Home

Model 363 Two-Channel SR/A-Type Processor User`s Manual

Contents

1. lf you are not sure what your Dolby level is read section 2 1 for more information 1 Remove the Model 363 from the signal path by pressing the Bypass button for each chamnel 2 Align the recorder to in house standards and check that recorder and console meters agree or bear a known fixed relationship 3 Restore the Model 363 unit to the signal path by releasing the Bypass buttons and set the SR off A switches on each channel to the center off position B RECORD ALIGNMENT 4 Press the Set up button and set the rec play switches on the Model 363 to the rec position red LEDs on 5 Send a 1 kHz test tone at Dolby level e g 0 VU from the console to the Model 363 6 Place the recorder in record line in or E to E 7 Adjust the rec in trimmers for Dolby level on the calibration displays equal brightness of the green LEDs 8 Adjust the rec out trimmers on the Model 363 for the correct reading e g 0 VU on the tape recorder meters As a quick check you may switch the Model 363 in and out of Bypass at this point and make sure the tape recorder meters don t change C PLAYBACK ALIGNMENT 9 Switch console meters to read the return signal from the tape recorder 10 Set the rec play switches on the Model 363 to play green LEDs on and adjust the play in trimmers for Dolby level on the calibration displays equal brightness of the green LEDs 11 Adjust play out trimmers on the Model 363 for Dolby level e g
2. 9V 9 F P DENOTES FLAME PROOF A1C3766 REV 1 7 1 SECTION 7 UNIT SERVICING IN CASE OF DIFFICULTY 7 1 Isolation of the Problem As a first step switch the unit into Bypass If the problem remains it is very unlikely that the Model 363 is at fault As a final confirmation that the unit is not at fault the XLR connectors at the rear can be unplugged and connected together thus removing the Model 363 totally from the signal path If the problem disappears with the unit bypassed the fault is either in the Dolby processing modules or the motherboard inside the unit In this case exchange the two modules if the fault changes channel then the problem is likely to be in the processing module If the fault is present on both channels it is possible that there is a fault in the power supplies Note that several functions are dependent on the type of module installed For example if the Cat No 450 A type only is inserted switching to SR will not illuminate the red LED and the module will remain in the off mode if the module is suspect it should be checked Testing equipment is available from Dolby Laboratories to verify the performance see Section 7 3 below alternatively refer to Section 7 2 Quick Check for a simple field test If a test extender Dolby Cat No 387 is available then a further check on the interface can be made by connecting together temporarily pins 26 30 34 and 36 Then with th
3. Release the Set up button and set the SR off A switches to the proper NR type for the recording The alignment is now complete J4 To console J3 From rec d2 To rec J1 From console J8 To console J7 From rec J6 To rec J5 From console WARNING Ch1 Check voltage selector and fuse Se rating before applying power No user serviceable parts inside Refer all service to qualified personnel S0nA Dolby Laboratories Inc r2 San Francisco amp London Made in U S A QA MIR mm J9 rec play remote J10 function remote E A2A3802 REV 2 Auto Compare Ref Tape ey Auto Compare DO Dolby moder33 mm DO Dolby SR A tss elo elo Cat No 300 Module E O Coo DooDoo O ODO Do Ree E pla SR off A normal checktape OUt rec gan n out in out in out GD Sct up Ch 1 Bypasss Ch2 normal check tape rec play _ DO Dolby SR A sects tecorsingy Cat No 300 Module COC ID A2A3802 REV 2 Figure 4 1 Front and rear details 5 1 SECTION 5 PRINCIPLES OF NOISE REDUCTION AND SPECTRAL RECORDING A A EWY O OCE E OR ERZE Z DOES 5 1 Introduction The Model 363 may contain Dolby A type noise reduction using the Cat No 450 module Dolby Spectral Recording Cat No 350 or both Cat No 300 There are two other analog Dolby systems B type and C type All employ some principles in common for completeness the following co
4. 0 VU on the console meters As a quick check you may switch the Model 363 in and out of Bypass at this point and make sure the console meters don t change 12 Release the Set up button and make sure the unit is out of Bypass The alignment is now complete D PREPARING FOR A RECORDING 13 Select A or SR on the individual 3 position switches as appropriate 14 Record a section of Dolby tone or noise on tape by putting the recorder into record and pressing the Set up button the the Model 363 unit Remember that Dolby noise is recorded on tape at 15 dB below Dolby level e g 0 VU and so will read 4 7 low on console and recorder meters However this level difference is automatically compensated in the Model 363 calibration display when Set up is selected Under these conditions when using Dolby SR the Cat No 300 or the Cat No 350 modules will go into the Auto Compare mode Auto Compare provides the user with an accurate audible verification that both the tape recorder frequency response and the calibration levels are set properly regardless of the indication shown on the tape recorder meters and Dolby calibration displays Listen for level differences between the pink noise signal coming from the output of the tape and internal Dolby noise generator in the Auto Compare mode The two LEDs on the front of the Cat No 300 and Cat No 350 modules indicate whether the monitors are receiving the internal reference signa
5. pueg Bulpis AAA LN pueg paxI4 pueg paxi4 9B61S 3437 MOT saBels 9437 PIN sabeis 9597 YbIH E L SON y SON 8 ON SHNI syingut gt HUID J011u02 J01u02 031005 UOREJNPOW uone npow uone npolw 41 4H Buimays jeayads Indul euBis 101 J Audio Eng Soc Vol 35 No 3 1987 March DOLBY quency the action of the second circuit is effectively substituted for that of the first The operation of the action substitution compressor can be characterized directly from the above description With an input signal V and an output signal Vo a first compressor transfer function F s and a second com pressor transfer function F gt s each being responsive to the respective applied signal we have Va VilFi s Foals Fis F 1 This equation shows that the overall transfer function is the sum of the individual transfer functions minus their product In other words to the extent that the transfer functions may overlap a factor is subtracted from the sum of the transfer functions The above type of action can be achieved with various circuit topologies the one used in the initial imple mentation of the SR system being shown in Fig 3 In this arrangement the compressor circuits are arranged in a stack Both circuits are fed in parallel and the output is taken from the top circuit which is configured as a three terminal network with terminals a b and c
6. A complementary audio signal encoding and decoding format called spectral recording SR for use in professional magnetic recording and similar applications is described The processing algorithm is highly responsive to the spectral properties of the signal A further characteristic used during encoding deemphasizes high level signal components in the frequency regions usually subject to channel overload The process results in a significant reduction of audible noise and distortion arising in the channel 0 INTRODUCTION In 1980 some 14 years after the introduction of A type noise reduction 1 the author began work on the development of the next generation system for general purpose professional recording and transmission A configuration that would employ the A type charac teristics as part of the new system with switchable compatibility was considered initially However this would not take full advantage of the new technology embodied in the C type system 2 nor would it readily allow the incorporation of some further new concepts Therefore the particular parameters of the A type sys tem were abandoned as a starting point for the new development However the basic principles which appear to be as valid as when they were first introduced were retained the use of a main signal path without any dynamic processing to pass high level signals coupled with a low level side chain compressor to pro vide dynamic action The design
7. A loud signal in one area of the spectrum will mask noise in that same area but the variation in noise in other areas may be audible Compandors of this type are inherently susceptible to this audible variation in noise level called noise modulation Noise modulation can only be eliminated by ensuring that the gains of the encoder and decoder m any particular part of the spectrum are fixed except when signals are present in that part of the 5 2 spectrum at levels above a defined threshold The output noise will then be constant in all parts of the spectrum except those where signals are present to mask the changes in noise A system with this property must be capable of changing gain in any one area of the spectrum without changing at others Clearly in a conventional compandor such as in Figure 5 2 any change in gain occurs at all frequencies equally so that a signal at any one frequency must inevitably alter the noise level at other unmasked frequencies Constant slope compandors have no upper or lower thresholds and hence possess the virtue that there is no need to ensure that the absolute signal levels in the compressor and expander are equal A superficially attractive idea is to use several constant slope compressors each fed by a separate band splitting filter Unfortunately because of the practical limits on filter slopes and the absence of compressor thresholds each band compressor receives signals albeit attenuated from
8. CAT Me 200 ONLY T z I SR 28 un 108028 1C805A IDENT d Zi E D ETER OUT A TYPE 32 isl 2 n OUTPUT xay r o SP mm i ists 1C8048 2M2 5 ENCODE DECODE 12 SE A l R802 3 lag SETUP NORM 19 j v 3 ecl 47K Reog mi ceo E L f METER GAIN SR A TYPE CS W 4 maAasos K an l R804 4 RE 1N 0UT MA e I 10 mabe s St D807 bea Shea EE 100K 6 WM A A 16 SYNC BUS LOGIC REF Dk E E i _AUTO COMPARE 56K 4 Ka i 8807 I w mnie I i 18k i FEO b NORM AUTO COMPARE ip H e 1058 1C804A W E R824 ee LM358 i i 1KS DS805 be l R825 h eo A1G8575 REV 10 Sie LG i i j D5806 i SZ MIR Ned dr d TR E Figure 6 5 Cat No 300 main path circuit jx f 6 15 101 R1 um ee 1 15V 18E r 71 Be R514 2 Di t A TYPE Sek SK6 100P 1N4004 100 Iconpressorl R528 K circuit 26 8 1x BS 10 2 103768 I r 5 gt i 23 GND icons 100K 1x RIS ER 470E I R526 56 97K6 1x C508 aln pera 1N4004 ER 1x 9v 15V QF501 arso2 J175 J175
9. Differential Network Playback Processor Output input Subtractor Fig 5 Basic block diagram of the noise reduction system Differential network shown at right is the same and playback the filters and compressors work under identical conditions both statically and dynamically products to be generated which may or may not be cancelled by reciprocal treatment following transmission With the method under discussion it is not only pos sible to confine overshoots to small values but to use relatively long attack times thereby reducing modulation distortion Referring to the differential network portion of Fig 5 the method used is to follow the compressor circuit linear limiter with a conventional symmetrical clipper nonlinear limiter A suddenly applied signal is thus momentarily passed without attenuation to the clipper the differential component is confined to an amplitude which results in negligible overshoot when added to the main signal In the present system the clipping level has been chosen to limit the overshoot to 2 dB with peak amplitude step inputs The addition of the low amplitude clipped signal to the large amplitude pure main signal results in momen tary distortion of a few percent but the degradation is so small and of such short duration 1 msec or less de pending on the frequency that it is masked by transient components present in the input signal as well as sub jectively atte
10. STEP 5 Insert the signal processing module s into the guides visible through the front panel The modules may be either Cat No 300 Cat No 350 or Cat No 450 The unit will operate with only one channel of signal processing installed or with a mixture of module types Circuitry in the unit frame recognizes which module is in each channel and prevents incorrect indications being given for the channel switches For example if the Cat No 350 which only has SR processing is installed switching to A type will not illuminate the green LED and the module will remain in the off mode STEP 6 Track identification stickers are supplied which can be used to differentiate between multiple units mounted together We suggest these are stuck on the front extrusion to the left of the left module cut out and to the right of the right cut out There is also space on the rear panel to add stickers a double set is supplied with each unit for labelling both back and front NOTE Prior to unit serial number 599 the remote mode selector only selected processing in out and was wired as shown on the next page 2 8 o 7 14 Remote Processing D l In out i 9 out 8 15 9 l 4 111 ps O MA Ramee SE SR green SES Processing ze G oO LED indicators S A type red 6 13 o RNA NOTE for simultaneous operation of Several channels pins 7 and 14
11. This deemphasis is effective not only on the steady state aspects of these signals but on all transient effects as well The output of the low level stage is then added directly For low frequency antisaturation the low fre quency deemphasis is effective on the secondary main path signal including the high level stage outputs The low frequency mid level signal is then added directly The mathematical basis for these arrangements is pro vided in 2 With the final combination of signals in the last adder an SR encoded signal appears at point B The encoded signal can be considered to comprise an unmodified component from the input plus an SR signal which carries all of the SR characteristics Thus the SR com ponent can be derived by subtracting the unmodified input signal at point A from the SR encoder output at point B This provides an SR signal at point C that can be handled and switched the same way as in the A type system This simplifies practical use of the system 3 2 Modulation Control Circuits Fig shows how the inputs of the various modulation control circuits are connected and how the resultant signals are distributed Modulation control signals MC1 MC7 are derived from the output of the second stage adder In this way the modulation control signals begin to have a significant influence at relatively low levels such as at 30 dB because of the contributions of the HLS and MLS stages the phase relationships betwe
12. de la mani re suivante Phase Brun Neutre Bleu Terre Vert Jaune WICHTIGER SICHERHEITSHINWEIS Dieses Ger t entspricht der Sicherheitsnorm EN60065 Das Ger t darf nicht mit F ssigkeiten Spritzwasser usw in Ber hrung kommen stellen Sie keine Gef e z B Kaffeetassen auf das Ger t F r das sichere Funktionieren des Ger tes und zur Unfallverh tung elektrischer Schlag Feuer sind die folgenden Regeln unbedingt einzuhalten o Der Spannungswahler muf auf Ihre Netzspannung eingestellt sein o Die Sicherungen m ssen in Typ und Stromwert mit den Angaben auf dem Ger t bereinstimmen o Die Erdung des Ger tes mu ber eine geerdete Steckdose gew hrleistet sein o Das mitgelieferte Netzkabel mu wie folgt verdrahtet werden Phase braun Nulleiter blau Erde gr n gelb NORME DI SICUREZZA IMPORTANTE Questa apparecchiatura e stata costruita in accordo alle norme di sicurezza EN60065 Il prodotto non deve essere sottoposto a schizzi spruzzi e gocciolamenti e nessun tipo di oggetto riempito con liquidi come ad esempio tazze di caffe deve essere appoggiato sul dispositivo Per una perfetta sicurezza ed al fine di evitare eventuali rischi di scossa elettrica o d incendio vanno osservate le seguenti misure di sicurezza o Assicurarsi che il selettore di cambio tensione sia posizionato sul valore corretto o Assicurarsi che la portata ed il tipo di fusibili siano quelli prescritti dalla casa costruttrice o Lapparecch
13. its purpose is to allow synchronization of a master Dolby noise generator in multi track applications of these modules 6 9 3 Power supplies The modules contain their own regulators and provide stabilized outputs of 9 V and 9 V for the module circuits 6 5 BALANCED INPUT AMPLIFIER RV101 J REC IN K201 REC RELAY TERMINED BY PRESENCE IN FROM CONSOLE lt i QF201 202 CAT No 300 350 450 l e I NR CIRCUITS I I I SR A OFF A UT I 0F801 802 h I SET UP r Se i NORH NORM BYPASS 1C103A l REC NORM ALS SAR ic1038 PLAY 1C201A BALANCED 26 1C803A OUTPUT 2 OUTPUT AMPLIFIER 1C705A i APS REC OUT I REC CAL A BALANCED 120 as I SE RV301 I K401 AMPLIFIER Be DOLBY TONE so n 158098 I PLAY RELAY PLAY CAL n OUTPUT 1 RV401 1C2018 FROM REC BALANCED GROUND IN CAT Nos 300 350 9 SET UP Gu T OUTPUT SR ONLY 110045 PLAY OUT AMPLIFIER nd GROUND IN CAT Nos 300 450 TAPE I SL TO CONSOLE 4 SEC I PINK NOISE z 22 OREF 4 SEC BYPASS qu A E NORM O CHECK O sa NORM l 1C1001 2 greets TAPE POWER L METER OUT REGULATOR gt 0 CA OOOO gt 9v GO SET UP 1ceosA Dg eX SR ONLY 108028 200Hz 4KHz I NOTES I I 1
14. or other variable gain means The dual control circuit arrangement described above is employed to obtain optimal performance under both simple signal single dominant signal and complex signal more than one dominant signal situations The modulation control signal MC3 is optimized in fre quency weighting and amount for simple signal con ditions in which the modulation control action is most useful Under complex signal conditions however the modulation control signal developed becomes larger and the subsequent modulation control action is then greater than necessary that is the de control signal output from the main control circuit is less than required Under this condition the control signal from the pass band circuit is phased in via the maximum selector circuit to control the overall action of the fixed band compressor circuit The output of the fixed band is fed through a buffer with an overall gain of unity to provide the reference for the sliding band filter This is the only signal output of the fixed band compressor circuit The sliding band control signal is derived from the stage output top of Fig 9 The signal is fed through a single pole high pass weighting network about 10 kHz different for each stage and is rectified The rec tified signal is opposed by modulation control signal MCI Since MCI also has a single pole high pass characteristic the ratio between the rectified control signal and MCI monitors th
15. servicing set up side chain skewing spectral specifications spectral recording Skewing stickers identification testers three head recorder transient troubleshooting unbalanced operation voltage voltage selection warble Dolby tone XLR wiring index 2 1 1 4 2 4 4 4 6 5 2 5 5 1 1 7 2 4 2 6 2 2 6 5 1ff 6 3ff 6 8 6 9 6 10 7 1ff 4 3 6 3 4 3 4 4 4 6 6 2 6 3 8 3 5 3 5 4 5 2 3 1 3 5 3 3 3 1 1 3 2 4 6 3 6 3 5 1 7 1 7 4 6 3 1 2 2 4ff 7 1ff 2 6 3 3 4 1ff 6 2 6 3 5 5 5 6 1 1ff 5 6 5 5 2 7 7 2 3 3 6 2 5 2 7 4 1 3 2 3 6 1 1 3 1 4 2 1 2 4 2 5 6 3 7 6 2 1 6 2 2 3
16. signals can be generated without the delay inherent in a low pass filter resulting in lower transient distortion 2 any transient distortion produced by the circuit is attenuated by the 800 Hz low pass filter and 3 noise generated by the gyrator circuit is attenuated by the filter The price to be paid for these advantages 1s the resulting high signal levels that the variable resistances RV and RY must be capable of handling at high fre quencies There is no active attenuation at all at very high frequencies since sufficient passive attenuation and phase shift are provided by the 800 Hz low pass filter Special control arrangements called high mode comprising complementary bootstrapping and control circuit gain boosting enable the fixed and sliding 800 Hz Out 80 Hz L T Fig 10 Low frequency stage block diagram steady state aspects J Audio Eng Soc Vol 35 No 3 1987 March 111 DOLBY band signal circuits to handle the required levels with low distortion and low noise Referring to the fixed band section lower half of Fig 10 the incoming signal is applied directly to the variable gain circuit Control circuit frequency weight ing is provided by cascaded single pole 800 Hz and 1 6 kHz low pass filters The corresponding filters in the high frequency stages are the 800 Hz and 400 Hz filters at the inputs of the circuits The main control circuit rectifies the filtered signal the resulting dc
17. ti rec 122V I r 4 8 I A2C3808 c REV 8 Model 363 Sony APR 5000 series I J9 2 6 I 3 channel ai 3 ZO I lopto iso ator l E e o o 4 8 j play d rec l Is i i use 3 pin male i D connector channel 1 pins 6 2 channel 2 pins 7 3 channel 3 pins 8 4 4253808 37 REV B Sony BVH 200C Modei 363 E 8E 03 3E 04 Audio B as l Amplifier 9 RER behind transporej existing cable __ Add plug socket to AP O6 board y on rear of BVH I CN373 4 BEE 7 A1 3 1 1 A o gt TTL 10 L 270 1 gt _Ae_ 3 2 A H j 27 SES 3 d I A4 4 4 l 5 5 i I i d E i Mase up a T connector as shown to para le existing connection A2C3856 a REV A 2 12 Studer A8C7 d ppc C odel 363 J9 b DA 2 transistor play D o rec dh o 8 note both channels I switch together LA2CS808 d REV B Studer A810 812 820 Mode 363 J9 14 m R lt a Ll transistor play o rec E use Studer connector A9C3809 a REV B part number 20 020 303 08 SECTION 3 APPLICATIONS ee ee OO TC E There are two ways of using the Model 363 in addition to a bypass condition These are 1 Switched operation where the mode is switched between record
