal blanking interval —(See artwork below.)

insertion gain — The gain (orloss) in overall signal amplitudeintroduced by a piece of equip-ment in the signal path. Insert i o ngain is expressed as a percent(Vout - Vin)/Vin x 100.

8-1

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 8-1. Horizontal blanking interval.

Page 38: NTSC Video Measurements The Basics - Educypedia

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) 701 77-261; Belgium 32 (2) 725-96-10;Brazil and South America 55 (11) 3741 8360; Canada 1 (800) 661-5625; Denmark 45 (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) 722-9622; United Kingdom & Eire 44 (1628) 403300; USA 1 (800)426-2200

7/97 FL4996/XBS 25W–7247–1

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 registered trademarks of Tektronix, Inc. All other trade names referenced are the service marks, trademarks or registered trademarks of their respective companies.