adequate.As is only to be expectedwith such a reasonably small-valuecapacitor of course. priced unit, there's no bar graph. I feel that Cirkit has come up with a range of real The meter performed beautifully on test. It was very winners with thesemeters. For the price the TM175 offers easyto use consideringthe number of ranges.To start with exceptional quality and features. You can't expect the we were rather dubious about the frequency counter, but build quality of more expensive machines such as those when it was checkedagainst our Racal unit we found that it produced by Fluke, but the test unit gave no cause for had a faster responseand a perfectly acceptableaccuracy. concern on this score. I wouldn't hesitate to recommend Having a capacitor tester is a real boon. Previouslythe best the unit to anyone in the service industry. My thanks to we could do was to carry out a resistancecheck to see a Cirkit Distribution, Park Lane. Broxbourne. Herts EN10 capacitor charge. This doesn't tell you much with a 7NQ for providing the test unit. NICAM Digital StereoSound Part 1 Eugene Trundle

For many years sound was the "poor relation" in what goes into the microphone and comes out of the . Most TV sets had a tiny, often side-mounted loudspeaker is a sound pressure wave. Conversion to loudspeaker which was driven by a barely adequate digital form, by quantisation (sampling the audio signal at . The situation remains much the samewith many regular intervals and assigning to each sample a digital TV designs. But the advent of hi-fi VCRs and the value), is done to overcome deficienciesin the intervening integration of audio and equipment has in recent systemsand equipment. When reproduced, the quality of years led to a great deal more attention being paid to digital sound, and particularly its signal-to- ratio, "with-picture sound". The f.m. TV sound system that largely dependson the number of quantisationlevels used. came into use with 625-line transmissionshas served us The ultimate, as used in the CD and R-DAT systems,is 16 well, providingquality soundto which receivingequipment bits. giving over 65,(X)0values each representing a signal seldom does justice. But it's incapable of carrying stereo. amplitude level. A 16-bit signal would require a wideband and various other shortcomings become noticeable when channel for transmissionhowever. As there is not room to it's processedby high-quality receivers, recorders. ampli- accommodatesuch a signal in the broadcastTV bands the fiers and loudspeakers. Nicam system uses l0-bit words. Ordinarily, these would Stereo TV sound has been available in W. Germany and provide a signal-to-noseratio of about 60dB. An improve- Japan for some years. The systemsused in these countries ment is achievedby continuously "moving the baseline" as employ analogue modulation of an audio or a it were. This gets close to the 84dB signal-to-noiseratio second carrier by a stereo-differencesignal - the systems provided by a 14-bit system. Just how this is done will be are in many ways similar to the technique used for explainedlater. v.h.f.-f.m. radio broadcasting.Encoding and modulation The 32kHz sampling rate used by the Nicam system systemsof this type don't fit easily alongsidea video signal provides a frequency responseof 0-15kHz in two channels in the tightly packed broadcast bands however, and many for stereo. Separationbetween the two channelsis total, so compromlsesare necessary. that dual-languagesound tracks can be broadcastwithout Research and development by the BBC led to the crosstalk.The Nicam systemcan carry data in one or both publication in 1986 of a specification for a radically channelsand is completely separatefrom and independent different TV sound system for broadcasting. It takes full of the existing f.m. sound channel at 6MHz (system I). advantage of modern digital technology and exploits the Before it was adopted the Nicam system had to be tested ability of chip manufacturers to mass-producevery com- under many different sets of conditions. It proved to be plex digital i.c.s at low unit costs.ln short, it representsthe extremely rugged, providing high-fidelity stereo sound current "state of the art". Setmaker Ferguson Ltd. did a even when the signal is so weak that the picture can hardly great deal of decoder designwork, which culminated in the be locked. Texas Instruments' single-chip demultiplexer that's cur- rently used in many TV and VCR Nicam designs. The System Overuiew system is officially known as NICAM 728 - Near Instan- taneously Companded Audio Multiplex, the 128 referring Many of the processesand techniques that the Nicam to the data rate. system uses are common to those employed by the compact disc system. Thus readers who are familiar with the CD system - primed, perhaps, by Joe Cicszynski's Digital Sound excellent seriesof articles in this magazine (March 1989- July 1990) - will have little trouble in understanding and Modern sound systemsare based on the use of digital servicingNicam equipment. Fig. 1 provides an overall view technologv. The first example that came along in the of the system,which we'll describein outline before going domesticelectronics field was the compact disc (CD) which into detail. was introduced by Philips and Sony in 1982. Pulse-code The studio sound feeds for the left and right channels are modulated (PCM) sound was first introduced in the video first pre-emphasisedand passedto an analogue-to-digital field with the Video-S format. We will soon have more converter which produces t4-bit audio samples("words") record and playback technology in the forms at a sampling rate of 32kHz. Thcse samplesare then t'ed to a of VHS-PCM and R-DAT. compressor which reduces the bit rate to 10 per word, Audio signalsmust start and end in analogueform, i.e. effectively extracting the ten most useful bits and discard-