18. which have generally been attributed to the differential method are in fact properties of the overall compression law produced see Fig 4a Good tracking high tolerance of gain errors in the audio channel avoidance of steady state overloading of highly equalized channels and negligible formation of modula tion products at high amplitudes are features of the overall transfer characteristic not of the method of forming it However the weaknesses of a direct ap proach to such a transfer characteristic would appear in AN AUDIO NOISE REDUCTION SYSTEM the usual forms overshoots high noise level high non linear distortion difficult reproducibility and poor stability NOISE REDUCTION SYSTEM BAND SPLITTING The advantages of the differential method of deriving the compression law are dependent upon the existence of a large ratio between the maximum amplitude of the signal in the main path and the maximum amplitude of the differential component it follows that the com pression threshold must be set at a low value Unfortu nately a low compression threshold is detrimental to good noise reduction properties With moderate and high level signals the noise reduction action technically disappears so that if only one full frequency compres sion band were used an unacceptably high degree of program modulated noise would be evident This dii ficulty has been overcome in the present system by splitting the differential component int
19. 10554 10561 10604 10609 10648 10720 10891 10798 52064 52073 52098 52099 54033 41018 42007 42003 54033 63276 63282 70107 45014 i RZ 7 7 LABEL FAN DUCT Go y FAN amp WIRING HARNESS ASSY A POLY TOP M5x5 4 PL ES COVER Cat No 386 PCB ASSY CAT No 373 PRIOR TO S N 650 XFMR amp WIRING 2 HARNESS ASSY SCREW PAN HD SOCKET M3x6 3 PL o 3 PL INSERTED ON SOLDER SIDE OF PCB PAN HD POZI _ Zei M3x8 2 PL a WASHER CRINKLE M3 2 PL G EXTRUSION LABEL S N p XLR HOUSING 8 PL NYLOC NUT M3 SCREW PAN HD POZI M4x12 A1A3763 REV 4 SCREW EUST e SCREW Ge E FLAT HD POZI Figure 7 1 Model 363 assembly drawing M3x8 2 PL M3x8 3 PL CAT 300 MODULE 2 PL SECTION 8 APPENDICES Part 1 Measurement of Noise and NR Effect Part 2 Dolby SR What It Is and What It Does Part 3 The Spectral Recording Process Part 4 An Audio Noise Reduction System Part 2 is a reprint of a semi technical description of Dolby SR which forms a good introduction to the concepts before reading the following two papers These next two papers in parts 3 and 4 were written and presented by Ray M Dolby the first on Dolby SR was given at the 81st Convention of the Audio Engineering Society Los Angeles November 1986 The second on Dolby A type noise reduction was given at the 32nd Convention in Los Angeles in April 1967 Both papers are reprinted fro
20. 20 dB or more to restore the deflection of the meter the meter amplifier is probably overloading and a quiet audio low or bandpass filter should be connected ahead of the meter In many cases a passive single pole low pass at 20 kHz will be adequate Assuming that noise sources ahead of the SR encoder and after the SR decoder are negligible Dolby SR will give an audible and a weighted measured noise reduction in the range 20 to 24 dB If less noise reduction is observed it is likely that the measurement is at fault as described above B Measurement of Noise Reduction Effect In many installations the extended dynamic range offered by Dolby SR means that noise sources which can normally be neglected such as input and output amplifiers may become significant Any measurement of the effectiveness of the noise reduction action must also measure the contribution of noise from these sources Simply replaying a biased no signal tape and turning the noise reduction system on and off will not give a figure for the practical efficiency of the system it will only measure the playback effect Unfortunately this playback only technique does not always give results which rolate well to the effect produced in the full record replay process The Dolby system is a complementary system recordings to be played with noise reduction switched on are not made with noise reduction switched off Therefore it is not sufficient to use a piece of blank tape for the
21. 30 V supply is generated by a simple voltage doubler circuit fed with the square wave of the Dolby tone generator IC1202c D1204 etc this voltage is used to hold the input FETs QF101 102 QF301 302 etc in the non conducting state A 2 phase ac fan is fitted at the rear of the unit for cooling and is supplied from the secondary winding of the power line transformer 6 9 Signal Processing Modules Refer to the block diagram Figure 6 4 and detailed circuit diagram Figure 6 5 Three types of modules may be used in the Model 363 Cat No 450 provides A type noise reduction Cat No 350 gives SR processing and Cat No 300 is a switchable module providing both SR and A type Modules may be interchanged with no adjustments changes in the logic are automatically made with the circuits acting on identification signals provided by the modules 6 9 1 Normal operation Figure 6 4 is a block diagram of the Cat No 300 and serves as the basis for describing all the modules Note however that the Cat No 450 containing only A type noise reduction circuits Figure 6 6 is considerably simpler and in fact differs significantly in its block diagram Nevertheless the description given below gives a good understanding of how it functions Audio signals selected by circuits outside the module pass through a bandpass filter to remove out of band components and via IC702 to the heart of the module At this point the level for Dolby level is exactl
22. Acoustics 1954 McGraw Hill New York Dolby R M A 20 dB Audio Noise Reduction System for Consumer Applica tions 1983 J Audio Eng Soc 31 3 pp 96 113 Dolby R M The Spectral Recording Process Presented at the 81st Convention of the Audio Engineering Society November 1986 A E S Preprint No 2413 Fielder L D Assessing the Low Level Sound Quality of Digital Audio Systems Presented at the 79th Convention of the Audio Engineering Society November 1985 Furrer W and Lauber A Raum und Bau Akustik Larm abwehr 1972 Birkh user Verlag Basel Gulick W L Hearing Physiology and Psychophysics 1971 Oxford University Press London Jakubowski H Program Level Metering in Digital Audio Technology 1985 J Audio Eng Soc 33 11 pp 905 910 Warren R M Auditory Perception A New Synthesis 1982 Pergamon Press New York Zwicker E Psychoakustik 1982 Springer Verlag Berlin DO Dolby Dolby Laboratories Inc Wootton Bassett Wiltshire SN4 8Q Telephone 0793 842100 Facsimile 0793 842101 Telex 44849 100 Potrero Avenue San Francisco CA 94103 4813 Telephone 415 558 0200 Facsimile 415 863 1373 Telex 34409 Dolby the double D symbol and Dolby SR are trademarks of Dolby Laboratories Licensing Corporation ODolby Laboratories 1987 S87 6824 7497 PAPERS The Spectral Recording Process RAY DOLBY Dolby Laboratories Inc San Francisco and London
23. Dolby SR has been carried out with close attention to the properties of human hearing especially the need to prevent any audible artifacts of signal processing At the lowest signal levels or in the absence of a signal Dolby SR applies a fixed gain frequency characteristic that reduces noise and other low level disturbances by as much as 25 dB Only when the level of part of the signal spectrum increases significantly does the circuit adaptively change its own spectral characteristics When this happens Dolby SR changes gain only at frequencies where change is needed and only by the amount required Adherence to this principle the principle of least treatment is critical to maintaining the extreme purity of sound audible in Dolby SR recordings Listening is the most demanding test Laboratory measurements using test signals show that Dolby SR recordings contain very little noise distortion and other impurities However meters cannot tell us how good Dolby SR tapes sound because equipment does not respond to a recorded signal in as complicated a way as the ear and brain The most important and reliable test of any signal processor is a careful comparison of line in and line out signals while a live recording is made in a quiet studio We urge engineers producers and recording artists to carry out such tests with Dolby SR and to compare their own Dolby SR recordings to those based on any other technology Why Dolby SR is needed All
24. Engineering Society J Audio Eng Soc Abstracts vol 18 p 704 1970 Dec 4 R Berkovitz and K J Gundry Dolby B Type Noise Reduction System Audio pp 15 16 33 36 1973 Sept Oct DOLBY 118 PAPERS THE AUTHOR Ray Dolby was born in Portland Oregon in 1933 and received a B S degree in electrical engineering from Stanford University in 1957 From 1949 52 he worked on various audio and instrumentation projects at Ampex Corporation and from 1952 57 he was mainly responsible for the development of the electronic aspects of the Ampex video tape recording system After he was awarded a Marshall Scholarship followed by a National Science Foundation graduate fellowship he left Ampex in 1957 for further study at Cambridge University in England where he received a Ph D degree in physics in 1961 and was elected a Fellow of Pembroke College Honorary Fellow 1983 During his last year at Cambridge he was also a consultant to the United Kingdom Atomic Energy Authority In 1963 he took up a two year appointment as a United Nations adviser in India and returned to England in 1965 to establish Dolby Laboratories in London Since 1976 he has lived in San Francisco where his company has established further offices laboratories and manufacturing facilities Dr Dolby holds a number of patents and has written papers on video tape recording long wavelength X ray analysis and noise reduction He 1s a fellow
25. Figure 2 8 Remote function wiring STEP 4 Other remote functions can be obtained by wiring a standard 15 pin male D connector to J10 function remote as shown Figure 2 8 Remote processing selection remote indication of signal processing selected depends on which processing module s are installed and remote mimic of the state of the Auto Compare function are provided for each channel independently Remote Set up control is also provided and as with the local control on the Model 363 is common to both channels It is also possible to link the processing selection and Set up functions on several Model 363 units using the J10 connector For common control of processing selection link all pins 7 and 14 together and separately link all pins 8 and 15 together The remote switch is wired as above to any pins 7 and 8 For remote operation the local switch must be in the off mode If the local switch is moved to either A type of SR it will take precedence over the remote switch and the unit and both local and remote indicators will switch to the mode selected by the local switch For common control of the Set up function link all pins 2 together and connect a switch between pin 9 on any unit and these linked pins In this case both the remote switch and any local switch will switch all the linked units into the Set up mode since tne switches are locking switches it is important to release the correct switch to cancel Set up 2 7
26. R509 C501 eem 1C501A RS16 AREN i mA 5 C503 C504 gen R507 105018 1040 B d 100N C502 1 LLI sl si T 5 g o0 1x R501 RS02 100N 100N 39 1 1 R515 C509 an 1M0 68P 2 10KO 1Z T R504 R505 GER SR 10K0 QF503 100K 240K 1NO MH990P TCS02A 1X dl 175 Y R518 DOLBY TONE CS IC5028 R517 10K0 J175 34 OUTPUT 2 ie QF505 QFS04 10E 5519 Ka 10K0 J175 in polo 460 27P R521 R608 10K0 1x 1C602A R520 R601 2n Ste 1C803A l ENCODE ZDE CODE 5123 C601A LH324 R612 4011 10KD J RSE OUTPUT 1 47K 3 4 3 02 n 3 RE13 1 3 10E 3 t Ch 2 100K TCSO4A 2 R609 2 2 100N SETUP NORM jo Eng gi 08038 A ay CS11 47K 7 L 100K 27P R524 6 R610 1C602C A R603 ana ease 7 SAM 1C602E SE a enne g T I oko 1 METER OUTPUT AK h0 5 4 D601 vc e 1C603C 1C602D 98611 La 9 10 2604 2 Do 1C603D 1C602F IN OUT 78 3 12 Im 14 15 47K 13 Do LOGIC REF p605 A TYPE 32 IDENTIFICATION 56K R607 18K Figure 6 6 Cat No 450 main path circuit i SE DS602 SET UP YELLOW NOTES UNLESS OTHERWISE SPECIFIED 1 RESISTOR VALUES ARE IN OHMS 54 2 CAPACITOR VALUES ARE IN FARADS 9 DIODES ARE 1N4148 4 BIPOLAR TRANSISTORS ARE DOLBY STANDARDS DEVICES PNP BC416 2SA970 OR SIMILAR NPN BC414 2SC2240 OR SIMILAR 5 DUAL OP AMPS ARE TLO72 PIN8 9V PIN4 9V 6 IC601 IS LM324 PIN 4 9V PIN 11 GND 7 10602 IS CD4049 PIN 1 9V PIN 8 GND 8 IC603 IS CD4011 PIN 7 GND PIN 14
27. Record playback changeover of each channel can be controlled individually from the front panel a tape recorder or a remote control allowing a single Model 363 to serve for stereo recording applications Front panel toggle switches select Dolby SR Dolby A type or no processing A setup button and four element LED calibration displays allow quick alignment using internally generated Dolby tone for A type or Dolby noise for SR As with all other implementations of Dolby SR the Model 363 features Auto Compare for verifying the performance of the audio system The Model 363 is also ideal for applica tions requiring dedicated encode or decode operation A pair of 363 units can be used for simultaneous record and playback on two channel tape recorders or for transmission DO Dolby SR A systems where the encoder and decoder are physically separated In single channel record ing applications a single unit can be used for simultaneous record and playback The Model 363 incorporates electroni cally balanced transformerless input and out put circuits Independent level adjustments for the record and playback signal paths allow accurate matching of existing line levels A Check Tape switch allows monitoring of either the line in signal or the encoded tape while recording Bypass buttons make it possi ble to remove either or both channels from the entire audio system for studio alignment The Model 363 is normally supplied with two Ca
28. SR A SWITCHING DE MODULE OF SR A TYPE 2 MODULE AUDIO AND OUTPUTS ARE 388mV OR BOTH SYSTEMS test SIGNAL INPUTS AND CONTROLS TO ELECTRONIC SWITCHES NOTE AUDIO CHANNEL 1 SHOWN IN RECORD MODE AND SR PROCESSING SET UP BYPASS SET UP BYPASS Figure 6 1 Model 363 block diagram LOGIC CIRCUITS L NORM NORM l A1B3799 REV I S1100 S902 S901 PROCESS CHECK TAPE NORM S903 904 CHECK TAPE REC PLAY LP BYPASS POWER TO LED INDICATORS NORM METERS ETC J9 3 REMOTE REC PLAY J9 4 S1102 1101 22V F0R RELAYS 101101 VOLTAGE SELECTOR FUSE CONNECTOR TRANSFORMER REGULATOR 15V RECTIFIERS ov 101102 j REGULATOR A SE GULATO 15V J10 7 REMOTE PROCESSING ON OFF fi R212 R206 20K0 0 1 2OKO 0 1 gt Ke REC 1C1014 R202 R205 gt RELAY F202 10K0 1C202A fe RO 4 a SECBYP R227 y175 R204 R207 5S6 KO 0 1 10 tp 22P 10K0 20K0 0 1Z o SCH czo R201 w i our R209 CJE RECORDER JON v oL 000 14 ONDE Ree Q Je 1K74 1 q CH ISK C20
29. System Interna tional Broadcast Engineer No 10 510 1965 10 L H Bedford Improving the Dynamic Range of Tape Recording Wireless World 66 104 1960 11 H Fletcher The Stereophonic Sound Film Sys tem J Soc Motion Picture Engrs 37 331 1941 12 J T Mullin Advanced Tape Mastering System Electronic Features IEEE Trans on Audio AU 13 No 2 31 1965 13 W A Munson The Growth of Auditory Sensa tion J Acoust Soc Am 19 584 1947 14 E Luescher and J Zwislocki Adaptation of the Ear to Sound Stimuli J Acoust Soc Am 21 135 1949 15 R L Wegel and C E Lane The Auditory Mask ing of One Pure Tone by Another and Its Probable Relation to the Dynamics of the Inner Ear Phys Rev 23 266 1924 16 J P Egan and H W Hake On the Masking Pattern of a Simple Auditory Stimulus J Acoust Soc Am 22 622 1950 17 C E Bos and E de Boer Masking and Discrimi nation J Acoust Soc Am 39 708 1966 18 I M Young and C H Wenner Masking of White Noise by Pure Tone Frequency Modulated Tone and Narrow Band Noise J Acoust Soc Am 41 700 1967 THE AUTHOR Ray M Dolby was born in Portland Oregon in 1933 and received his B S degree in electrical engi neering from Stanford University in 1957 From 1949 52 he worked at Ampex Corporation on various audio and instrumentation projects and from 1952 57 was mainly responsibl
30. an exploded view of the unit This allows access to much of the circuitry with the modules in place and to all with them removed The module edge connectors are also accessible and are labeled and temporary connections can be made at the connector to simulate the module connections to assist in tracing signals If this analysis reveals the faulty components then the complete interface assembly can also be removed to allow access to both sides of the board Alternatively since the board may be safely operated out of the case removal may help in fault finding 7 5 7 3 Disassembly numbers as refer to Fig 7 1 Disconnect power from the unit and remove cover 2 Unplug the three pin connector P13 linking the power transformer to the printed circuit board Turn the unit over and undo the two screws 16 holding the heat sink to the chassis Do not undo the larger single screw holding the transformer Undo the three screws 18 holding the front panel extrusion to the main chassis Turn the unit right side up and remove the two counterbored screws 7 holding the extreme left and right hand top side of the front panel extrusion to the chassis sides Remove the front panel extrusion 3 from the chassis Unlock each XLR body from its shell Three different types of locking system are employed in different styles of connectors two made by Switchcraft and one by Neutrik Study the diagrams in Figure 7 2 to identify which
31. as from a subthreshold signal situation the overshoot suppression threshold is set at its lowest point The overshoot sup pression effects are then phased in gradually as a func tion of increasing impulse level Under relatively steady state but nonetheless changing signal conditions the overshoot suppression effects are gradually phased out as a function of increasing signal levels this action further ensuring low overall mod ulation distortion from the system The phasing out effect is achieved by increasing the overshoot sup pression thresholds The thresholds are controlled by signals that are the same as or derived from the mod ulation control signals used to control the steady state characteristics whereby a tracking action between the transient and steady state behavior is obtained This arrangement results in both well controlled overshoots and low modulation distortion Both primary and secondary overshoot suppression circuits are employed the latter acting as fall back or long stop suppressors In the high frequency circuits the secondary overshoot suppressors improve the per formance just inside the stop band that is in the 400 800 Hz region In the low frequency circuits these additional suppressors improve the performance under extreme complex signal conditions such as high level low and mid frequency transient signals in combination with high level high frequency signals In the low frequency circuits a further ove
32. dia gram shows only the basic parameter determining ele ments the practical circuits of course contain other details such as buffering amplification and attenuation The high level mid level and low level stages have the same basic block diagrams and schematics The main distinctions are that the ac and de circuit gains are increased for the mid and low level stages Referring to the block diagram each stage comprises a fixed band section on the bottom and a sliding band PAPERS section on the top each with its own control circuits The fixed and sliding band circuits are fed in parallel and the output signal is taken from the sliding band circuit The sliding band variable filter is referenced to the output of the fixed band that is the fixed band output is fed directly to the bottom end of the sliding band variable resistance RV This connection results in the action substitution operation discussed previously At any given frequency the overall output will be the larger of or some combination of the fixed and sliding band contributions If there is a signal situation in which the fixed band output is negligible then the sliding band predominates Conversely if there is little or no sliding band contribution the output from the fixed band will still feed through to the output through RN In this way the action of one circuit augments that of the other and as the occasion requires may be sub stituted for that o
33. goals of the new system were set high The new technology called spectral recording SR should provide master recordings of the very highest quality especially with regard to audible signal purity Yet the system should be practical and economical for routine applications being suitable for easy and trou blefree use in a wide variety of professional recording and transmission environments Certain new techniques to be described provide the required signal quality and Presented at the 81st Convention of the Audio Engineering Society Los Angeles 1986 November 12 16 J Audio Eng Soc Vol 35 No 3 1987 March practicality but result in circuit complexity Reliance has been placed on improved circuit implementation and manufacturing techniques to overcome the problems of complexity and to ensure economical production of the new system 1 BRIEF OUTLINE The goal of the spectral recording process is to modify the various components of the incoming signal in such a way as to load an imperfect recording or transmission medium in the most rational way Generally high level signal components at both ends of the spectrum are attenuated whereby a better match with the overload characteristic of the medium is provided At the same time low level components of the signal are amplified substantially ina highly frequency selective way These effects are reversed during reproduction restoring the original signal The result is a s
34. in the meter amplifier automatically compensate for this low level and raise the signal to the same level as Dolby tone In both record and playback modes the signal from the From recorder input socket is selected by switches in the frame and enters the module at pin 26 When A type processing is selected this input signal is sent to the monitor output via the normal position of IC803B and to the Meter output via the normal position of IC802B The return signal from the tape normally Dolby tone in this case can therefore be heard and displayed on the calibration display When SR is selected the input signal is normally Dolby noise It is therefore fed to the meter output via a bandpass filter and amplifier selected by IC802B allowing an accurate and steady display Output 1 receives the Dolby noise and reference pink noise alternately with four seconds of each selected by IC803B This Auto Compare check is an extremely critical check for small errors in gain or frequency response of the overall record playback process Identification of the two signals is simple as the pink noise is uninterrupted while the tape Dolby noise has the characteristic nicks occurring every 2 seconds In addition there are indicators on the front of the module to help identification these indicators can be duplicated at a remote location through a D connector on the rear of the Model 363 The sync output from the module pin 16 is not used in the Model 363
35. in those channels can be linked with one processing in out switen I 3 18V tape 2 2 gy V 3 10 Mi Remote I Auto Compare indicator i S 680 red 4 reference l 8 15 2 gt 15V Remote Set up set up loperates an both HE channels pin 2 of severa units I may be linked 1 5 15V A42C3808 6 REV C BEFORE S N 650 Figure 2 9 Remote function wiring prior to unit number 650 Note that the operation is somewhat different from that described previously especially in that relationship between local and remote controls In addition connections for the remote Auto Compare indicator are different It is possible to link this older processing on off control on several Model 363 units using the J10 connector For common control of processing in out link all pins 7 and 14 together and connect a switch between pin 8 or 15 on any unit to these linked pins Note the difference from later units with the master remote switch in the on position the local switch on each channel can still be used to turn off that particular channel With the master switch in the off position the local control is disabled and processing remains switched off If the wiring for an older unit is connected to a later unit or vice versa no damage will be done but control will not be effective The local and remote mode and nr type indicator of any unit will always indicate the actual operating mode o
36. increases in level the band slides upward As with the fixed band circuit the unnecessary sliding results in a loss of noise reduction action and the modulation of signals at higher frequencies when the sliding band varies under the control of the 100 Hz signal Fig 7 b shows the operation of the same circuit with modulation control Minimal sliding occurs when the 100 Hz signal is varied over the same range of levels as in Fig 7 a Thus the sliding band compressor is also made essentially immune to strong signals outside its pass band The effect of modulation control is further illustrated by Fig 7 c and d except that the frequency of the dominant signal is changed to 800 Hz a frequency within the pass band of the circuit Ideally sliding is 10dB No signal 5 o 20 200 2000 20000 FREQUENCY Hz a Frequency response curves with 100 Hz signal at the levels indicated no modulation control 10 dB No signal B pum 9 20 290 2000 20000 FREQUENCY Hz c Frequency response curves with 800 Hz signal at the levels indicated no modulation control SPECTRAL RECORDING PROCESS required to go only so far as not to boost the 800 Hz signal above the 0 dB reference level Thus in the Fig 7 c response without modulation control the sliding produced by the 800 Hz signal at levels above 10 dB is excessive Fig 7 d shows the response of the circuit with modulati
37. independent of frequency at least in the pass band A sliding band circuit has a fast recovery time for nondominant signals at the pass band end of the spectrum and a slow recovery time for nondominant signals at the stop band end of the spectrum The choice of integrator recovery times is therefore a matter of compromise between this recovery time situation and the amount of steady state and modulation distortion obtained The compromise is made much easier by the use of the action substitution technique In particular the fixed band provides a definite and rapid recovery time for the overall system so that the sliding band can employ longer time constants than would otherwise be desirable This results both in low modulation dis tortion and a fast recovery time Thus the action substitution technique provides the advantages of fixed and sliding band circuits while avoiding their disadvantages In other words there is a significantly improved adherence to the ideal of least 10 dB FREQUENCY OF f DOMINANT SIGNAL a Fixed band compressor characteristic 10 dB FREQUENCY OF f DOMINANT SIGNAL b Sliding band compressor characteristic 10 dB FREQUENCY OF f DOMINANT SIGNAL c Action substitution compressor characteristic Fig 4 Types of compressor characteristics J Audio Eng Soc Vol 35 No 3 1987 March SPECTRAL RECORDING PROCESS signal treatment In the level region somewhat above the circuit
38. inst lld p korrekt n tsp nning O o Konrollera att s kringarna r av r tt typ och f r r tt str mstyrka s som anvisningarna p enheten f reskriver o Enheten m ste vara jordad genom anslutning till ett korrekt kopplat och jordat el uttag o El sladden som medf ljer denna enhet m ste kopplas enligt foljande Fas Brun Neutral Bl Jord Gr n Gul BELANGRIJK VEILIGHEIDS VOORSCHRIFT Deze unit voldoet aan de EN60065 veiligheids standaards Dit apparaat mag niet worden blootgesteld aan vocht Vanwege het risico dat er druppels in het apparaat vallen dient u er geen vloeistoffen in bekers op te plaatsen Voor een veilig gebruik en om het gevaar van electrische schokken en het risico van brand te vermijden dienen de volgende regels in acht te worden genomen o Controleer of de spanningscaroussel op het juiste Voltage staat o Gebruik alleen zekeringen van de aangegeven typen en waarden ND o Aansluiting van de unit alleen aan een geaarde wandcontactdoos o De netkabel die met de unit wordt geleverd moet als volgt worden aangesloten Fase Bruin Nul Blauw Aarde Groen Geel NOISE REDUCTION UNIT DOLBY 563 DOLBY 363 SR A NOISE REDUCTION UNIT he Dolby Model 363 provides two chan nels of noise reduction switchable between Dolby SR Spectral Recording and Dolby A type The economical 1 U high unit is intended for all kinds of audio facilities music recording video postproduction broadcast multimedia and film
39. level noise reduction system has also been developed recently by Mullin 12 Presented April 25 1967 at the 32nd Convention of the Audio Engineering Society Los Angeles REQUIREMENTS Referring to Fig 1b it is possible to draw up a set of requirements which any noise reduction system must meet if it is to be used without reservation in high quality recording or transmission channels The system will necessarily be of the complementary type Recording out or Processor Transmission Noise a Non complementary system b Complementary system Control 1 Record Recording Playback or Processor Processor Transmission Noise Fig 1 Noise reduction systems illustrating two basic types with reference to effects on signal The control path shown in b is optional Overall Signal Quality Requirements 1 The output signal should not be perceptibly differ ent from the input in frequency response transient re sponse and dynamics stereo signals thus should be perceptibly free of image wandering or shifting Reprinted from Journal of the Audio Engineering Society October 1967 RAY M DOLBY 2 The system should not introduce perceptible non linear distortion of transient or steady state signals at any level or at any frequency or combination of frequencies the overload point should be substantially above the nor mal peak signal level 3 The system s
40. of the differential method In the system under discussion using four bands with compression thresholds of 40 dB below peak operating level the frequency divisions are Band 1 80 Hz low pass Band 2 80 Hz to 3 kHz band pass Band 3 3 kHz high pass Band 4 9 kHz high pass Bands 1 3 and 4 are conventional 12 dB per octave filters while Band 2 has a frequency response which is complementary to that of Bands 1 and 3 The outputs of all the bands are combined with the main signal in such a wav as to pro duce a low level output from the recording sending proc essor which is uniformly 10 dB higher than the input signal up to about 5 kHz above which the increase in level rises smoothly to 15 dB at 15 KHz The figure of 10 dB represents a compromise between a number of design factors not the least of which is the desirability of minimizing sensitivity to gain errors RAY M DOLBY in the audio channel At the high end of the spectrum an extra 5 dB of noise reduction is obtained without appreciable disadvantage in this regard Bands 3 and 4 contribute approximately equally in this region so that with normally encountered sounds the output of Band 3 is usually compressed substantially before the threshold is reached in Band 4 The maximum compression ratio is thereby reduced and the possibility of occurrence of program modulated frequency response under gain error conditions is decreased Because of the use of four bands with conseq
41. of the same type and rating can be placed in the upper position Close the compartment door making sure it clicks firmly into place SPARE FUSE UPPER OCTAGONAL PEG ROUND PEG OLTAGE SELECTOR WHEEL FUSE A2F4037 Figure 2 2 2 3 2 2 Connections STEP 1 Connect the Model 363 to mains power STEP 2 Connect audio cables to the inputs and outputs For optimum EMC performance wire the shields of the audio cables to the outer shells of the XLR connectors not to pin 1 which should if possible be left open If other items of equipment connecting to the Model 363 have unbalanced inputs or outputs use twin screened cable and wire the connectors at the Model 363 end exactly as if the circuits were balanced perform the unbalancing at the other end Ch 1 J To console J3 From rec J2 To rec J1 From console J8 To console J7 From rec J6 To rec J5 From console Figure 2 3 NOTE 1 Current IEC wiring convention calls for XLR pin 2 to be high hot and pin 3 low cold In a balanced system the distinction is arbitrary provided there are no phase inversions through the unit the Model 363 maintains phase As the Model 363 is fully floating it is unimportant which of the two signal pins is grounded so long as the same convention is used on all inputs and outputs Nevertheless in the interests of maintaining international standardization we suggest the IEC recommendation is followed NOTE 2 It is n
42. polarities as shown in Figure 2 5 channel 2 pin numbers shown in parentheses Many tape recorders have suitable switching voltages available note the switching input on the Model 363 uses an opto isolator to provide a floating input The input current is roughly 5 mA essentially independent of voltage Switching voltages higher than 25 V can be accommodated by adding an appropriate series external resistor J9 Tape rec play Recorder remote Model 363 pe rec 4 to 25V local i y opto isolator switch z 3 0 AN o T L Ji 1K 1 ie gt Play Nm 4 I A rec play rec play ov 4 8 4 U I l A2C3810 Figure 2 5 Rernote control of mode voltage operated Jg rec play remote Hodel 363 I zie T 2 B 1K i E opto isolator local Recorder 3 M 5 e i Ed gt l I E l amp ae PEE play rec play 2 4 t8 rec Ll i l l l 5 9 A2C3810 Figure 2 6 Remote control of mode using isolated contact Alternatively some recorders provide an isolated make contact which can be used by wiring the connector to pick up the supply voltage in the Model 363 Figure 2 6 Wired in either manner the unit is switched to Record using either the local rec play switch or by the remote connection provided
43. record mode and is muted in the play mode This prevents a positive feedback loop being formed in the recorder under certain combination of Model 363 and tape recorder monitor switch settings In the play mode another path is made available if the Model 363 is in the Set up mode and the signal processing has been turned off With these conditions met there is a calibrated path from the From console input to the To recorder output which allows a simple calibration procedure using console tones to be used see Sections 3 and 4 The switching for the To console output is inside the processing module and will be described later If the unit is being used with a three head tape recorder it is possible to monitor the reproduced signal in either the record or play modes The check tape switches operate the play bypass relays K401 K801 connecting the To console outputs directly to the From recorder inputs Note that if SR or A type processing is being used the signal heard will be an encoded signal and so cannot be used for critical comparison with the input signal but only as a simple check of the recording process 6 5 Metering Each module provides a metering signal of 388 mV 6 dBu when the internal signal is at Dolby calibration level This signal is applied to a comparator circuit which at Dolby level gives equal illumination of the two green LEDs As the signal moves away from this level one of these LEDs fades down while the oth
44. recordings and communication systems have definite dynamic range limits However a simple measurement of maximum level and noise level does not reliably indicate how recordings made with such a system will sound Such a test says nothing about noise that appears only in the presence of a signal or about system behavior when the signal is at the over load level Analog tape for example saturates gently digital recordings on the other hand clip fully if maximum level is reached even for less than a millisecond Because Dolby SR increases recording headroom considerably there is less risk of under or over recording The engineer s working space is increased and there is greater freedom for creative effort Dolby SR provides effective protection during original record ing during mixdown when equalization or specialized signal processing requires the lowest possible noise level and when multi generation copies are needed The analog tape recording window 20 O dB 20 40 80 100 20 200 2k 20k The limits of unassisted analog recording using a standard professional recorder and tape at 15 ips are sketched in Figure 1 The limit at high signal levels actually a gradual overload is at the top of the clear area The noise level is the bottom of the clear area Both the overload level and the noise vary with frequency The central open part of the sketch can be thought of as a window through which the signal must f
45. special Neutrik tool can be used Turn the screw anti clockwise to unlock the shells 8c If the connector is in the style of Figure 7 1 c simply undo all of the top mounted captive countersunk screws a few turns to allow the connector bodies to slide out of the housing The motherboard 5 can then be withdrawn through the front of the tray Once out it can be reconnected to the transformer CAUTION Do not leave the mother board assembly complete with processing modules powered outside the chassis for extended periods of time The heatsink will become very hot and the regulators may shut down preventing further troubleshooting The transformer assembly can be removed by removing the green yellow ground wire from the transformer to the ground lug undoing the bottom fixing screw then sliding the power input module upwards out of the chassis CARE Live terminals are present on the rear of the power input module when the rubber boot is removed Re assemble unit by reversing the above steps 7 5 7 6 Parts List Most parts are available locally however the following lists Dolby part numbers for items which are either special or may be difficult to obtain locally They can be ordered from local agents or Dolby Laboratories offices in London or San Francisco Component capacitors 2 5 Circuit Reference only channel 1 listed Doiby Part Number 330p C1215 25003 680p C1214 25008 1n0 C203 403 901 1114 25010
46. tape noise source 8 4 The tape to be played back with noise reduction switched on must be recorded with noise reduction switched on Similarly the tape to be played back with noise reduction switched off must be recorded with noise reduction switched off Only by doing this can the effect of extraneous noise sources on the total noise be properly assessed The correct test procedure is as follows a Feed a signal at normal reference level from the console record a test length play back the tape and confirm the output level is also at reference Remove the source signal if the test oscillator is merely turned down ensure there is really no signal at its output a small leakage of tone would invalidate the noise reduction measurement Record a length of tape with the noise reduction system off and at a convenient timing point turn the system on Rewind the tape and play back the section recorded without noise reduction with the noise reduction switched off and measure the output noise At the point at which the system was switched on in the recording switch on the system and measure the reduced output noise The noise reduction effect is the difference between the measurements in d and e above The exact figure will depend on the weighting method used but should be over 10 dB for Dolby A type noise reduction and over 20 dB for Dolby SR DO Dolby SR Dolby spectral recording What It Is and What It Do
47. the fixed and sliding bands in any particular stage all of the stages operate independently Depending on the levels and spectral conditions in each stage fixed band operation is used whenever it provides best per formance sliding band operation is substituted when ever it has an advantage The substitution is effective on a continuous and frequency by frequency basis Even though the frequency division of the stages 1s nominally 800 Hz the use of what are effectively single pole band defining filters results in a significant overlap region between the high and low frequency stages The high frequency stages extend their effects down to about 200 Hz the low frequency stages extending their effects up to about 3 kHz This overlap together with the use of action substitution contributes to the achievement of a very good spectral tracking effect Output IER Compressor 1 Fig 3 Another action substitution compressor configuration J Audio Eng Soc Vol 35 No 3 1987 March PAPERS under all frequency and level conditions The practical significance is that an excellent noise reduction effect is obtained in the presence of signals and that the system has a remarkable tolerance to gain and frequency response errors in the signal channel A further aspect of action substitution relates to the transient recovery characteristics of the system A fixed band compressor circuit has a recovery time that is essentially
48. the level returning from the tape recorder and if the Model 363 is switched to record Dolby noise at low level will be the output to the recorder The gain of the calibration display is increased so that Dolby noise at the correct level will give an Dolby level indication This mode is also used to align the Model 363 to Dolby noise played back from a previously recorded tape The Model 363 will start an auto compare sequence when Dolby noise is detected at the From recorder input Where both tracks in a two channel situation are being replayed first start the playback of the Dolby noise and then press the Set up button this ensures that the Auto Compare switching in the two channels is synchronized However we recommend listening to one track at a time 4 4 Set up To help alignment of the Model 363 a Set up button is provided Pressing the Set up button will set the unit to the correct alignment mode which depends on the setting of the SR off A switch Initially the Model 363 must be aligned to the normal signal levels present in the studio To help this alignment a special mode is selected if the Set up button is pressed when noise reduction is in the off position 4 4 1 When the SR off A switch is in the off position the signal paths in Set up are To Recorder Calibration display To Console Play out The calibration display may be switched between the record and play signal paths using the front
49. the local switch is left in the play position By connecting pin 1 to pin 5 or 9 the local control is inhibited and only the remote switching system will effect the change Figure 2 7 At the end of this section starting on page 2 8 there are specific diagrams for several of the commoner two track recorders J9 rec play remote Model 363 I grounding i s this line di sables I local switch Tape I Recorder 5 ORS 4 to 25V l l local I i switch a opto isolator disabled 8 03 NL 1K I h l Ki b rec play rec play ov 4 8 M 4 I J I l I A2C3810 Figure 2 7 Disabling local 363 rec play switch Mode 363 J10 Remote Function 5 to i5V AN SR P O off o du 14 o 8 15 A type l l l 4 11 A GW Remota T SR green 5 333 Processing o LED indicators A type red 6 13 NOTE for simultaneous operation of several channels pins 7 and 14 l in those channels can be linked with ore processing SR off A switch x gt 15V tape 2K SEN 34 1181 7 Remove Auto Compare indicator gt dz 680 red refererce I aul P x gt FIN Remote Set up 1 set up l loperates on both cmm channels oin 2 l of several units I may be inked A2C385 7 After S N 650 REV B
50. the other bands and responds with gain changes It can be shown that a change in input signal at any particular frequency that causes x dB gain change in one band causes exactly x dB change in all the other bands Thus constant slope split band compandors also inevitably lead to audible noise modulation The solution is to employ a defined low level input threshold below which the frequency dependent gain or loss of the processor is constant Together with appropriate frequency response adaptation it is then possible for the processor to keep its sub threshold gain or loss except in those areas of the spectrum where high level signals mask the noise If the gain or loss in unmasked regions of the spectrum is constant then there can be no noise modulation Such a response adaptation and low level threshold are features of all the Dolby systems If the level of an input signal at a particular frequency increases abruptly an encoder must reduce its gain in response to that new level This gain reduction occurs over a finite time during which the encoder output level will be excessive this excess level is known as overshoot In general the magnitude of an overshoot corresponds to the degree of gain reduction and its duration to the response attack time Provided the overshoot does not lead to overloading of the recording or transmission system it is harmless However for high level signals overshoot can cause transient distortion and non complementar