846 TELEVISIONSEPTEMBER 1990 ----

I S l l r I ) )

Tone Bal ance Stereo wide

Fig. 1: Block diagram of the Nicam system, (a) at the studio/transmitter end, (b) at the receiver end.

ing the rest. At this stage an error bit is added for error It remains to convert the data back to analogue form, checking and correction in the receiver. In the following using a DA convertor, then to filter the result to obtain the interleaver stage the bits are shuffled like a pack of cards. original sound signals.User controls are incorporated into This is done to distribute the effects of propagation errors the audio , whose design is critical if the low or interference when the original order is restored at the and high signal-to-noiseratio of the transmitted receiving end. Scrambling is then done by adding to the signal are to be maintained. Thc performance, rating and interleaved data an effectively random data stream, so that positioning of the loudspeakers are also crucial to the the sidebands of the transmitted Nicam signal have no realisation of the full potential of Nicam sound. strong, fixed-frequency components that could cause interference. Modulation comes next. For system I the Nicam An lncongruity subcarrier is at 6.552MH2. It's phase modulated by the data pulses, using the DQPSK system that we'll describe The incongruity with many types of programme of later. With digital signalsphase modulation is economical havingthis wide, high and full-bodiedsound alongside the of bandwidth, a factor that's a key to being able to squeeze necessarilylimited picture area is somethingwe won't yet another signal into the tightly-packed 8MHz slot dwell upon here. The systemcomes into its own with available for a system I TV channel. The Nicam carrier is concertsof any type, and with any Nicamprogramme the added to the 6MFIz f.m. sound carrier fed to the larger the screenthe better. In one jump the TV sound transmitter, where it joins the vision carrier (at a level of systemhas got itself severalstrides ahead of the picture -20dB relative to the peak vision carrier level) on its way transmissionand displaycapabilities. to the transmitting aerial. At the receiver the tuner converts the Nicam carrier Sysfem in Detail to an i.f. of 32.95MH2. This is demodulated and selected by a filter for application to the first section of the Nicam Although it's not essentialfrom the servicingpoint of decoder, the DQPSK demodulator. The Nicam signal view to understandwhat goes on in the studio and the emerges from this as a data stream. It then undergoes a encodingand even the decodingprocesses, a reasonable series of processes that reverse those carried out at the knowledgeof how the systemworks doeshelp with fault broadcasting end. First comes a descrambler which sub- diagnosisand Nicam decodersetting up. Nicam decoders tracts from the data stream the random data added at the are now becominga common feature in top-of-the-range transmitter to smooth out the sidebands.De-interleaving, TV setsand VCRs. in accordance with a simple codebook held in memory. then takes place. As a result the order of the original 10 + Pre-emphasis 1 bit words is restored. The words are next expanded to 14-bit form in accordance with a scale factor transmitted First, pre-emphasis.No soundor visiontransmission or with the data. Error correction on the basis of the parity recording system is complete without a pre-emphasis bits is then undertaken. At the end of this decodingprocess system,which givesa boost to the high-frequencycompo- we have a real-time data stream of 14-bitwords, containing nents of the signal. During reproduction the balance is information for both the left and risht sound channels. restoredby reducingthe amplitudeof the higher frequen- cies:noise is reducedin proportion.The pre-emphasiVde- emphasischaracteristic follows the CCITT J17 recommen- - emDhas r5 dation,with a boostof 6.5dB at 800H2.Fig. 2 showsthe characteristiccurves.