51. threshold of the relevant compressor circuit is to tend to create a bal ance or equilibrium between the compressor circuit control signals and the modulation control signals Under these conditions there is a significantly reduced gain reduction or sliding of the variable filters as a 90 Side Chain Signal Total Signal a 7 d A Main Path Signal a In the pass band a low threshold and strong limiting characteristic are required 90 Side Chain Total Signal Signal Main Path Signal b In the stop band phase shift results in the side chain signal having negligible influence on the total signal amplitude therefore a higher threshold and weaker limiting can be used Fig 5 Phasor diagrams dual path compressor 104 PAPERS function of increasing input signal levels Fig 6 illustrates the action of modulation control with a high frequency fixed band compressor circuit The circuit has a low level gain of about 8 dB and an 800 Hz high pass characteristic Fig 6 a shows the response of the circuit in the absence of modulation control Ideally there should be no attenuation in re sponse to a 100 Hz signal because the overall shape of the envelope is such that there is negligible signal boosting at 100 Hz Nevertheless with a conventional compressor circuit as shown here when the 100 Hz signal increases in level there is a reduction of low level signal boosting over the whole frequency band The unneces
52. threshold the signal more closely approaches fully boosted conditions in the encoding mode with a consequently improved noise reduction effect in the decoding mode For signals at higher levels the tech nique of modulation control described below is em ployed 2 3 Modulation Control In the A type B type and C type systems the signal from the side chain is highly limited under high level signal conditions This high degree of limiting begin ning at a low level threshold is responsible for the low distortion low overshoot and low modulation distortion which characterize these systems A closer examination shows that it is unnecessary to utilize such a low threshold and such a strong limiting characteristic under certain signal conditions In par ticular whenever the side chain signal departs from an in phase condition with respect to the main path signal then the threshold can be raised Moreover after an appropriate degree of limiting has taken place at a given frequency in order to create the desired overall compression law then it is unnecessary to continue the limiting when the signal level rises even further Rather the level of the side chain signal can be allowed to rise as a function of a further increase in signal level whereby it stabilizes at some significant fraction of the main path signal level In the fixed band portions of the spectral recording process the above arrangement results in conventional perform
53. type recordings is generated by applying a square wave to the input of a filter IC1204c switches between precisely defined 6 V and 6 V supplies the switching rate is 850 Hz 1C1202a Q1204 modify this frequency to 935 Hz for 30 ms every 750 ms producing the characteristic warble sound frequency rather than amplitude modulation is chosen so that meters do not indicate any level change during the time of the modulation 6 3 Pink noise is obtained by taking the output of a white noise generator IC1205 and passing it through a pinking filter The pink noise also passes to Q1203 which inserts a 20 ms nick every 2 seconds this is Dolby noise 6 8 Power Supplies Incoming AC power passes through a voltage selector switch and to the transformer giving nominal operating voltages of 100 V 120 V 220 V and 240 V Full wave rectification is used the transformer secondary is centre tapped The power supply is conventional separate positive and negative IC regulators are used to provide 15 V and 15 V used for all circuits In addition a simple circuit supplies 22 V for the relays The supply to these relays is arranged to provide delays in both switch on and switch off to prevent large transients in the outputs On switch on the bypass relays remain open in the bypass mode for about one second provided by C902 R914 and C1302 R1314 On switch off the relay is opened when the 15 V supplies fall to about 7 V A
54. unity for instance Somehow to achieve this gain over as broad a range of frequencies as possible in the presence of higher level dominant signals is the task of the system Thus in a superposed action compressor circuit represented in Fig 2 a signal is fed into a first com pressor circuit The output from this circuit represents the completed part of the total potential action The input signal minus the completed part is therefore the uncompleted part This is so derived and fed into the next compressor circuit which has some different characteristic The output of the second circuit is then added to that of the first augmenting the action of the first In an extreme condition in which the output of the first circuit may be negligible at a particular fre J Audio Eng Soc Vol 35 No 3 1987 March SPECTRAL RECORDING PROCESS PAPERS uoneinjyestue Aouanborj o pue 2uonba1J g81q pur 3uimoys pesjoads auenbaxJ AAo pue Kouonboi 48I4 1011409 uonenpour spueq SUIPI s pue Spurq paxy uonninsqns 001138 8u11a8 uote orumu p 2uonbo1 Ao pue quiu 3uanba41 u8I ais nu ss220Jd jo 24nj 2j UEU sous ure 1er ss2204d Surp10221 0112ads weaseip xoo q srg indino IeuBis ied reuBis uie 3 ino ap03u3 JEuBIS us us apo eg Buimays rU EE je1mads 11 uon uon e1njesnuy e2njesiyuy 3H Jappy 3BE1S 15L pueg Buipi s pueg Buipi s pueg Buipiis H pueg paxI4 5
55. which have been investigated the syllabic compressor and expander compandor technique Fig 2 has been the subject of the most development effort Since the noise reduction system to be described in this paper may be roughly classi fied as a compandor it is worth noting some of the limi tations of previous approaches to compression and expansion Well known compandor difficulties which by now are regarded as classical include poor tracking between recording sending and reproducing receiving both stati cally and dynamically high sensitivity to gain errors in recording or transmission inadequate dynamic range high noise level vs high distortion overshooting with tran sient inputs audible modulation product generation un der dynamic conditions distortion of low frequencies by control signal ripple modulation and production of no ticeable signal modulated noise effects Comparison of compandor performance with the pre viously listed requirements for high quality applications shows that the normal compression and expansion ap proach is inadequate Compandors have thus been found to be usable without qualification only in relatively KH Recording Variable So Circuit in Variable Sow Tronsmison Control Circuit Dale A Fig 2 Layout and input output transfer characteristics of a compandor noise reduction system low grade narrow band applications such as te
56. zH 008 90W 1092 95 i zine WNnuixEA T L d ooz T S O Asewuid 4 BN LW CS SO 29 nee S O 41 zH EY S O 1eui11d d N N pueg paxij z ZH 9 L H 008 POW ZH 00 S O iepuo as S O Jl 41 id pe TT BIW LOW S2W jo o Ee POZNA H 009 pueg BuIpIIs ul 113 J Audio Eng Soc Vol 35 No 3 1987 March DOLBY sliding band final integrator circuit In both the fixed and the sliding band circuits the effects of the primary overshoot suppression circuits are maximized for the most significant transient signal situation that is a single impulse or toneburst starting trom a subthreshold signal level A side effect of the use of smoothed MCS and MC7 signals is that the ov ershoot suppression levels for low and medium fre quency transient signals are raised under certain complex signal conditions especially those in which relatively steady state high frequency signals at high levels are also present To compensate for this effect secondary overshoot suppression signals are derived from the fixed band main control circuit rectifier and are diode coupled to the fixed band and sliding band final integrator cir cuits The secondary overshoot suppressors have higher thresholds than the primary suppressors and operate only rarely because of the unusual circumstances for which they are designed The secondary overshoot suppression signals are generated from the frequency weighted point 800 Hz and 6 kH
57. 110 dB We will stay with that setting as we look at various recording system windows Several interesting facts are visible in this figure One is that the noise of the tape will be audible only in a restricted range of middle frequencies that is where the auditory threshold is lowest This is because noise or distortion components at higher and lower frequencies even if only slightly below the threshold are totally inaudible Another observation is that if the audible noise in the mid range frequency band could be reduced by 20 25 dB no noise would be heard at all Disturbances produced by the presence of a signal When no signal is present the only low level defect that can be measured is tape hiss However in the presence of a signal the analog tape recording window closes further as other artifacts are added to the signal layer by layer Figure 9 shows several components of noise and non linearity that can appear in the presence of a signal The signal is shown as the vertical bar at a and is at a level that causes 3 harmonic distortion These harmonics are shown in correct scale at b Modulation noise which appears only when a signal is present is spread over a wide range in the spectrum c The bottom layer of noise tape hiss d is caused by statistical fluctuations in magnetic domain orientation in the tape coating Available dynamic range Another interesting fact is shown in Figure 10 Analog recording m
58. 2 ms smoothed ver sion of MC6 the fixed band steady state modulation control signal the effect is in the direction of improving the steady state and overshoot suppression threshold tracking ona steady state basis However the thresholds must also track on a transient basis This is the function of the high frequency transient modulation control signal MC8 which is a high frequency weighted peak detected signal that opposes the primary overshoot suppression signal in the time interval before MC7 be comes effective The overshoot suppression signal is then diode coupled to the final integrator circuit of the fixed band circuit In the generation of the sliding band primary over shoot suppression signal the output of the variable filter is fed through a 200 Hz single pole filter top of Fig 11 in order to reduce the effect of the circuit at low frequencies as in the fixed band circuit The signal is rectified and then opposed by modulation control signals MC5 and MC7 to provide an adequate degree of tracking between the steady state threshold and the overshoot suppression threshold on a steady state basis As in the fixed band circuit MC8 provides the required degree of tracking on a transient basis The resultant overshoot suppression signal is diode coupled to the J Audio Eng Soc Vol 35 No 3 1987 March 129 SE o nuoo 1U91SUEA pue oje1s peojs ureisvip YOo q 2321s Aouanbay MmoT J BLT S O 31 l T
59. 2n2 C1202 25014 3n3 C1213 25017 6n8 C1203 1204 1205 1212 25021 capacitors 5 22p C104 105 201 202 304 305 401 402 24082 68p C103 303 1206 24123 10n C106 206 207 21071 15n C1211 21073 33n C1210 21077 220n C101 102 301 302 1201 1209 1216 21135 capacitor 10u 50v C1117 1118 22072 card guide 62281 connector edge J11 12 71133 connector XLR female J1 3 5 7 71145 plus 71146 connector XLR male J2 4 6 8 71144 plus 71146 connector fan 3 pin PL14 70121 plus 3 x 75004 connector 9 pin male D J9 71141 connector 9 pin female D cable mounting 70113 plus 9 x 75050 connector 15 pin female D J10 71142 connector power 70106 connector transformer PL13 70111 plus 3 x 75014 cover screws M 5x5 60199 cover unit top up to S N 649 63277 cover unit top after S N 650 63280 diode reference DZ1203 44054 display calibration DS100 32043 duct for fan 63283 extrusion front panel 63274 escutcheon control panel 63275 fan and wiring harness 83099 fuse 250 mA 1 25 56033 fuse 250 mA 20 mm 56035 fuse 500 mA 1 25 56003 fuse 500 mA 20 mm 56036 fuse holder IEC 70108 fuse holder US 70109 insulator for guide 63284 integrated circuit 101205 45014 integrated circuit 4556 10203 403 44099 integrated circuit LM317T 181101 44025 integrated circuit LM337T 101102 44042 integrated circuit TL431 DZ1203 44054 jumper grounding J1101 74087 Component motherboard up to S N 649 motherboard after S N 650 opto isolator 4N28 potentiometer multi t
60. 36 x a PROC R226 Cem R208 ENG mo 220K U REC OUT 10K0 gt Te A n sim c206 CAL 0 1 UTT ar q TON 1C2028 ts INPUT l MO de K201 D201 l bs R229 R230 7 e s Se RDO 6K98 11 mim C209 Ed 54 GND 0UTPUT SC fess 0902 I CAT NO 1 5 6K98 Feios l 300 35 EIE W0 dw 1cz04 41 C204 30v 1 l 0 350 I Y TL072 ps 3 v BL 18 DOLBY TONE output 1 36 T gt t ng l V401 PLAY 25K Y 207 DOLBY NOISE I cuv QUT A 1 l I Kl z 1K74 SA E PN PINK NOISE wn 38 u AAA Rate 301 0202 I ki E E R311 ENC R1024 1C201A 1K74 ae ewcrec URS Lys PLAY 3 a l I a ix RELAY PROC Coen PLAY SET 10 SET UP NORH METER OUTPUT 36 E 1C402A re RECORDER T 102018 R404 3 R407 To s GH E l gt 10K0 20K0 0 14 o CONSOLE SR 78 sR A TYPE l 0 11 lows K EN I I gt 10K0 0 15K MC4DI GI Q INO IN fE in out sro D304 sv I E 4 _ LOGIC REF 104028 0 1 R416 22E1 D401 wns l uw y 4556 2940 0 1 1 15V Jne Jie S SE ARA zg 28 M ninic C406 3420 R417 8 l 1008 10K0 10K0 POCA Ye TOME 0 1 10 17 DOLBY NOISE AGD re IDENT I PINK NOISE E 10204845 C404 A y AS y E eng M AUTO COMPARE DE v mie 1C9U ninc SZA RELAY ROUND 15V 4 inii EXTERNAL TO MODEL e SR GRN 12 REMOTE PROCESSING Do qe SWITCH J10 D303 I al R328 p SR s302 1 I yor 15 D304 I Mec so D om 1C904D E actyee jg R936 A TYPE RED l 6V i I U REF I I J10 1C904A v l MHR 5 R311 CD4049 REMOTE SR LU
61. 987 March DOLBY the 400 Hz high pass filter at the input Thus the overall quiescent subthreshold frequency response of the circuit is that of a single pole 800 Hz high pass network The low frequency stages have a complementary 800 Hz single pole low pass characteristic which overall results in optimal combination of the signals from the high and low frequency stages The fixed band output from RV that is variable gain circuit is fed to two control circuits the main control circuit middle of Fig 9 and the pass band control circuit bottom of Fig 9 In the main control circuit the signal is rectified and opposed by the mod ulation control signal MC3 The resulting de signal is smoothed by an integrator circuit with a 15 ms time constant the overall steady state control signal char acteristic in this and all other stages is average re sponding The control signal is then fed to one input of a maximum selector circuit which passes to its output the larger of two signals applied to the input The fixed band output is also fed to the pass band control circuit bottom of Fig 9 which comprises a 1 6 kHz single pole high pass filter a rectifier and a smoothing circuit 15 ms The pass band control signal is applied to the other input of the maximum selector circuit The output of the maximum Selector circuit is further smoothed by a 160 ms time constant and is used to control the fixed band variable resistance RV
62. As mentioned previously Fig 1 shows the basic J Audio Eng Soc Vol 35 No 3 1987 March PAPERS layout of an SR processor While the whole system comprises an encoder and a complementary decoder the figure as drawn shows a switchable configuration which generally is the most useful one The main signal path transmits high level signal components To this is added in the encoding mode and subtracted in the decoding mode the output of the side chain circuitry called the SR signal point C The stage circuits as well as the spectral skewing networks and antisaturation networks above are driven from point A See 1 for a mathematical explanation of these arrangements A secondary main path which does not include any antisaturation is employed as the basis of the side chain to which the outputs of the high level and mid level stages are added in the first stage and second stage adders respectively The low level stage and modu lation control circuits 1 7 are driven directly from the output of the second stage adder Modulation control circuit 8 is driven from the output of the spectral skewing network as will be discussed The antisaturation effects are created in the dashed block labeled stage signal combiner and antisaturation The arrangement shown provides a high frequency deemphasis effect on the secondary main path signal which includes the output of the high level stages and on the high frequency mid level stage signal
63. DO Dolby Model 363 Two Channel SR A Type Processor Users Manual Users Manual For Model 363 Two Channel SR A Type Processor Serial No 650 Onwards Dolby Laboratories Incorporated U S A 100 Potrero Avenue San Francisco CA 94103 Tel 415 558 0200 Fax 415 863 1373 U K Wootton Bassett Wiltshire SN4 8QJ Tel 01793 842 100 Fax 01793 842 101 WARRANTY INFORMATION USA Warranty on the product covered by this manual is subject to the limitations and disclaimers set forth in the warranty disclaimer originally shipped with the product and also printed on the back of the invoice All requests for repairs or information should include the unit model number and serial number to assure rapid service Dolby and the double D symbol are registered trademarks of Dolby Laboratories Licensing Corporation 1996 Dolby Laboratories Inc ISSUE 8 W96 034 Dolby Part No 91122 TABLE OF CONTENTS SECTION 1 INTRODUCTION AND SPECIFICATIONS 1 1 TA TOFOdUCHON usus eR RED Ee ee E ER Ba a e ee ee 1 1 SECTION 2 INSTALLATION 26564 ccs dux RERE 504 oe A 2 4 2 1 Installation 2 1 22 Connections ui foes an xe Rack dece bates Aiaka Ropa e Vue AA eg d 2 3 2 3 Connection Diagrams for Specific 2 Track Hecorders co c 2 8 SECTION 3 APPLICATIONS 3 nosse dE EC bU bg XR RR Rr ROI Eier dU dad ces Sidi 3 1 3 1 Switched operation ura cei uer mac Soke RXCR
64. DS 7 XC n6 A a eee R ere SS SS SS SS SS SS SS SS SS 4 P CH FET s ARE J175 N CH FET s ARE J210 r SR CIRCUIT m m mm y LLS FB 4 5 DIODES ARE 1N4148 r STAGE SIGNAL COMBINER gl I 15 6 SINGLE OP AMPS TLO7O g anti saturation Ser 3 ANTI SATURATION 8V TO PIN 4 l 13 es c713 7 DUAL OP AMPS TLOG2 TLO72 LF442 NES512 NES632 LMIS8 gt LLS S8E L MLSNLFNFB 8V TO PIN 8 bing 18 9V TO PIN 4 Il Dp 17 8 CMOS SWITCHES ARE CD40538 RCA ONLY STATIC SENSITIVE 4 ceros Cum PECTRAL SKEWING E SR E HLSXLFASBE _ XSV SHITCHES SHOWN WITH LOGIC IN LOW STATE 220N 100N 22K1 39V TU PIN IG ee R717 R719 1C703A STAGES 27 anm s MLS HF FB 9V TO PIN 7 l u UI gt HLS LF sam 1 i GND TO PINS 6 8 I p 5 9 X INDICATES CAT No 300A EDGE CONNECTOR PIN 1 R796 1 dos L INDICATES INTERBOARD CONNECTORS J1 J2 I HLSNHFASB Mis I 35 24 l Sa l l konk msvenso 00 D M cr s 37 L EE OKA E ee ee pe 22U R732 yw Y pa er 23 HLS LF FB PA oc EE oo A8 s 1k82 2 SS ONP 33 HLSNHFNFB I POWER SUPPLY AND REGULATION cape 20KO SE KE ES J L5 i Ri zal I 2 5 esa AI a 12 TO No LH317T ED yoy L SSS Se eS SS in c SSS ZKU JE Se ee Se d 15V m Vin Vout T POT OO OTI 2 Eb R7 i AOT d i A TYPE CIRLUIT cs vi 8 G3 c iuo R750 DZU A TYPE OMITTED ON CAT No 350 vn l H tQ G2
65. Diagnosis eR e R n 7 5 Disassembly 7 6 Parts List issu Ry pes b OR Sa SRG Hoa DE ESR NOR f eee d ng mew ex om ais SECTION 8 f PPBENDICES qe Ru n ra Re AE SECTION 1 INTRODUCTION AND SPECIFICATIONS Bee 1 1 Introduction The Dolby Model 363 is a two channel signal processing unit designed to accept modules providing either A type noise reduction or SR Spectral Recording processing or a combination of the two The unit can operate with a mixture of module types or with only a single module installed The unit occupies 1U 44 mm 1 75 of standard 19 rack space Provision is made for up to two plug in modules for signal processing The Cat No 450 module provides A type noise reduction the Cat No 350 SR processing and the Cat No 300 contains switchable SR A type One module is required for each channel in the Model 363 Circuitry in the frame allows selection and indication of the type of processing provided in the installed modules Units prior to serial number 650 differ from later units in that no cooling fan was fitted and the remote control for noise reduction allowed only processing in out and not selection of type of processing The Model 363 is shipped with the following 1 Model 363 mainframe Signal processing modules as ordered Cat No 300 for switchable SR or A type processing Cat No 350 for SR processing only Cat No 450 for A type processing on
66. H R7 7 4l 10K K 27 COMPRESSORS 1C803C GND Sec 4 21 30 G4 R722 7K50 R1429 40538 24 um T 4 E RE R721 10K0 c BKE el uss ET dei 65 ME BAND4 25 SC w ls Je 1 is S EE H SE GND A 6 BAND3 Jaca V o earn BI CON I T R724 os J2 7 1 P EOS 3 poe R1223 1i 12v TO 38 pipe Win Vout AB SE 201 EE 5 100K isv 40 EH Lum i4 S CY 106 ERE cD R1128 l FP 97K6 EPEE TA ZZS Z EE zl Y 1C605A H TLO72 ee YE EE gom ar EE TAE um we Prid ua zi MANET PREZ PE E trs e ur tabou ee RS T F eu INPUT BANDPASS FILTER aen T 7 TRSUT I Toon Km SE E I H DECODE R809 e OUTPUT AMPLIFIER AND SWITCHING 8 6 dar E At wt 7 100N 100N kg ai E C724 200K R742 e701 R 022 c702mm Ty MES 5 el R708 R708 t TON I 10K c7a mos cep R705 C705 C706 Kl Bess 10K0 18 1C803A 2 ane 2 7 10K0 I i L 240KS INO Fab H I Wy R741 NORM pe I IC701A R704 2 I lop T d See I TL072 100K 10K0 R744 i ra e re Mer tie oh i Agen ea SCA SS D ONE J A sere alo w lps ouor KOL Lj crosa YE i 1C802A A _ 4 E TLO72 8 amp dBr RSS Pero n J DOLBY NOISE 0 S R740 EAR ioo C723 EE CI N t Pis T I 6 der DOLBY TONE 18 w C801 108038 27P R748 8 8 6dBr 4K99 7 GA R747 H OMITTED ON CAT No 350 RSA das Im 7 I mag 1040 I I Aa OUTPUT 1 i D 51k QF804 4K99 GE R808 10 TL072 Y E 200K Fef doll J PINK NOISE mue METER OUTPUT CIRCUIT d 8 6 d r