Sampling and Auantisation Fig. 2: The CCITTJ17 emph- asis characteristic. The samplingrate for DA conversionis 32kFIz, which Pre - emDhasis means that digitised samples of the original analogue soundwaveforn occurat 31'25gsecintervals. To prevent o o'2 o4 Bzrsl + Frequ€ncy kHz aliasing(this occurs when the lower sidebandsof the

TELEVISIONSEPTEMBER 1990 847 M LSB Scale Coding Prolection Table 1: Scale-factorsignalling. raclor range range -), i' XXXXXXX '.x'/ x//x /x ltt I 1 Coding range Protection range Scale-factor code XXXXXXX X\X '/x/,x tro 2 2 lXXXXXX \Z*, r0l 3 3 1 1 111 ]X XXX XX Xd 0ll t L 2 2 110 ? ? l XXX X XX XX 100 5 5 101 4 4 011 0 XXX X xx XX 010 5 6 5 100 0 0xx X XX XX l5'l I 6 6 1000 010 0 OXX XX XX R lot 7 001 ) I l lXX X XX XX l00i I 5 7 000 I txx XX XX -15'l I 0xx X XX 0r0 5 6 eachword. It's usedin the decoderas a truth checkon the 0 XXX XX 100 5 5 word'ssix mostsignificant bits. The systemused is referred X xxx XX ,.fi OI] L L to as evenparity. xxx X xx l0r 3 3 We now have a protectedll-bit word. We must next XX XXX lr0 2 2 sendthe decodera scalefactor to indicatewhich bits have XX xxx X lll I r been deleted during the compressionprocess. If the removedin comprersionprocess X = lrrelevanl brls missingbits left-hand in the W7)= 1D476I are on the side Fig. 3 scale "nt factorwill enablethem to be recreated:if they are on the Fig. 3: The digital compression system - 14 to 10 bits. The bits in the shaded areas are deleted during transmmission: right-handside they are lost forever,but this is of no great those in the wedge on the left can be reconstituted at the momentsince the wordsof which they form a part are not receiving end. crucial onesfrom the noisepoint of view. A carefullook at Fig. 3 showsthat 10-bitresolution is sampling frequency overlap with the upper end of the given to large signals(coding range 1), 1l-bit resolution audio frequency spectrum), and the distortion to which this appliesin range2,I2-bit in range3 and so throughto the gives rise, the upper frequency limit of the audio signal 14-bit range 5 which representsthe quietestand most must be held at or below half the sampling frequency, i.e. vulnerablesounds. This "progressivecoding" systemis at or below 16kHz. For this purpose a filter with a very veryeffective, giving bandwidth economy while losing little sharp cut-off at l5kFlz is inserted in the L and R channel in comparisonto a linearcoding system. feeds prior to the sampling gates. For scale-factorsignalling purposes the 1l-bit (10 + 1 The 14 bits per sample used give a total of 16,238 paritybit) wordsare grouped together in blocksof 32,each possible sound signal levels. L and R signal sampling is blocklasting for lmsec.With eachblock a 3-bitcode word carried out simultaneously.A samplescarry L signalsand is sentto indicatethe scalefactor. You will at oncesee that the B samplescarry R signals. there's a degreeof inaccuracyin this. Since only one scale-factormassage is sentfor 32consecutive words, some Digital Compression wordswill not receiveoptimum expansion. In factwhat the scalefactor tells the decoderis the magnitudeof the largest The 14-bit rate is reduced to 10 bits to conserve samplein each block. This approximation increasesthe bandwidth. The significanceof each bit in a word depends quantisationnoise in the final audiosignal, but this noise very much on the sound level represented by that alwaysoccurs where the busynessof the signaldisguises or particular word. The least significant bit in a word can masksit. The scale-factordata occun sufficientlyfrequent- influence the final signal by only one part in 16,000odd. ly for the fastest perceptible loudnesstransitions to be While this may be important in a delicate, quiet passage,it tracked- this is the key to subjectivenoise reduction. It's matters not one jot when the sample representsa very loud this "running approximation" characteristicof the scale- sound or a rapid transition between two widely-different factor signallingthat givesthe systemthe first two wordsof levels. its name, near-instantaneous.The averageerror is in So we come to Fig. 3, where the possiblecombinations practicevery smalland is imperceptibleto the listener. that could go to make up a 14-bit word are laid out. The Sofar we'vediscussed two of the fiddle-factorsinvolved shaded areas show the coding scheme. The most significant in Nicam coding, bit reduction for compressionand the bit (MSB), the 14th (on the left), passes through the l-in-32 scale-factorsignalling. Now we come to the biggest compressionsystem unchanged. The 13th bit is discardedif fiddle-factor,the way in which the scale-factordata is sent. it's the same as the 14th; the 12th bit is likewise discardedif In effect it's includedfor nothing, using the parity bits in it's the same as the 13th and 14th, and so on with the 12th, the datawords. The scalingfactor is sentby modifyingthe 1lth and 10th bits. When this has been done we are left polaritybits in accordancewith a complextable that's held with words that have between 10 and 14 bits. If a particular in memoryin the transmitterand receiver.The receiver's word has more than 10 bits a sufficient number of bits is decoder extracts the 3-bit scale-factorwords by using removed at the leastsignificant bit (LSB) end to reduce the what's called majority-decisionlogic - this processalso word to 10 bits. The digital signal has thus been restoresthe original parity pattern.