67. R circuits are built exclusively by Dolby Laboratories High Level Stages Mid Level Stages Low Level Stage Sliding Band Fixed Band Encoder Sliding Band Sliding Band Fixed Band Fixed Band Sliding Band Fixed Band Sliding Band Fixed Band LF HF Signal LF 8 HF Encoded signal Input Spectral Anti Anti to recorder Skewing saturation saturation Decoder Low Level Stage Mid Level Stages High Leve Stages Sliding Band Sliding Band Sliding Band HF LF LF amp HF Signal De Anti De Anti Spectral Output saturation saturation De Skewing BE 6822 Figure 5 10 Block Diagram of Dolby SR 5 7 20dB 0 dB 20dB DECODE 20 200 2000 20000 FREQUENCY Hz Figure 5 11 Low level response of Dolby SR processor 6 1 SECTION 6 CIRCUIT DETAILS EE a a ET ET OO 6 1 General The unit is functionally split into various parts Each audio channel comprises input and output stages together with various switches monitoring mode selection etc and a separate signal processing module In addition there is a common power supply and calibration generator section Apart from the modules and the power line transformer all the circuits are contained on one printed circuit board called the motherboard which may be removed for servicing The circuits on this main board are relatively simple in operation in the event of any f
68. R1020 l INDICATOR J10 bt MO I uer RIOS gt d E Ing e DS1001A I REMOTE A TYPE 109048 16904F v d INDICATOR J10 pace I lun R910 They BYPASS 909 A AA EE ra EE av d aet la EA A A AAA A SE A 1C904C du 2U2 R1021 1C9058 T 0 ai 8 FLASH gt 4 TC R918 m sog KO 1K0 on 10K 1510018 E BYPASS D913 se SET UP FLASHING PTA DOLBY RED d passt pou vc bebo o dE GEET EN PO OW ie OB MOK Ve zz ja See ee L ee e GREENS LEVEL CAL REC PLAY SHIT Dee raus R1022 Y SHIT 3 LOCAL REC PLAY SWITCH Kees CN M 0 8 PUE 1 5 g 8038 AR EPPO So R919 1K0 R1023 15V 30 1cso3c ly B DS1001D s 10 1 RED dB RECBYP shcs02 U MEL e 169034 i C1002A FULLY FULLY CD4011 h GC OFF ON i 10K0 1x INTENSITY REC J SERIAL NUMBER 650 ONWARDS 4 i i3 PROC CHANNEL SHOWN ADD 400 TO REFERENCE C902 A DS30S 1C902E DESIGNATORS FOR CHANNEL 2 A1C3735 REV 3 SH 2 Bees REC REDI JS J10 CHANNEL 2 PIN NUMBERS ARE SHOWN Do IN PARENTHESIS 00 R924 5 4 E 1C9028 pssog D912 PLAY DRN 6 9 6V 1C1201A gt 6V REF R1210 R1230 uud R1292 AN R1205 R1206 ISK 100N Imo TEEN R1231 204 T PERDRE 15V 1c1202C 3 E R1214 SIRO y 1C1204C 12 D21203 Ee 1C12030 v R1224 Ee E NEUE TL431 R1216
69. R1217 sieved Eech V externac T0 7 57K6 57K6 12 TONE R1207 mover 363 I i 1 1 14 l 13 R1218 Y 4K7S I I C1203 minC 1204 min C 1205 wx mm Ng 6N8 6N8 l l J10 a ZET o REMOTE SET UP l SWITCH ue R1109 I L d Ru SET UP a a a zl LUS R1212 pn 120K 112018 pulse C1114 R1213 b ne 220K D1105 LEI 10 SET UP C1209 R1234 15V 01101 FLASHING SC are 1C1203A R1242 R1243 BE638 YELLOW ezon 17 53 19K0 1380 1012038 e R1244 D21101 22V R1235 R1236 Ri237 R1238 4053 i E E PINK IC1204A R1239 C1214 22E NOISE R1233 Sisek 64 9 S20KS 4K75 13K0 680P EE 220K ja n m 1 1x 22v 8 p SENT C1210 Ciani C122 EIS tia s Vs 124 1241 C1215 dU e Tos JIN Sm SN GNS AMINI AT e HS af um 1 Vin Vout 15V 1N4004 R1103 R1248 9IK D1205 os ere dote PPE penso eros C1105 R1245 2200U ma 100N Bi 100N 01102 15K 1N4004 KON ov R1116 FUSE AND ix 1C1203C VOL TAGE 15K 2 16 R1247 SELECTOR ORN 120v PO 5 SY DOLBY R1105 1C12028 R1250 120k NOISE RIO 1108 T 4475 T 2 t ti c1109 S 1KI E t n 4 1U0 CH CH2 IW 100N 1x f 1N4004 oe R1252 R1251 91203 D1104 SIE Es 4K75 Vin Vout gt 15V ix 1N4004 Ln337 1C1102 Y C1110 C1111 a Saa p li u n NOTES UNLESS OTHERWISE SPECIFIED CND LINK reco E a 1 RESISTOR VALUES ARE IN OHMS 5i 10 OdBr 0 775V rms I I 2 CAPACITOR VALUES ARE IN FARADS 11 2777 DENOTES COMPONENTS EXTERNAL TO PZJI4 Ven 3 DIODES ARE 1N4148 KE CHASSIS 1 4 BIPOLAR TRANSISTORS ARE DOLBY STANDARD DEVICE
70. Re see ee e ance BU St d 3 1 3 2 Encode Decode Operation 3 3 3 3 Bypass seem hee tne th h 3 5 SECTION 4 LEVEL STANDARDIZATION 4 1 4 1 Level standardization for Dolby systems 4 1 4 2 Level standardization for Dolby A type noise reduction eee 4 2 4 3 Level Standardization for Dolby Spectral Recording 4 3 4 4 Set up LL ille hh 4 5 4 5 Alignment 4 5 SECTION 5 PRINCIPLES OF NOISE REDUCTION AND SPECTRAL RECORDING 5 1 5 1 Introduction iom e 544040 tag Xp RO eR Fe 46g Rede bdo ee e M d 5 2 General uisus ns BE EES ee deep ane 5 3 Dolby A type noise reduction sl R n 5 4 Dolby B type noise reduction allen e R e 5 5 Dolby C type noise reduction llis HR M SECTION 6 CIRCUIT DETAILS ouod b E RE ERE XXE RU Re EE BA General lt 5 20588864 ed bed Eu A edo PAW UA QC e RUE Kitt i 6 2 Motherboard Audio Circuits 6 3 Monitoring and Mode Switching 6 5 Metering 6 6 LOQ G Q 0 nep innui rere ep eene 6 7 Calibration GeneratorS ee 6 8 Power Supplies zana kh t t 6 9 Signal Processing Modules eee n ng SECTION 7 UNIT SERVICING IN CASE OF DIFFICULTY 7 1 Isolation of the Problem 7 2 Quick Check for Dolby A type NR and SR Processing 7 3 Testers cameras me dex D USER IE RO e x Real W Rao RC due te dLa 7 4 Motherboard Fault
71. S MOTHERBOARD CAT No 386 CONTROL BOARD DE Aa RED PNP BC416 2SA970 OR SIMILAR 12 THIS DOCUMENT APPLIES TO MODEL 363 WITH kee 106 En NPN BC414 2502240 OR SIMILAR SERIAL NUMBER 650 UPWARDS EARLIER UNITS 3 5 DUAL OP AMPS ARE 4558 WERE WITHOUT FAN ASSEMBLY AND USED THE l i PIN 4 15V PIN 8 15V CAT NO 373 CONTROL BOARD AICI IIS REV ISHAI L J 6 QUAD OP AMPS ARE TLO74 EXCEPT ICIO02 4 1402 WHICH ARE LH324 ed See EEN Ka PIN 4 1SV PIN 11 15V 7 1C902 904 1302 1304 ARE CD4049 PIN 1 15V PIN 8 LOGIC GND 8 IC903 1303 ARE CD4011 10905 1305 ARE CD4001 PIN 7 LOGIC GND PIN 14 15V 9 JCI204 IS CD4053 FINS 6 7 8 6V PIN 16 6V 6 11 SE CIRCUI SRZUMJ IN AT Ne SHOWING A REGUL ATUR MIMIC ECR TAT No Bac E CI SEDLCT ION AIT ap D CAT Nc 452 IRCL17S RF CVED EY S7 PROCESSING WITH If dic eS CTION SWITZHES HARD w MAN Figure 6 4 Cat No 300 block diagram 6 13 NOTES UNLESS OTHERWISE SPECIFIED 1 RESISTOR VALUES ARE IN OHMS 2 CAPACITOR VALUES ARE IN FARA
72. The output of the bottom circuit is fed to the reference terminal c of the top circuit It can be shown that the signal components at the output terminal b are those specified by Eq 1 The usefulness of the superposition technique can be appreciated by consideration of Fig 4 a and b The advantages of fixed band compressor circuits Fig 4 a arise from the fact that all signal frequencies within the band are treated equally in contrast with sliding band action Fig 4 b Thus the appearance of a dom inant signal component actuating the compressor results in a loss of noise reduction effect that manifests itself in a uniform manner throughout the band 2 dB in the example shown The loss is not concentrated in any particular frequency region as it is in sliding band cir cuits note the 5 dB loss shown in the example of Fig 4 b The main significance of this is that if the recorder or transmission channel has an error in gain and or frequency response there is no undue exaggeration of the error at other nondominant signal frequencies In sliding band circuits the amplification effect may be significant the midband modulation effect discussed in 2 Compressor 2 F2 V Compressor 1 Fy Fig 2 Basic layout of action substitution compressor 102 PAPERS In contrast the advantages of sliding band com pression and expansion circuits derive from the fact that all signal frequencies are not treated equal
73. ailure it would be possible to repair them in the field using identical components as replacements On the other hand the modules are extremely complicated and contain large numbers of components specially selected during the manufacturing process If there is a module failure the module must be returned to Dolby Laboratories or its agents for repair More notes on servicing are found later in Section 7 6 2 Motherboard Audio Circuits Refer to the block diagram drawing Figure 6 1 and detailed circuit drawings Figure 6 2 for the channel circuits and Figure 6 3 for the common circuits All diagrams are at rear of this section following page 6 4 Input signals from the console and the tape recorder enter the unit via separate XLR connectors and pass to identical balanced input amplifiers Some component values are 0 1 tolerance in this section to provide for excellent performance figures without the need to adjust values during manufacture Each input amplifier has a 10 turn potentiometer at its output to allow calibration of input levels to that required in the Dolby signal processing modules A hard bypass is provided by relays at the outputs which connect these outputs directly to the appropriate inputs when either the bypass function is selected or the unit is not powered Note that on powering the unit there is an initial one second delay provided by IC905B and IC1305B in the operation of these relays to prevent any thumps or clic
74. al is the least treatment The operating goal of the encoder is to provide fixed predetermined gains for all frequency components of the signal with corresponding atten uations in the decoder If a large signal component appears at a particular frequency or frequencies then the gains should be reduced at those frequencies only in accordance with predetermined compression laws for restoration of the signal during decoding In other words the compressor should try to keep all signal components fully boosted at all times When the boost ing must be cut back at a particular frequency the 100 PAPERS effect should not be extended to low level signal com ponents at other frequencies The audible effect of this type of compression is that the signal appears to be enhanced and brighter but without any apparent dynamic compression effects The ear detects dynamic action primarily by the effect of a gain change due to a signal component at one frequency on a signal component at some other frequency some what removed If the ear cannot detect dynamic effects in the compressed signal then 1 it is unlikely that noise modulation effects will be evident in the decoded signal and 2 it is unlikely that signal modulation effects will be evident in the decoded signal if there should be a gain or frequency response error in the recording or transmission channel In the SR process two new methods are used that greatly reduce the circuitry re
75. also need to record on the tape be sure to align the record section of the recorder as well 3 Restore the Model 363 unit to the signal path by releasing the Bypass buttons B PLAYBACK ALIGNMENT 4 Set the rec play toggle switches on each channel to the play position green LEDs on 5 Set the SR off A toggle switches to SR or A as appropriate for the incoming tape Replay the Dolby noise SR or Dolby tone A type from the tape and then press the Set up button When using Dolby SR the Cat No 300 or the Cat No 350 modules will go into the Auto Compare mode Auto Compare provides the user with an accurate audible verification that the playback frequency response and decode calibration levels are set properly regardless of the indication shown on the calibration displays Listen for level differences between the pink noise signal coming from the tape and the internal Dolby noise generator in the Auto Compare mode The two LEDs on the front of the Cat No 300 and Cat No 350 modules indicate whether the monitors are receiving the internal reference noise signal red LED or the Dolby noise from the tape recorder green LED 6 Adjust the play in trimmer on each channel of the Model 363 unit until the tape and reference levels audibly match in the Auto Compare mode or for equal brightness of the green LEDs 7 Release the Set up button and set the SR off A toggle switches to the off position 8 Switch the console meters to read
76. ance in the pass band in phase frequency re gion However in the stop band region the modulation control scheme causes the limiting threshold to rise and the degree of limiting to be reduced The possibility of doing this can be appreciated by consideration of the phasor diagrams of the two conditions shown in Fig 5 In the pass band in phase condition the side chain signal and the main path signal add directly Therefore a relatively low threshold must be maintained at all pass band frequencies Fig 5 a However in the stop band the effective amplitude contribution of the side chain signal may be minimal due to the phase difference between it and the main path signal Because of this it is possible to raise the threshold significantly _and to reduce the limiting strength once the desired amount of attenuation has been obtained at a given frequency Fig 5 b The result is that large signals in the stop band do not cause signal modulation in the pass band and consequently create an impairment of the noise reduction effect achieved during decoding Similar considerations apply in the SR sliding band circuits By way of introduction in the B type and C type sliding band circuits a variable filter follows a fixed filter which has proved to be an efficient and reproducible arrangement At frequencies outside the pass band a pure two pole filter would result in overall amplitude subtraction from the main path signal because of the lar
77. and past president of the AES and a recipient of its Silver Medal Award He is also a fellow of the British Kin ematograph Sound and Television Society and of the SMPTE and a recipient of its Samuel L Warner Me morial Award the Alexander M Poniatoff Gold Medal and the Progress Medal In 1979 he and his colleagues received the Scientific and Engineering Award of the Academy of Motion Picture Arts and Sciences In 1986 he was awarded the British title of Officer of the Most Excellent Order of the British Empire OBE J Audio Eng Soc Vol 35 No 3 1987 March An Audio Noise Reduction System RAY M DOLBY Dolby Laboratories London England A noise reduction system which is suitable for use with high quality audio recording or transmission channels is described A special signal component derived from four band splitting filters and low level compressors is combined with the incoming signal during recording or sending During reproduction the additional component is removed in a complementary way any noises acquired in the channel are attenuated in the process Practical features of the system include 10 dB unweighted noise reduction imperceptibility of signal modulated noise effects level frequency response overall accuracy of reproduced signal dynamics low distortion low internal noise level and stability of characteristics INTRODUCTION In an audio recording or transmis sion channel noises of varying
78. and play One Model 363 unit two channels can therefore be used for a two track recorder and 2 Dedicated Encode or Decode operation where the channels are permanently assigned Two Model 363 units are required for a two track recorder 3 1 Switched operation The processor channel is switched either into the RECORD signal path or into the PLAY signal path The switching may be controlled manually from the front panel of the Model 363 or more often automatically by the tape recorder Controls on the Model 363 are identified on the fold out drawing at the end of Section 4 Since each channel of the recorder has only one processor it is not possible to decode the signal off tape during recording In normal operation the console output is heard during recording A switch labelled normal check tape is provided to allow the operator to hear the playback signal directly without any processing In both record and play modes the encoded tape can be monitored by using the check tape position of the switch The switch should usually be left in the normal position The output signal is then the unprocessed line in signal while recording or the properly decoded signal during playback This switched operation is commonly used with two track tape recorders in recording studios and sometimes with video recorders 3 1 1 Signal paths in RECORD Model 363 Processing module out Output 1 SS To Recorder Calibration
79. complementary compressors controls cross fade meter dBu de emphasis disassembly Dolby level nicks noise noise synchronization tone dual path encode decode operation encoder equalization escutcheon expander extender extrusion fault finding FET filter CCIR ARM fluxivity fuses generators calibration grounding headroom heatsink high level signals identification module indicator auto compare installation 5 3 Appendix part 4 1 3 8 3 2 1 3 5 4 1 4 3 4 5ff 5 5 5 6 Appendix part 3 4 3ff 4 3ff 1 3 8 3 5 4 7 1 7 2 3 5 6 1 6 3 7 1 5 5 4 1ff 4 3 4 4 6 2 4 5 8 3 7 5 1 2 3 1ff 5 1 5 2 4 1 5 4 5 5 8 3 5 2 1 2 2 10 6 2 1 2 5 1 7 3 4 2 4 4 4 6 4 3 6 3 4 3 4 4 4 6 6 2 4 4 4 2 4 3 4 4 4 6 6 2 5 2 3 3 5 1ff 5 1 7 5 5 1ff 7 1 7 3 7 5 7 2 6 1 6 2 8 3 4 2 1 1 1 4 2 2 6 2 7 5 1 3 4 2 8 3 7 4 5 3 2 6 6 4 2 1ff labels identification level Dolby low level signals mainframe malfunctions metering mimic auto compare modulation noise module signal processing block diagram circuit diagram motherboard nicks Dolby noise Dolby pink weighting noise reduction A type B type non complementary effects normal check switch operation bypass encode decode switched opto isolator pinking filter power supplies pre emphasis quick check re assemble regulators remote operation
80. cuitry takes control In contrast with the high frequency situation the low frequency general strategy is to derive the primary overshoot suppression signals from signal points that do not include any control circuit frequency weighting This is because the required control circuit weighting networks of the low frequency stages are low pass in character resulting in delays The high pass networks used for control circuit weighting in the high frequency stages do not introduce delays However because of the lack of a weighting factor in the primary overshoot suppression signal there is no inherent tracking between the steady state and overshoot suppression thresholds of the circuits involved particularly in the stop bands Therefore further modulation control techniques are employed to obtain the required tracking The secondary overshoot suppression signals are derived from a point in the fixed band circuitry that provides adequate tracking in both the fixed and the sliding bands Referring to Fig 11 the fixed band primary overshoot suppression signal is generated by passing the variable attenuator output through a 200 Hz single pole high pass filter middle of figure This filter reduces the influence of the primary overshoot suppressor at low frequencies allowing the more gentle low frequency overshoot suppressor to take over the transient control function The signal is rectified and then opposed by modulation control signal MC7 a
81. degrees of avoidability arise in the channel itself Apart from correction at the source any scheme to reduce the audibility of such noises which may include hum crosstalk print through hiss and other undesired signals can be classified broadly into one of two types Fig 1 a non complementary in which the signal is post processed only thereby producing an overall alteration of the signal while reducing noise and b complementary in which both pre and post processing are used the attempt being to produce no overall alteration while reducing noise Examples of non complementary noise reduction sys tems include simple tone controls and filters In more sophisticated forms the filtering action may be made dynamic as in the Dynaural system of Scott or the multiple band diode expander system of Olson Auto matic signal controlled attenuators which function simi larly on a wideband basis have also been described 23 The noise problem has also been attacked in comple mentary ways the simplest method being the use of optimized equalization characteristics 5 Various types of compressors and expanders have been devised both of the instantaneous variety employing nonlinear networks and of the syllabic type in which linear variable gain devices are controlled in accordance with the signal envelope9 in some systems pilot tones are used in the expansion process 10 11 An electronically switched two channel low level high
82. der i H 1 i 1 TH 1 E 5 sat r JA DCH H mam UBER REN ji 10 20 Hz 50 100 500 1000 2000 5000 10000 20000 40000 o if Figure 5 9 Low level C type encoder response Betacam and U matic are trademarks of Sony Corporation and MII is a trademark of Matsushita Electric Industrial Co Ltd 5 6 5 6 Dolby Spectral Recording Introduced in 1986 Dolby Spectral Recording SR is a professional system which combines all the advantages of fixed bands as in A type with the spectral adaptation of sliding bands as in B type and C type It also employs spectral skewing and anti saturation but applied at low frequencies as well as high SR uses three high frequency and two low frequency stages in tandem with a crossover at 800 Hz see Figure 5 10 Together they result in a reduction in audible noise of 24 dB see Figure 5 11 taking into account the frequency dependent sensitivity of the ear The SR processor response adapts to the input spectrum to obtain full unchanged boost except in the immediate neighborhood of dominant frequencies In this way the system reduces not only noise but other unwanted signals such as tape modulation noise and distortion products introduced between the encoder and decoder The spectral skewing and multi stage processing give greater tolerance to level and frequency response errors compared with A type nr Dolby S
83. ditory window 120 dB 100 80 60 40 20 20 200 2k 20k One way to define ideal sound reproduction is to show the limits of the human hearing system as a window as we did for analog tape recording in Figure 1 Such an auditory window is sketched in Figure 7 The top of the graph corresponds to 120 dB a continuous sound pressure level that is unbearably loud at 5 kHz sound changes to pain at about 110 dB It is safest of course to leave a margin between such a level and the highest continuous sound level one aims at reproducing The boundary of the window near the top of the plot is 6 dB below the threshold of pain at each frequency The bottom of the window is the threshold of hearing sounds at this level are just audible to a listener with sensitive hearing The level of the background noise in a very quiet recording stu dio may be 10 to 15 dB A recording system with a window like that shown in the figure could be played back without audible noise or overload even if the highest level signals were literally at deafening levels Playback of an analog recording We can learn more about tape recording by superimposing the analog tape window on top of the auditory window Sliding the analog window up or down corresponds to playing a tape at higher or lower level In Figure 8 we have set the playback gain so that the maximum level signal that can be put on the tape lines up with a continuous sound pressure level of