compressed, and as long as the decoder knows what the Read ordGr+ compressoris up to it can expand the signal back to 14-bit length at the receiver. Fig. 4: The Nicam in- terleaving system. For Protection and Scale Factor protection, bits that are initially adjacent are scat- oader It's necessaryto add to each digital word some tered. protection againstthe possibility of transmission/reception errors. For this purpose a parity bit is added to the end of

848 TELEVISIONSEPTEMBER 1990 728 bits = 'lms

ossignmenlof sixly-foull1-bit soundt words (01 to D6a) to A ond I chonnels

bitr

frome olignmenl oddit ionol dofo wor6

t -'-reserve f rome ,,.r-*W sound controlbits swiichingflog l.s.b m.s.bti I

porityI Fig. 5: The Nicam frame format, showing the arrangement of digital data for a stereo sound signal before interleaving. Diagram reproduced by courtesy of the BBC.

The scale-factor code words are shown to the right in amongstseveral words the damageto eachis minor. It can Fig. 3. There are five different compressionpatterns, each be repairedby parity correctionand/or error concealment being assigneda range figure from 1 to -5.Table 1 shows as necessary,techniques that are universallyused with how the three-bit code words convey scale factor and PCM soundsystems. protection data. The three scale-factorbits can give eight combinations.As shown, some of the "spares" are used to Housekeeping signal to the decoder the amount of error-protection required by each sample. This can be used in the receiver The digital signalnow hasenough protection to enableit to give extra protection to the most significant bits of the to withstandthe rigoursof transmission."Housekeeping" samples. data hasto be addednext so that the sectionsof data can be identified and synchronisedand the decodingprocess Interleaving canbe controlled. Fig. 5 showsthe compositionof a broadcastdata frame - The parity systemgives significant protection to the data the transmitteddata is arrangedin 728-bitframes. The words. But impulsive interference and similar corruptive "housekeeping"data comes first, ratherlike the field sync influences could blow large holes in the data stream. So pulsein the video waveform.As the duration of a frame is further protective measuresare required. The degree of lmsec,the datarate is 728KbitVsec.There are of courseno protection called for depends on the likelihood of and gapsbetween frames. probable pattern of data comrption. Nicam data, coming The frame alignmentword (FAW) synchronisesand sets off-air, doesn't require the massive error protection used up the decodingprocess in the receiver.It consistsof eight with the CD, V8-PCM and DAT systems.So far from the bits and is always01001110. Next comethe five bits C0 to intricaciesof CIRC, the interleaving systemused with the C4, which are used for decoder control and switching. Nicam system is simple. Their functionsare as follows. The frame flag bit C0 is set It's achieved by writing the data into memory and then at 1 for eight successiveframes and to 0 for the next eight reading it out non-sequentially, as shown in Fig. 4. The frames. This unchangingpattern defines a 16-frame readout order, which is held in a ROM address-sequencer, sequenceand is usedto synchronisechanges in the type of ensures that bits which were originally adjacent are informationbeing carried. Bits Cl, C2 and C3 provide separatedby at least fifteen other bits. An error burst in applicationcontrol information- seeTable 2. Note that C3 the transmission path can corrupt several consecutive bits remains unchangedthroughout' these options. It's at in the broadcastdata stream. But as the error is distributed present spare but is assignedto future sound and data coding options. The reserve sound switching flag C4 causeddecoder design problems at one stage.It's set to 1 when the Nicam channel carries the same sound as the Table 2: Control bit functions, main (f.m.) soundcarrier and to 0 otherwise.This permits fall-back switching should the Nicam carrier fail or Control bits Contents of 704-bit sound/data block shutdownwhen a datatransmission is in use. Cl C2 C3 The nextdata block, ADO to AD10, is at presentunused - it's reservedfor "additional data" and it will be 0 0 0 Stereosignal with alternateL and R interestingto seewhat this will be. samples 0 1 0 Two independentmono signalsin alternateframes, e.g. dual-language The Sound/data Block sound 1 0 0 One mono soundsignal and one data This initialsection of the trameaccounts for 24 bits.The channelsent in alternateframes remaining704 convey stereo/dual-channelsound data. 1 1 0 One datachannel There are sixty four 11-bit words, the A channel (stereo