84. e amount of boost of the dominant signal The boosting of the nondominant signals is maintained toward the spectrum ends even though the level of the dominant signal is relatively high in the range 30 dB to 20 dB This boosting action spectrally tracks the dominant signal frequency or frequencies It is advantageous to have a steeply rising boosting effect away from the frequency of the dominant signal component In this connection the SR circuit profits from the steepness enhancing effect of cascaded stages The low frequencies have two stages of steepness com pounding the use of three stages at high frequencies further improves the effect These characteristics are particularly evident in the high level areas of the probe tone curves The curves show that the encoder circuit tends toward keeping all low level signal components boosted at all times Only those components above the threshold are subject to a reduction of boosting With regard to the 20 200 2000 FREQUENCY Hz 20000 Fig 15 Low level encoding characteristic in the presence of an 800 Hz signal at the levels indicated 116 PAPERS overall encode decode system the advantages of this type of characteristic are 1 a powerful noise reduction effect in the presence of signals and 2 a relative tol erance of level and frequency response errors in the channel between encoder and decoder 4 4 Audible Results As with the investigation o
85. e for the development of the electronic aspects of the Ampex video tape recording system Awarded a Marshall Scholarship followed later by a National Science Foundation Graduate Fellow ship he left Ampex in 1957 for further study at Cam bridge University in England where he received a Ph D degree in physics in 1961 Working in the Cavendish Laboratory at Cambridge from 1957 63 Dr Dolby studied various properties of long wavelength x rays particularly as applied to elec tron microprobe analysis In 1961 he was elected to a Research Fellowship of Pembroke College Cam bridge During his last year in Cambridge he was also a Consultant to the U K Atomic Energy Authority In 1963 he took up a two year appointment as a United Nations Adviser in India serving as a member of an advisory team to the Central Scientific Instru ments Organization In 1965 Dr Dolby left India and returned to England to establish Dolby Laboratories He holds a number of patents and has written papers on video tape recording long wavelength x ray micro analysis and noise reduction Dr Dolby is a member of the Audio Engineering Society INDEX A Te cm nu nnn gt zzsaenwww n A type A weighting accessories alignment anti saturation Auto Compare Auto Compare mimic Average responding meter B type back to back testing calibration concepts generators methods CCIR ARM chassis disassembly check tape compandor
86. e module removed and the extender inserted a signal at either the From console or the From rec input connectors as determined by the rec play switch should appear simultaneously at the To rec and To console outputs and also if at or near Dolby level on the calibration display If the signal does not appear at any of these places in the normal record play modes then there is a problem in the motherboard The processor modules have been designed for accuracy reliability and long life There is no need for adjustment the critical parts of the circuit are pre adjusted during manufacture with selected fixed value components using custom designed test equipment and procedures These selected components determine the standard Dolby A type and SH processor characteristics and any attempted repair by the user may result in degraded performance The modules should be returned to either Dolby Laboratories or the local agent for exchange and or repair 7 2 Quick Check for Dolby A type NR and SR Processing A quick test of the Dolby processing may be carried out in the field to check the outline characteristics of both the record encode and play decode modes Tones at various levels and frequencies are sent to the Model 363 and the outputs at the To recorder and To console are measured Alternatively if as is usually the case two channels are available simple back to back checks can be made by putting one channel into the record mode and the othe
87. e signal attenuation this ratio creates an end stop effect on the sliding band ac tion The result is smoothed first by a time constant of about 5 ms different for each stage and finally by a time constant of 80 ms The smoothed control signal 110 PAPERS is then used to control the sliding band variable re sistance RV A single control circuit suffices in the sliding band circuit because the 10 kHz high pass control weighting network tends to offset the effect of complex signals on the modulation control voltage de veloped MC1 A modification is made in the sliding band control characteristic at low levels Signals from the fixed band circuit are combined in opposition with the sliding band output signal combining circuit at right of Fig 9 The effect is in the direction of simulating the de rivation of the sliding band control voltage from the voltage across the sliding band variable filter only that is from the voltage across RV This tends to raise the sliding band threshold at high frequencies which reduces unnecessary sliding of the band The 10 kHz control weighting network provides the correct amount of control signal for the variable filter at medium and high levels but it produces the undesirable side effect of lowering the threshold at high frequencies The dif ferential control signal derivation method counteracts this threshold lowering effect The overshoot suppression O S arrangements are also
88. ection of the response of the recorder To prevent these errors Dolby noise is recorded at a lower level When the Set up button is pressed and SR is selected on the Model 363 the calibration display circuit gain is increased so that Dolby noise will still read Dolby level on the Model 363 display even though it is not recorded at Dolby level on the tape In addition the Dolby noise is filtered in the meter circuit only to remove the low frequency components of the pink noise which would otherwise give a display signal which would be difficult to read As well as providing level information Dolby Noise also makes checking the frequency response quick and easy Audibly comparing the Dolby noise recorded on tape with a Dolby noise generator will show up any differences caused by either gain or frequency response errors in the complete record replay chain this test is remarkably easy to perform and is extremely critical This comparison is made even easier by a feature called Auto Compare During Auto Compare the monitor output of the Model 363 switches between the the signal replayed from tape and the internal Dolby noise generator in the sequence shown below In general if Auto Compare Sounds OK all aspects of the complete recorder Model 363 chain are operating and are calibrated correctly Note that the noise from tape is Dolby noise and so has 20 ms gaps whereas the reference noise from the generator has no gaps This pattern makes it easy t
89. ects audible or inaudible are possible Dynamic action of Dolby SR at moderate levels FREQUENCY Hz If the signal increases in any part of the spectrum some adjust ment of gain must be made to ensure that overload will not take place This is done in a very gradual way by Dolby SR so gradual that there is no danger of producing audible modulation of the signal or any other audible effect The principal mechanism of Dolby SR is a group of ten fixed and sliding band filters with gentle slopes Those with fixed bandwidth are electronically controlled to vary their gain those with fixed gain can be adjusted to cover different fre quency ranges By selecting and combining from the group the Dolby SR control circuit can create an infinite number of filters through which the signal must pass before it is recorded During playback filters are automatically created that are the exact opposite of those used during recording Filter selection and adjustment is controlled by a continuous analysis of the signal spectrum and a process called action substitution Action substitution determines which of the two types of filters will predominate and how each must be adjusted to produce the optimum composite filter Figure 12 Even when the signal level increases substantially the system is designed to deviate as little as possible from the fixed character istic shown in Figure 11 in accordance with the principle of least treat
90. educes the level of the incoming signal at extremely low and extremely high frequencies Spectral skewing desensitizes Dolby SR to minor aberrations in tape to head contact and azimuth align ment which might cause fluctuations in high frequency response and to head bumps or low frequency variations in alignment tape levels Although these effects are often inaudi ble they can disturb the operation of other signal processing systems Recording complex signals at high levels 15 0 dB b a 10 20 20 200 2k 20k Hz 16 Dynamic Processing input 4 kon Anti saturation Network output Headroom is as important as any other property of a recording system Analog tape recording as engineers know has a gradual or soft clipping characteristic In digital recording there is no saturation region at all there is simply clipping in which the same digital word is recorded over and over as long as the signal remains above the limit The Dolby SR circuit contains a feature designed specifically to deal with extremely high levels at low and high frequencies The anti saturation charac teristic is sketched in Figure 15 and the circuit configuration shown schematically in Figure 16 Low level signals pass through the side chain for processing as signal level increases an increasing proportion of the signal follows the direct path which applies no dynamic signal processing By introducing attenuati
91. en the modulation control signals and the signals in the control circuits of the several stages are also optimized In the generation of MC8 which is used for low frequency stage overshoot suppression inhi bition under high frequency transient signal conditions J Audio Eng Soc Vol 35 No 3 1987 March SPECTRAL RECORDING PROCESS the influences of the HLS and MLS stages are unde sirable MCS8 is therefore derived from the first feed point of the stages just following the spectral skewing networks Fig 1 also shows the distribution scheme of the modulation control signals MC1 MC3 are used for the high frequency stages MC5 MC are used for the low frequency stages In Fig 8 the basic layout of the modulation control circuits is shown MCI controls the high frequency sliding band circuits The signal from the takeoff point is fed through a 3 kHz single pole high pass filter full wave rectified all rectifiers in the system are full wave and fed in opposition to the control signals generated by the high frequency stages An all pass phase shift network is used to optimize the phase of the MC signal in relation to the stage control signal at low frequencies this reduces control signal ripple MC is also smoothed by a two stage I ms integrator and is employed as MC2 to oppose the operation of the high frequency sliding band overshoot suppression circuits the ov ershoot suppression thresholds thereby track the steady state
92. er gets a little brighter eventually only the one green LED is illuminated As the signal continues to change this green LED in turn cross fades into a red The low red LED goes out when the signal level is about 11 dB below Dolby level Exact levels and cross fade appearance are given on the detailed circuit diagram 6 6 Logic The logic circuits accept simple control voltages from the internal and external control switches and apply logical rules to provide appropriate control signals to the various FET switches note that in addition there are logic circuits in the processing modules which also enter into the circuit functional design Signals from the modules are also used to indicate which module is in place and to prevent false indications being given For example if a Cat No 350 module is being used which only has SR capability there will be no ground at pin 32 of the module This information is sensed by IC905c so that if A type is selected by the front panel switch the A type light will not be illuminated and no processing will be selected in the module 6 7 Calibration Generators Circuits provide Dolby tone Dolby noise and pink noise for use in the calibration of the Model 363 to other equipment and signal routing systems The level of these signals is 388 mV note that the value read for pink and Dolby noise is critically dependent on the type of meter used Dolby tone used for calibration and identification of A
93. erties Steady State Properties Referring to the steady state transfer characteristics shown in Fig 4 the noise reduction requirement to gether with the desirability of interfering as little as possible with high level signal components dictates a reproducing expansion curve of the type shown in Fig 4b that is the gain at low levels must be reduced while a unity gain condition should prevail at high levels The required differential component transfer characteristic shown in Fig 4c is then determined being linear up to the compression threshold rising slightly with increas ing input and finally decreasing with larger inputs In practice such a characteristic is formed by deriving the compressor control voltage from a combination of feed forward and feedback signals The recording compression transfer characteristic shown in Fig 4a is complementary to the reproducing characteristic amplifying low level signal components in order to compensate for the corresponding deficiencies produced by the noise reduction action during reproduc tion Comparison of the differential method of forming the compression and expansion laws with the conventional approach depicted in Fig 2 shows that the scheme has several advantages Nonlinear and modulation distortion are both reduced since the compressor limiter contri bution is negligible at high levels System noise problems AN AUDIO NOISE REDUCTION SYSTEM are alleviated the
94. es Dolby Spectral Recording Dolby spectral recording Dolby SR is a new professional studio mastering system that yields recordings with exceptional purity of sound Several important technical advancements are intro duced in the new system One is a substantial extension of available headroom which allows the use of a uniformly high maximum recording level at all audio frequencies Another is the practical elimination of the influence of noise and non linearity on the reproduced sound These advancements are achieved by new circuit functions adaptive to the signal spectrum and by the consistent application of minimum processing to the signal the principle of least treatment Recording with Dolby SR is extremely simple It can be used with any modern professional recorder installation is rapid operation is simple and reliable and routine maintenance is unnecessary Its unique Auto Compare feature allows immedi ate dependable validation of recording system performance without additional instruments or delays in the studio work schedule Plug in module construction and compatibility with existing Dolby mainframes provide convenience and economy for studios that use Dolby SR The principle of least treatment The ear is the final destination of all audio signals and the most sensitive instrument for their analysis An ideal audio device or system would impose no audible limitation on the signal passing through it The design of
95. esponse The signal is coupled through a diode to the fixed band final integrator circuit and is also used suitably biased for secondary overshoot suppression in the sliding band circuit also coupled through a diode The low frequency overshoot suppression signal bottom of Fig 11 is developed by tapping the rectified but unsmoothed output of the pass band control circuit of the fixed band circuit The signal is opposed by MC6 to desensitize the circuit to high level high fre quency components The signal is further opposed by the resulting fixed band smoothed control signal in a 114 PAPERS negative feedback fashion When the fixed band control signal has risen to a sufficient level there is no further need for any low frequency overshoot suppressor ac tion The signal is then highly amplified and limited peak rectified and smoothed by an integrator with about a 20 ms decay time constant The resulting high am plitude pulses are fed through a differentiating network with a time constant of the same order as the integration time constant to provide low frequency overshoot sup pression impulses of defined strength for distribution to the fixed band and sliding band final integrators via high value resistors and series diodes The result is a decaying constant current charging of the ca pacitors of the final integrator circuits This is in contrast with the higher peak currents and correspondingly more abrupt contro
96. f other psychoacoustic devices such as loudspeakers measurements of the SR process can provide only a partial characterization The rest must be obtained by detailed listening using a wide range of source material and a variety of re cording and transmission media Generally the audible encoding effect of the system is to create a dense bright sounding signal as sent to the recording channel but with little or no apparent dynamic action Harmonics overtones and small scale components of the sound including noise are all en hanced At high signal levels the antisaturation char acteristics cause a high and low frequency audible dulling of the encoded signal when applied to the re cording channel this treatment results in a significant reduction of recording distortion The overall audible encoding and decoding effect of the SR process is simply to create a clean and accurate sounding replica of the input signal Tape bias noise and modulation noise are significantly reduced Also a reduction of intermodulation distortion is achieved by the low frequency noise reduction capabilities of the process as well as by the effects of the antisaturation characteristics used during encoding Furthermore the decoding portion of the system re duces harmonic distortion generated by the recording channel Steady state third harmonic distortion is typ ically reduced to less than one half fifth harmonic distortion to less than one quarter
97. f the other The incoming signal is fed through an 800 Hz single pole band defining filter This is followed by a 400 Hz single pole filter which attenuates the low frequency signal levels fed to both the fixed and the sliding band circuits this reduces waveform distortion and complex signal transient distortion at high signal levels The filter also forms part of the fixed band control signal weighting network The output signal is taken from the sliding band stage and is fed through a 400 Hz network having a reciprocal characteristic to that of 3 kHz All Pass bi gt ES id ims To HF LL MC2 TT Stages 400 Hz EN gt MC3 800 Hz MC1 MC Input op 200 Hz Pass po Network N MC4 2ms LL MC5 TT 1 6 kHz bl ToLF E Hz MC6 Stages 2ms E MC7 TT MC8 Input FEL Double Differen MC8 tiator Fig 8 Basic layout of modulation control circuits These circuits reduce modulations of the gains and frequency response characteristics used in system especially at signal levels significantly above the compressor thresholds 108 J Audio Eng Soc Vol 35 No 3 1987 March SPECTRAL RECORDING PROCESS PAPERS weJdeip xoo q 28E1S 2uanba4I u H 6 314 ZHAO L l ft gt T ZH 008 ZH 00 10128 35 EDN 4H unulixelA N i S O pueg paxi4 Auo soi S O 7 Aaepuoras sau LL ZOW Y TT S O Hewud ul zH 00 2H 008 pueg ButpIIs 109 J Audio Eng Soc Vol 35 No 3 1
98. f the unit 2 3 2 9 Connection Diagrams for Specific 2 Track Recorders use Otar channel Ampex VPR 9 VPR 6 VPR 80 y Mode 363 I 9 Audio 1 anc 2 ccnnectors I where fitted I I T Za 7 I r rec Tie 9 12 z 4 8 l A C3856 b REV A Orari Mode 363 J9 dg 2 6 link 1 l La 3 l I I I o 4 8 P play Rd rec l 5 9 l il l i connector part number CN 220220 I MTR 10 12 MTR 20 AcC3808 b REV B pins 1 2 1 2 pins 3 4 tu pins 11 12 5 6 pins 13 14 7 8 2 10 Panasonic AU 650 Model 363 9 16 pin plug r o 2 6 Ext NR GG n 10 11 l I a 4 8 transistor O rec EE L a 4 ote ground return N groune must be provided by screen of audio cables 5 8 een o 2 la 1 Ji J5 1 4 A IN 3 9 o O2 A l ioi J2 J6 t 3 2 12 o o2 gt 7 13 of i D IBN 3 J3 J7 3 16 of Do 2 l 1 J4 UB l 3 A C3855 REV A from console to rec from rec to console Sony MC JH 110 series Model 363 J9 l lay I I Pla 3 3 7
99. g In addition the processing is dependent on both the level and frequency of the signal being recorded This gives great flexibility in the way that the processing adapts to the incoming signal but it means that for correct operation the levels in the playback processor must be the same as those in the record processor In most studios all this really means is that the tape recorder should be at unity gain which is the normal situation To ensure that the play processor gets the same signal level as the original record processor which often is in a different studio it is necessary to record a test signal onto the tape to indicate how the Dolby processing was set when the recording was made For Dolby A type this signal is called Dolby tone and for Dolby SR it is called Dolby noise 4 2 4 2 Level standardization for Dolby A type noise reduction Every Dolby A type noise reduction unit contains an oscillator that produces a distinctive reference signal at a closely defined internal level called Dolby level and a calibration display to indicate this Dolby level When the Set up button is pressed Dolby tone is fed into the record signal path and the calibration display indicates the level in the play processor In products designed before the introduction of Dolby SR the Set up function is controlled by a button labelled Dolby tone Dolby tone is made distinctive by frequency modulating the basic 850 Hz frequency upward by 10 for 30 m