TELEVISIONSEPTEMBER 1990 849 left) and B channel (stereo right) samplesbeing sent output sequencelasts for 5II (2e- 1) bits and thus runs alternatelythroughout the period - 32 of.each. Sincetwo through almost one and a half times durins the frame soundchannels, plus the initial data,are being transmitted period. as a singleserial data streamthe bit rate of eachchannel is One of these PRSGs is used at each end of the system, approximatelydoubled. This form of time-compressionis a i.e. at the transmitter and the receiver. Each is reset to comrnon technique known as time division multiplex 111111111at the end of the last bit of the frame alignment (TDM). It's easyto put into effectby writingthe datainto word, which is not scrambled, so that the PRSGs run in a RAM in real time and readingit out at approximately synchronismthroughout the rest of the frame. The PRSG twice the rate at suitableintervals. TDM is usedin many sequenceis added "modulo-two" to the data bit stream on data storage and transmissionsystems to match signal its way to the modulator - modulo-two addition is the densityto channelbandwidth. action of an exclusive-orgate, whosebehaviour is shown in The order of A and B samplesshown in Fig. 5 represents Fig. 8. Fig 9(a) shows the action of the scrambler at the the sequencebefore bit interleaving(see above). Only the transmitter end, working on a data stream that consistsof 704bits of sounddata are interleaved,which is why Fig. 4 010101010tbr the purposesof illustration. Fig.9(b) shows startswith bit 25 andends at bit728. With monauralsound the reciprocal process at the receiver end, after which the the data block is arrangedslightly differently, see Fig. 6. In original data sequence is restored. this case two 32-word blocks (n and n + 1) are placed end-to-endin a singleframe, carryingthe completemono Modulation and Transmission signal.Since the datarate is unchanged,the next frameis available for other forms of data if required. Odd There are various ways of modulating a carrier wave numbered frames carry the mono sound data and even with information. The best known are amplitudemodula- numberedones the "information" data. To indicatethis tion, asused for terrestrialvision broadcasting, m.f. radio, conditioncontrol bits C1-3 changeto 100,switching the etc., andfrequency modulation as used for satellitevision, decoderto the mono mode. TV soundand f,m. radio. A digital signalhas only two If two independentmono signalsMl and M2 (e.g. states.1 and0. In termsof bandwidththe most economical dual-languagesound) are being transmitted,Ml is carried method of modulation tor use with such a signalis phase in the odd-numberedframes as just describedwhile the M2 modulation:the carrier frequencyremains constant but its signalis carried in the same mannerin even-numbered phaseis altered by the modulatingdata. The simplest frames.The controlbits C1.