100. ge phase angles created Therefore the type of filter which has been employed is only quasi two 103 DOLBY pole a single pole fixed filter plus a variable shelf characteristic The same type of arrangement is used in the spectral recording process with a one octave difference in the stop band direction between the variable filter turnover frequency under quiescent conditions and the fixed filter cutoff frequency Above the threshold at a par ticular frequency the variable filter slides to the turnover frequency needed to create the overall main path plus side chain signal compression law As the input level rises and once an overall gain of about unity is ob tained when the variable filter cutoff frequency is about two to three octaves above the dominant signal frequency there is no reason for further sliding of the variable filter At this point the modulation control arrangement counteracts the sliding As with the fixed band circuits this technique prevents unnecessary modulation of the signal The above effects in both the fixed and the sliding bands are created by circuits called modulation control circuits Suitably filtered or frequency weighted signals from the main signal path are rectified and in some cases smoothed and are fed in opposition to the contro signals generated by the control circuits of the various stages The result at higher signal levels relatively beginning at about 20 dB above the
101. gments is accomplished in two ways First the reference segments are 4s of continuous pink noise and the tape segments begin with a nick have a nick in the middle and end with a nick this time sequence is easily identified with a little practice Second colored lights identify the two different signals The new calibration facility gives the recording and production personnel a useful control of the recording process At any time a check of the recorder can be made the result can be heard immediately and conclu J Audio Eng Soc Vol 35 No 3 1987 March SPECTRAL RECORDING PROCESS sions drawn about whether adjustments might be nec essary With tape and signal interchanges it is possible to tell quickly whether there is any error or misunder standing about levels equalization azimuth and the like If the original recording of calibration noise stays with the tape the quality of the ultimate playback even after copying can be retained Thus the compar ison function serves to ensure that the recorder and spectral recording process provide on a routine basis the signal quality and reliability of which they are ca pable 6 CONCLUSION A new professional recording format designated spectral recording has been described The objective of the new encoding and decoding process is to record and reproduce audio signals with a high degree of au dible signal purity The system employs a dual path multilevel staggered act
102. he audible noise floor of a Dolby SR recording is normally the noise of the microphone amplifiers console electronics or electronic instrument amplifiers In a studio with playback gain set as shown in the figure and no signal present the background noise of a Dolby SR recording is below the threshold of human hearing and cannot be operationally improved Ee GE 22 Dolby SR works with every modern recorder Dolby SR processing can be used with any modern professional analog tape recorder or high quality audio communication system This means that nearly every recording studio and communications facility in the world is already equipped to install and use Dolby SR Often changing over to Dolby SR will only require the removal of the Dolby A type noise reduc tion modules already installed and their replacement by Dolby SR modules of the appropriate type Its practical design makes recording with Dolby SR easier than recording without it and editing mixdown copying maintenance and other studio procedures are generally simpler because of the features of the new system Full information on Dolby spectral recording equipment is available on request from Dolby Laboratories Inc Technical note The following list contains sources used to prepare the illustrations in this booklet and suggestions for further reading Copies of publications marked are available on request from Dolby Laboratories Inc Beranek L
103. higher order har monics are even further reduced Thus especially if the recording medium has a hard clipping characteristic the subjective cleanliness of the signal at high recorded levels is significantly improved 10 FREQUENCY Hz Fig 16 Low level encoding characteristic in the presence of a 3 kHz signal at the levels indicated J Audio Eng Soc Vol 35 No 3 1987 March PAPERS 5 CALIBRATION The spectral recording calibration procedures are conceptually similar to those of the A type B type and C type systems That is signal levels in the decoder circuit ideally should match those in the encoder circuit even though the SR process has been designed to be more tolerant of gain and frequency response errors than these previous systems For tape interchange standardization it is also preferable if at least within a given organization the reference level of the encoder and decoder corresponds to a known and fixed flux level Whether or not a standardized flux is used for this the matching of the decoder to the encoder is accomplished by a calibration signal generated in the encoder and recorded on the tape this allows the tape replay gain to be set correctly using the meter in the decoder unit Most problems in the studio use of noise reduction and indeed analog recording in general can be traced to incorrect level settings and or frequency response errors in the recorder This may be becau
104. hould have a low internal noise level and should not generate any additional perceptible noises in the presence of signals 4 All the above requirements should be met in tandem operation of the system e with multiple processing and de processing of the signal Requirements Relating to Recording or Transmission Channel 1 The output from the recording sending processor should be suitable for transmission through one channel of normal audio bandwidth 2 Correct operation should not be dependent upon linear phase frequency response in the channel 3 Normally encountered errors or fluctuations in gain and frequency response of the channel should not cause audibly significant changes in the system output 4 The system should not modify significantly the over all steady state or transient overload characteristics of the channel Interchangeability Requirements 1 The operating characteristics of the system should be fixed and reproducible 2 The processing units should be sufficiently stable with time temperature and other factors to permit inter change of recordings or channels Noise Reduction Requirements 1 The amount of noise reduction should be percep tibly similar for all types of noises encountered 2 The noise reduction action should be perceptibly free of signal modulated noise effects with any normally encountered combination of program material and noise COMPANDORS Of the possible noise reduction methods
105. iatura deve avere un collegamento di messa a terra ben eseguito anche la connessione rete deve avere un collegamento a terra o Ilcavo di alimentazione a corredo dell apparecchiatura deve essere collegato come segue Filo tensione Marrone Neutro Blu Massa Verde Giallo AVISO IMPORTANTE DE SEGURIDAD Esta unidad cumple con la norma de seguridad EN60065 La unidad no debe ser expuesta a goteos o salpicaduras y no deben colocarse sobre el equipo recipientes con liquidos como tazas de cafe Para asegurarse un funcionamiento seguro y prevenir cualquier posible peligro de descarga o riesgo de incendio se han de observar las siguientes precauciones o Aseg rese que el selector de tensi n est ajustado a la tensi n correcta para su alimentaci n CED o Aseg rese que los fusibles colocados son del tipo y valor correctos tal como se marca en la unidad o La unidad debe ser puesta a tierra conect ndola a un conector de red correctamente cableado y puesto a tierra o El cable de red suministrado con esta unidad debe ser cableado como sigue Vivo Marr n Neutro Azul Tierra Verde Amarillo VIKTIGA SAKERHETSATGARDER Denna enhet uppfyller s kerhetsstandard EN60065 Enheten far ej uts ttas for yttre verkan samt f rem l inneh llande v tska s som kaffemuggar far ej placeras p utrustningen F r att garantera s kerheten och gardera mot eventuell elchock eller brandrisk m ste f ljande observeras o Kontrollera att sp nningsv ljaren r
106. ight be thought to be deficient in dynamic range at very low frequencies because saturation occurs at lower tape flux levels in that part of the spectrum However the opening in the analog window at low frequencies is actually larger than the opening in the auditory window The same is true at high frequencies If signal components at different frequencies were simply recorded at different gain settings the effective dynamic range of analog recording could be extended considerably What Dolby Spectral Recording Does Dolby SR at low levels FREQUENCY Hz The simplest way to suppress noise and other low level record ing defects is to use as high a recording gain as possible From what we have already seen an even better idea would be to use the optimum gain in each part of the spectrum This is what Dolby SR does at very low signal levels The result is a form of fixed equalization that does not change as long as the signal level stays below a certain threshold When the recording is played back the same equalization is applied in reverse and any background noise is lowered by the same amount Figure lla The upper curve in Figure 11b is the noise level of a typical professional tape at 15 ips The lower curve shows the change in this level that results from the use of Dolby SR fixed low level equalization Since there are no dynamic changes taking place the equalization is fixed no dynamic side eff
107. ignificant reduction of distortion and noise in both the absence and the pres ence of signals The process has a number of layout and operating characteristics in common with the A type 1 B type 3 4 and C type 2 noise reduction systems The SR process takes these developments considerably fur ther in the same general direction An understanding of the new system requires reference to the technical papers on these earlier systems the C type paper 2 being particularly relevant Referring to Fig 1 which will later be described in detail a main signal path is primarily responsible for 99 DOLBY conveying high level signals A side chain signal with the SR process characteristic is additively combined with the main signal in the encoding mode and sub tractively in the decoding mode whereby an overall complementary action is obtained The SR stage layout resembles that of the C type system except that three levels or stages of action staggering are used high level mid level and low level HLS MLS LLS There are various advantages arising from the use of multilevel stages including accuracy and reproducibility low distortion low ov ershoot and action compounding for good spectral discrimination For the high level and mid level stages both high frequency and low frequency circuits are used with a crossover frequency of 800 Hz The low level stage is high frequency only with an 800 Hz high pass charac
108. in the lower 35 40 dB of the dynamic range Similarly there is no action in the top 25 dB of the total dynamic range A linear dynamic characteristic prevails in these two regions a bilinear characteristic 4 1 2 High Frequencies Dynamic action occurs in the range from about 62 dB to 5 dB That is there is no action in the lower 20 25 dB but full boosting or the top 25 dB of the dynamic range a linear dynamic characteristic in these regions In the intermediate level regions of dynamic action the effects of the multilevel stages are joined together to create a compression ratio of about 2 1 Referring to the high level low and high frequency portions of the curves the effective antisaturation of the system can be seen with the combined effects of the spectral skewing and antisaturation networks the SR stages and about dB of wide band level com pensation built into the coefficients of the stage signal combiner Fig 1 The overall result is an antisaturation effect of about 2 dB at 5 kHz 6 dB at 10 kHz and 10 20 dB 0 dB 20dB DECODE 20 200 20000 2000 FREQUENCY Hz Fig 13 Low level subthreshold encoding and decoding characteristics J Audio Eng Soc Vol 35 No 3 1987 March SPECTRAL RECORDING PROCESS dB at 25 Hz and 15 kHz At high frequencies this amount of antisaturation significantly reduces distortion re duces signal compression effect
109. introduced in 1966 and manufactured only by Dolby Laboratories It provides 10 dB of nr over most of the audio spectrum rising to 15 dB at very high frequencies The requirement for variable frequency response is met by dividing the frequency range into four bands each with its individual compressor see Figure 5 5 A high level signal in one band does not affect the other bands where noise may not be masked and hence in general the system does not give audible noise modulation Band 4 9 kHz High Pass Filter Compressor Band 3 3 kHz High Pass Filter Compressor Adder Band 2 80 Hz 3 kHz Band Pass Filter Compressor Band 1 80 Hz Low Pass Filter Identical Networks Compressor Network Compressor iin Compressor Network Recording Output or transmission Input Figure 5 5 Block diagram of A type noise reduction A type nr is in widespread use throughout the recording broadcast and film industries Itis an essential ingredient in the film release format known as Dolby Stereo 5 4 5 4 Dolby B type noise reduction B type nr is a consumer system intended primarily for use with low speed tape especially the Philips compact cassette It was first introduced in 1969 It reduces tape hiss by 10 dB Unlike A type nr it uses only one frequency band instead of providing variable gain within a fixed range of frequencies it provides fixed gain or cut i
110. ion arrangement of two low frequency compressor stages and three high frequency compressor stages each with a fixed band and a sliding band The outputs of the bands are combined in a unique way called action substitution which results in an unusually responsive treatment of the signal with respect to both frequency and level a technique referred to as modulation control augments the spectral tracking abilities of the system Spectral skewing contributes to a tolerance of channel errors and the employment of both high and low frequency antisaturation techniques results in a sig nificantly improved channel overload characteristic 7 ACKNOWLEDGMENT The author is indebted to Douglas Greenfield for the printed circuit layouts mechanical design and general assistance with the first practical embodiments of the new system Thanks are also due to many other indi viduals particularly to Henry Bottino for the design of testers to Brad Teague for preproduction testing and to Michael Smith Christopher James and Martin Fried for production engineering 8 REFERENCES 1 R M Dolby An Audio Noise Reduction Sys tem J Audio Eng Soc vol 15 pp 383 388 1967 Oct 2 R M Dolby A 20 dB audio Noise Reduction System for Consumer Applications J Audio Eng Soc vol 31 pp 98 113 1983 Mar 3 R M Dolby A Noise Reduction System for Consumer Applications presented at the 39th Con vention of the Audio
111. ispl From Recorder display To Console Play in Play out 3 1 2 Signal paths in PLAY Model 363 D Pon R h i VAS Processing module AAA To Recorder From Console Calibration displa From Recorde pay To Console Play out 3 1 3 Signal paths in RECORD and CHECK TAPE Model 363 D R n H U Processing module To Recorder From Console Calibration display From Recorde To Console Play out 3 1 4 Signal paths in PLAY and CHECK TAPE Model 363 RA be Processing module To Recorder From Console Calibration display From Recorde To Console Play out 16 u05 4444044444 Encode Decode operation 3 2 1 Spo AEjd ui I2UUE09 opoui 291 U BUUEUJ no Aeld Ino AEJd 9josuo2 OL Si m 49pJ029H u044 Aejdsip UORELGIIEJ UOREJGI EJ no 2ow EE 9I0SuO2 uio4j 49p41O29H 9p9o9u3 OL e npoul buisso9204d 9jnpotu Buisso204d 3 5 3 2 Encode Decode Operation Two processors are needed for each channel of the recorder One processor is permanently in the record signal path and one in the play signal path This arrangement is used where it is important to listen to the signal off tape or E to E during a recording or where encoding and decoding must be carried out simultaneously This method is used with film and most video recorders and transmission lines and may also have applications with
112. it if it is to be recorded Recording a signal A music program constantly changes in level and frequency content A moment in such a program might have the spec trum shown in Figure 2 The signal level varies with fre quency The gain setting that would give the best recording at middle frequencies would cause the high frequencies to overload One objective of Dolby spectral recording is to achieve as nearly optimum a level as is possible at all audio frequencies Finding the right level Figures 3 and 4 show what happens when the recording gain is adjusted the signal spectrum moves up or down in the window Even when the recording gain is set at the highest safe value as in Figure 3 much of the capacity of the record ing system that is the spectral space above the tops of signal components and inside the window is unused Dolby SR makes use of this capacity The simple compander The simple or broadband compander compressor and expander was first used in attempts to increase the dynamic range of recording systems During recording such a com pander increases gain when the overall signal level is low some companders also reduce gain when signal level is high During playback the compander s action is reversed high level signals regardless of their frequency cause all frequen cies to be played back at a high level Figure 5a while low signal level or no signal causes playback gain to drop Figure 5b The mea
113. ks reaching the monitoring circuits Each input amplifier has two FETs across the input line under normal operation these are held off by a 30 V supply generated from the Dolby tone oscillator circuit IC 12020 When the unit is not powered they act as a short circuit to prevent the non linear distortion which would arise if large signals are applied to the inpat operational amplifiers in their unpowered state In this mode the input impedance drops to about 10 kohm The output amplifiers are also balanced and provide the same output level even if one side is grounded ie in unbalanced use although of course the maximum output level will be halved Separate 10 turn potentiometers are provided for each output 6 3 Monitoring and Mode Switching The operation of the various switches is best understood from the block diagram where the switches are drawn in such a manner as to emphasis their function rather than their circuit The simpler block diagrams in Section 3 are also useful as they show the signal paths individually for each mode of the Model 363 The actual realization of these switches is a combination of logic circuits and FET switches and is shown on the detailed circuit diagram Examination of one switch will allow the remainder to be understood 6 2 The switching of the signal to the To rec output is a good example Under normal operation the To rec output is fed from the processing module Output 2 signal in the
114. l red LED or the Dolby noise from the tape recorder green LED The alignment procedure is correct when the levels between the tape and reference levels audibly match Note Dolby level on tape will now have a fixed flux level which may be compared with the magnetic reference tape used for the recorder alignment e g Dolby Level is 4 dB above 185 nWb m It is possible that the level of Dolby noise or Dolby tone recorded on the tape in this last step will produce a slightly different reading on the calibration display from that observed in step 10 above This is due to the tolerance of many parts of the complete audio chain the accuracy to which the recorder calibration adjustments can be made particularly if automatic setting of these controls is used and the agreement between meters or displays in the console recorder and the Model 363 If a small difference is observed as evidenced by the two green LEDs on the Model 363 calibration display not being equally bright we recommend that the rec out trimmers on the Model 363 be slightly adjusted to compensate for this build up of tolerances 4 8 TAPE LEVEL REFERENCE Aligning to Tapes Recorded in Other Studios A INITIAL STEPS 1 Remove the Model 363 from the signal path by pressing the Bypass button for each channel 2 Align the recorder using the tones on the incoming tape and check that recorder and console meters agree or bear a known fixed relationship If you