-3change to 010to switchthe phasemodulation system is a two-phaseone, with for decodersound routing. example ones setting the carrier phase at 0" and zeros Data information can take many forms. Computer settingit at 180".This is easyto detectat the receiver. programs or software are perhaps the most likely. If Bandwidthsaving is possibleby usingfour-phase modula- conditionalor restrictedaccess is requiredby the broadcas- tion, in whichthe carrierhas four possiblestates - 0", 90o. ter or his agent,perhaps the sparebits C3 andAD0-10 will 180" and 270". Even more complexphase-shift systems be used to provide this feature. No data information havebeen devised, with eightphases, but the smallerthe format for this purposehas to date been published. phaseshift the more difficultreliable detection becomes. The quadrature phase-shift keying system used for Scrambling Nicam hasfour carrierphase states. At the transmittera serialto two-bit parallelconverter changes the serialdata Now that we've considered the composition of the into a seriesof two-bit pairs. Thesepairs can take one of Nicam data streamin detail it shouldbe obviousthat there four forms - 00.01. 10 or 11. Each of thesemoves the are certainfixed patterns.Since the data streammodulates carrier phaseby a different amount, as shownin Table 3. a carrier whose sideband structure depends on the Thus the carrier phase has four possible states and is characteristicsof the modulation, the energy in the movedfrom one to anotherdepending on the composition sidebandscan be evenedout by scramblingthe data at of eachtwo-bit pair. Note that with a 00 bit pair the carrier random. This is a form of energydispersal. Total random remainsat its 0oshift, i.e. its rest,state. Fig. 10represents scatteringof the bits is not possiblebecause there's no way theseconditions. of descramblingsuch information at the receiver.A simple Since each phase change occurs at two-bit intervals circuit that generatesbinary digits in what appearsto be a insteadof once per bit and the maximumshift is half a randomway but whoseoutput is predictableand repeat- cycleof carrier (either way, in effect), the useof DQPSK able is availablehowever. It's called a pseudo-random greatly reducesthe sidebandspread. The sudden phase sequencegenerator (PRSG) and consistsof a clocked jumps depictedin Fig. 10 can be smoothedout to reduce shift-registerwith one or more taps that feed back to its the sidebandsfurther. With a 2-bit data rate of 346Kbits/sec input. Fig.7shows the one usedwith the Nicamsystem. It and a carrierfrequency of 6.552MH2,there's a minimum consistsof nine stages,with a tap betweenthe fifth and of eighteencarrier cycles between each phase change. By sixthfeeding one input of an exclusive-orgate whose other feedingthe data to the DQPSK modulator via a low-pass input comesfrom the final stage.If you work it out, the filter we can changeit from a squarewave to a triangular sequencestarts 0000 0111 1011 etc. Onceit getsgoing the one. This hasthe effectof "smearingour" the carrierphase changesin time so that they are more evenly distributed Table 3: Nicam carrier phase changes for each along the eighteencycle periods available. possible combination of 2-bit data pairs. With the broadcastsystem I as used in the UK the normal f.m. sound carrier is at 6MHz above the vision -7d8. lnput data Phase of Nicam carrier carrier at a relative level of For use with Nicam transmissionsit's reducedto -10dB. The Nicamcarrier is 00 0' at 6.552MLIZ (exactly nine times the bit rate) and is 01 -90" transmittedat a level of -20dB with respectto the vision 10 -270' carrier. It's important to appreciatethat these levels are 11 -180', quoted as voltages,so that the power ratio of the peak

850 TELEVISIONSEPTEMBER 1990 728 bits= 1ms

ossignmenlof sixty four 1'l- bif sound+ poralytYords (D1to 05a) to compondingblocks n ond n + 1

n32 n +1).1

oddilionoldolo

reserve sound conlrolbils switchingflog porlry

Fig.6: Nicamframe format for two signal transmissions.Compare with Fig.5. The n and n + 1 channelsmay carry dlfferent language sound tracks. Diagram reproduced by courtesY of the BBC.

Oaia 00

QPSK ca rrier

-180' ll Reset pulses derived DI6n {a) {b) Fig. 70: Changing the phase of a carrier to convey data. ln this case there's a phase change with different bit-pairs in the transmitted data. A graphical presentation is shown at (a) while (b) shows the phase shift conditions.

sovnd carrier azFM

Fig. 7: A PRBS generator using a shift register and two / r Dioital sound carrler exclusive-or gates, with a tap betuveen the fifth and sixth t/- I stages of the shift register. ll I f --f--1--AIBIF ) 0l0lo a --+-f5-\ 0lr )) >*' Fiq. 8: Modulo-two addition lr B4/--'/ 11011 - binary data. ; 1lrl0 IDTtsl _,_ - 1 sideband _ gMh2 channel_ l @@ r Fig. 11: Frequencyspectrum of a system ITV channel with S Nicamcarrier. The total spread of the Nicam signal is about 000001r 110 00000111r011 700kHz. etc' 0l010l0l0etc 0l0l0l0l0elc or, with co-channelreception conditions, to French system

0r01110 L transmissions which have an amplitude-modulated etc. sound carrier at 6.5MHz. {a} Transmit (b) Receive l0ZB0-l The Nicam systemhas been adopted by some European (a) the transmitterend, (b) Fig.9: Action of the scrambler, at countriesand many others have expressedinterest. For the the reciprocalprocess in the receiver. Continental system B/G, with the f.m. sound carrier at 5.5MHz, the Nicam carrier frequency is set at 5'85MHz. visioncarrier to the Nicamcarrier is 100:1.Fis. 11 shows This is rather close to the f.m. sound carrier, but can be the spectrumof a systemI TV channelwith Niiam sound. accommodated with critical modification of the data- At a main transmitter the Nicam carrier is fed to the shaping sideband filter. The Nicam signal bandwidth is klystron power amplifier used for the f.m. sound signal. reduced from 700 to 500kHz with system B/G. For optimum efficiency this normally operatesin a very The Nicam format is compatible with the sound and data non-linearmode and is very sharplytuned to the relevant transmission arrangements used with the MAC/packet channel frequency. For correct operation with Nicam system that's been adopted for the BSB satellite TV signalsits conditionshave to be re-tweakedfor more linear transmissions.Thus the chip setsdeveloped for one can be operationand for greaterbandwidth to allow for the extra used for the other. which is useful for both set designers phase-modulatedcarrier. A great deal of trouble is taken, and chip manufacturers. These chip sets have to sort out as we've seen,to keep the Nicam carrier'ssidebands low the rather complex Nicam transmissionsat minimal cost, and quiet. There's no interferenceto the other carriersin which is no mean task. Next month we'll be considerine the channelor to the adjacentsound and vision channels- the receiver and decoder. TELEVISIONSEPTEMBER 1990 851