115. l voltage changes produced by the rela tively low impedance primary and secondary overshoot suppressors The use of the low frequency overshoot suppression method results in low waveform distortion of relatively slowly changing low frequency signal impulses applied to the system The control signal recovery characteristics are similar to those of the high frequency circuits although being about half as fast because of the longer time constants employed As in the high frequency stages the recovery time is favorably affected by the use of action substi tution and modulation control but is further augmented by the use of speed up diodes 4 OPERATING CHARACTERISTICS The practical results of the various methods and cir cuits which have been described are given in Figs 12 16 which show some of the main measurable op erating characteristics of the SR process 20 dB 10 0 dB 10 u 20 E 2 je sch a 30 5 o 2000 20000 20 200 FREQUENCY Hz Fig 12 Encoder characteristics single tone J Audio Eng Soc Vol 35 No 3 1987 March PAPERS 4 1 Dynamic Compression Action for Steady State Dominant Signals Referring to the single tone encoder curves shown in Fig 12 several features may be noted 4 1 1 Low Frequencies Dynamic action occurs in the range from about 48 dB to 5 dB with respect to reference level That is there is no action but full boosting
116. lby C type noise reduction C type nr was introduced in 1980 and is used in consumer audio cassette recorders and in the audio channels of professional Betacam MII and U matic SP video recorders It operates in a manner similar to B type but offers 20 dB of nr It achieves the steeper filter slopes required to give adequate nr at high frequencies in the presence lower frequency dominant signals by employing two overlapping processor stages in tandem operating with offset staggered thresholds and with an action extending two octaves lower than B type see Figures 5 8 and 5 9 Frequency shaping spectral skewing at the input of the encoder desensitizes the processor to the effects of high frequency response errors Additional shaping in the main path anti saturation lowers the amplitude of high frequency high level signals before they are applied to the tape reducing high frequency distortion and self erasure The decoder contains complementary circuits to restore the frequency response the amount of nr at the highest frequencies is decreased but this is where the ear is least sensitive to noise Virtually all C type circuits are built under license Control Control Variable high pass filter Control circuit Variable high pass filter Control circuit To recording medium Anti saturation Spectral skewing Figure 5 8 Block diagram of C type enco
117. lephone circuits NOISE REDUCTION SYSTEM DIFFERENTIAL METHOD A noise reduction system which is capable of meeting the listed requirements has been developed and is de scribed below In normal compression or limiting a primary object is to modify high level signal dynamics it is thus un fortunately necessary to subject the signal as a whole to the hazards of passage through a variable gain system In applying compression techniques to the noise reduc tion problem in which the objective does not include modification of signal dynamics it is unnecessary and undesirable to operate upon high level signal components noise amplitude in a high quality channel is only of the order of 0 1 of maximum signal amplitude It would clearly be preferable to generate a small correction or differential component which could be appropriately sub tracted from the signal thereby cancelling or reducing noise while leaving the larger aspects of the signal untouched The differential treatment of the signal in the present noise reduction system is illustrated in Fig 3 The net Recording or Record Transmission Playback Fig 3 Basic layout of noise reduction system In prac tice the operators G and G comprise identical sets of four filters and low level compressors works operators G and G are signal multipliers con trolled by the amplitudes frequencies and dynamic properties of the signals fed into them D
118. letter L or coloured red Safety Notices IMPORTANT SAFETY NOTICE This unit complies with the safety standard EN60065 The unit shall not be exposed to dripping or splashing and no objects filled with liquids such as coffee cups shall be placed on the equipment To ensure safe operation and to guard against potential shock hazard or risk of fire the following must be observed o Ensure that your mains supply is in the correct range for the input power requirement of the unit o Ensure fuses fitted are the correct rating and type as marked on the unit o The unit must be earthed by connecting to a correctly wired and earthed power outlet o The power cord supplied with this unit must be wired as follows Live Brown Neutral Blue Earth Green Y ellow IMPORTANT NOTE DE SECURITE Ce materiel est conforme la norme EN60065 Ne pas exposer cet appareil aux claboussures ou aux gouttes de liquide Ne pas poser d objets remplis de liquide tels que des tasses de caf sur l appareil Pour vous assurer d un fonctionnement sans danger et de pr venir tout choc lectrique ou tout risque d incendie veillez observer les recommandations suivantes o Leselecteur de tension doit tre plac sur la valeur correspondante votre alimentation r seau Cr o Les fusibles doivent correspondre la valeur indiqu e sur le materiel o Le materiel doit tre correctement reli la terre o Le cordon secteur livr avec le materiel doit tre cabl
119. levels from 10 dBr to 10 dBr can be adjusted to give Dolby level Built in Dolby tone and Dolby noise generators for calibration and channel identification Analog Audio Outputs XLR connectors balanced 20Q output level balance within 1 dB into symmetrical 600Q load output float better than 40 dB 50 Hz 1 kHz maximum signal level into 600Q 26 dBr balanced 21 dBr unbalanced output line levels in the range 10 dBr to 10 dBr can be adjusted to give Dolby level Remote Control 9 pin male D connector for individual channel record play changeover for remote control by the tape recorder connector may be wired to allow remote only or combined local and remote opera tion changeover time 3 ms maximum 15 pin female D connector for remote control of SR Off A in out and Setup functions also provides a remote status indication of A SR and Auto Compare function Front Panel Controls and Indicators Individual channel controls A OFF SR Selection of processing type NORMAL CHECK TAPE Allows checking of the nondecoded signal from tape during recording or playback RECORD PLAY Control of record play switching BYPASS Provides relay controlled hard bypass of all circuitry LEVEL CONTROLS Four multi turn level controls per channel for setting levels to and from the console and to and from the tape recorder SETUP For use during alignment and calibration Calibration Display Four LED display per channel for setting D
120. lters rectified and used to oppose the steady state fixed band control signals MC6 is also smoothed in a two stage 2 ms integrator forming MC7 which is used to control the low frequency fixed band overshoot sup pressors This smoothing is necessary in the low fre quency fixed band stages because unlike the situation in the high frequency fixed band stages there is no fixed phase relationship between the stage signals and the overshoot suppression signals MC7 is also used in a supplemental way to control the low frequency 107 DOLBY sliding band overshoot suppressors MCH is used to control the overshoot suppression circuits of both the fixed and the sliding band low frequency circuits MC8 compensates for the fact that no frequency weighting is used in the generation of the low frequency primary overshoot suppression sig nals High frequency transient signal components are detected and used to oppose the operation of the LF primary overshoot suppression circuits The signal from the MCS takeoff point is fed through a 5 kHz high pass filter rectified double differentiated with 15 j s time constants and peak held with about a 30 ms time constant The resultant high frequency transient mod ulation control signal MC8 is then employed to oppose the low frequency overshoot suppression action 3 3 High Frequency Stage Fig 9 shows both the steady state and the transient control aspects of the high frequency stages The
121. ly 1 AC Power Cable 1 Cable mounting 9 pin D connector Fuses one 20mm one 1 25 1 Manual 1 Set of identification labels 1 2 1 2 Regulatory Notices UL Troubleshooting must be performed by trained technicians Do not attempt to service this equipment unless you are qualified to do so The ground terminal of the power plug is connected directly to the chassis of the unit For continued protection against electric shock a three pin correctly wired and earthed power outlet must be used Do not use a ground lifting adapter and never cut the ground pin on the three prong plug WARNING On multi voltage units check that the unit has been set to the correct supply voltage and that the correct fuse is installed To reduce the risk of fire replace the fuse only with the same type and rating UK Connections For the United Kingdom WARNING THIS APPARATUS MUST BE EARTHED As the colours of the cores in the mains lead may not correspond with the coloured markings identifying the terminals in your plug proceed as follows e The core which is coloured green and yellow must be connected to the terminal in the plug which is marked with the letter E or by the earth symbol or coloured green or green and yellow e The core which is coloured blue must be connected to the terminal which is marked with the letter N or coloured black e The core which is coloured brown must be connected to the terminal which is marked with the
122. ly In particular compression expansion and noise reduction action are well maintained above the frequency of the dominant signal component in high frequency circuits and below the frequency of the dominant signal com ponent in low frequency circuits This action main tenance effect except on a one to one basis is absent in fixed band circuits Clearly it would be desirable to have the benefit of fixed band operation on the stop band side of the dom inant signal frequency and sliding band operation on the pass band side The action substitution technique provides this useful combination In Fig 4 c the re sponse of an action substitution compressor to the signal conditions of the two previous figures is shown As is seen the output is primarily from the fixed band for frequencies up to the dominant signal component and from the sliding band above that frequency Conversely for a low frequency stage the output is from the fixed band for frequencies down to a low frequency dominant component and from the sliding band below that fre quency This cooperative effect is particularly useful in the level region from the circuit threshold up to some 20 dB thereabove In the SR process action substitution operation is used in both the high and the low frequency circuits Thus both fixed band and sliding band dynamic actions are used in each of the five stages a total of ten com pressor circuits While there is an effective interaction of
123. m the Journal of the Society 8 3 Measurement of noise and noise reduction effect of Dolby SR A Measurement of Noise Human hearing is not uniformly sensitive with frequency At low levels those in which unwanted background noises exist it is most sensitive over a middle range of frequencies and progressively less so at frequencies further from the middle Beyond about 20 Hz and 20 kHz it is infinitely insensitive that is these frequencies cannot be heard irrespective of level by the large majority of people However the spectrum of noise delivered by most audio systems is roughly uniform with frequency white containing significant noise contributions at the extremes of the audio spectrum and usually well beyond those extremes Dolby SR has been designed to reduce noise in accordance with its audibility Operating on signals at the extremes of the audio spectrum or beyond where noise is not a problem is counter productive it increases the possibility of audible side effects and obviously has no audible benefit in reducing noise SR therefore reduces noise most over the middle of the spectrum has only a modest effect at the extremes and indeed slightly degrades noise at subsonic frequencies less than 20 Hz The design of SR in accordance with the audibility of noise means that for sensible results noise measurements must also take into account the audibility of noise A simple wide band unweighted noise measurement
124. m the point of view of attack time is a product of short recovery time as well as short attack time As with the attack aspects of the matter the recovery problem in the present system is solved jointly by the differential method itself and by suitable choice of char acteristics of the nonlinear control signal integration circuitry the smoothing time constant is made long under equilibrium conditions but is decreased appropri ately for large abrupt reductions in signal amplitude In this way low frequency distortion in the recording sending processor output is readily held to negligible values at high and low levels and to moderate values a fraction of a percent at intermediate levels while the recovery time is made sufficiently fast that perceptible noise modulation effects are avoided following cessation of the signal Because of the undistorted character of the recording sending processor output signal the system does not de pend upon subsequent distortion cancellation during re production for correct operation Phase errors in the audio channel thus are not troublesome the signal may be re recorded a number of times or be sent through transmission lines both being important applications in which nonlinear phase frequency characteristics prevail Also the signal may be processed and de processed re peatedly with negligible cumulative distortion effects It may be remarked that some of the operating char acteristics discussed
125. master tape recording When using two model 363s with a two track recorder it is possible to use one Model 363 to provide two channels of encoding and the other to provide two channels of decoding or alternatively to use one Model 363 to provide encoding and decoding for track 1 and the other for track 2 We would recommend the latter approach as shown on the next page which has some advantages in the Set up mode notably in the ease of entering this mode 3 3 Bypass Bypass is used during alignment or to remove the Model 363 from the audio paths When the power is removed the unit is bypassed automatically In bypass the Model 363 inputs are connected in parallel across the outputs of the console and recorder The inputs to the recorder and console are connected by relay contacts to these inputs with the noise reduction process disconnected from the signal path as shown below The impedance in shunt across the two inputs is about 20 kohm with power on and about 10 kohm with power off Model 363 GI A U 1 Processing module gt Rec out To Recorder From Console From Recorder To Console H Play out 4 1 SECTION 4 LEVEL STANDARDIZATION AND OPERATION Eel 4 1 Level standardization for Dolby systems Dolby A type and Dolby SR are complementary noise reduction systems that is the processing applied during playback is a mirror image of that applied when recordin
126. ment Dolby SR action at high levels B 2008 When a high level component appears in the signal spectrum Dolby SR assumes the kind of characteristic shown in Figure 13 In this example a single tone at 800 Hz and at a level of 0 dB has been applied to the system input Dolby SR reduces recording gain but only at and near the frequency of the tone and only by the amount needed to prevent overload Above and below this part of the spectrum the Dolby SR curve returns to the fixed low level characteristic This action has results that are especially impressive when listening to a Dolby SR record ing without decoding Although bright in sound the program does not appear to have been subjected to any dynamic process ing Since the only other system action consists of fixed equali zation at very low levels it is not surprising that during decoded playback no trace of processing can be heard except for remark able clarity of reproduction Masking and Dolby SR 14 800 Hz The changes that take place in the Dolby SR circuit are adap tive that is the system filter always adjusts itself to maintain the highest practical gain at every frequency as the signal spectrum changes The human ear and brain also respond to these changes in the signal spectrum one such response is a form of signal processing known as masking one of the most extensively studied aspects of hearing Masking is the concealment of a low level so
127. n the decoder within a band of variable width It can be considered as a high frequency emphasis of fixed magnitude whose start and stop frequencies slide upwards along the frequency axis so as not to boost the dominant high level spectral components of the input while providing a fixed 10 dB of nr at frequencies above those dominant components see Figures 5 6 and 5 7 At any one frequency the output input characteristic of the encoder displays gentle compression permitting complementary expansion in the decoder The fixed magnitude ensures that noise not masked by the input signal has a fixed level and therefore no noise modulation is perceived The vast majority of B type circuits are built under license from Dolby Laboratories Licensing Corporation by over 250 world wide licensees who include all the major manufacturers of consumer tape recorders Dolby Laboratories manufactures small numbers of professional B type processors for use in the preparation of pre recorded tapes audio cassettes and VHS video cassettes Control Control Variable high pass filter Variable high pass filter Control Circuit Control Circuit Output Input Low speed Recording t Noise Figure 5 6 Block diagram of B type noise reduction mech gain of encoder dB 20 200 2000 20000 Frequency Figure 5 7 Family of response curves for B type noise reduction 5 5 Do
128. ng Soc Vol 35 No 3 1987 March 105 DOLBY The modulation control aspects of the SR process result in an encoding action which is remarkably free of noticeable signal related modulation effects Working together with action substitution modulation control contributes to the goal of least treatment in providing a highly boosted audibly stable encoded signal 2 4 Overshoot Suppression A side effect of the modulation control scheme is that at high signal levels the amplitudes of the signals from the several stages are relatively high in comparison with the situation in the A type B type and C type systems Because of this it is not possible to employ simple overshoot suppression diodes as in these previous systems A more flexible but necessarily more complex scheme operating directly upon the control signals is used In common with the A type B type and C type sys tems the SR process features overshoot suppression thresholds that are significantly higher than the steady state thresholds this results in low modulation distor tion The overshoot suppression thresholds are set about 10 dB above the relevant steady state thresholds The net result is that for most musical signals the overshoot suppressors rarely operate the compressors are con trolled by well smoothed signals When the suppressors do operate the effect is so controlled that modulation distortion is minimal Under extreme transient conditions such
129. ntains brief descriptions of all More detailed descriptions of A type and SR can be found in the appendices in Section 8 5 2 General In sound recording or transmission the higher audio frequencies are often pre emphasized to improve the signal to noise ratio see Figure 5 1 However the equalization characteristic must be chosen so that even in the worst cases there are no detrimental effects material rich in high frequencies must not cause distortion Therefore the allowable boost with fixed equalization is limited and the degree of noise reduction is modest Pre De Input emphasis emphasis Output Recording or transmission Noise Figure 5 1 Fixed pre and de emphasis Systems which improve the signal to noise ratio by compression in the encoding mode followed by expansion in subsequent decoding are known generally as compandors Figure 5 2 shows the block diagram of a typical system Fixed e Recording Input gt Pre varane or Variable Ge Output emphasis transmission Gain emphasis Control Circuit Control Circuit Figure 5 2 Typical compandor The variable gain blocks change gain under the control of the signal level most commonly with a straight line relationship when the compressor output is plotted against its input using decibel scales such a system is known as a constant slope compandor The noise level at the output of the expander rises as the signal level rises
130. ntial Component Fig 4 Input output transfer characteristics of the noise reduction system The compression characteristic is formed by adding the differential component to the input signal the expansion characteristic is formed by subtracting the differ ential component from recorded transmitted signal in accord ance with negative feedback configuration shown in Fig 3 A related matter is the tracking behavior of the system with channels having nonlinear phase frequency response For a given rms value the peak and average values of a complex wave depend on the phase relationships of the various frequency components With a channel of uncertain phase response it is in principle necessary to RAY M DOLBY control the compression and expansion operations using the rms value of the signal a procedure which at best is inconvenient However in practice a combination of peak and average values is a sufficiently accurate indi cator of the rms value to permit the use of relatively simple rectification and smoothing circuits in the present system such circuits are used Good tracking is thereby obtained even when the signal has suffered considerable phase distortion A further channel compatibility aspect concerns the possibility of overloading channels with frequency dependent overload characteristics The overload prop erties may be further complicated if pre emphasis is used Since the pre emphasis is usually based on the ene
131. ntrol circuit suffices for the sliding band The same type of low level control characteristic modification is made in the low frequency circuits as in the high frequency circuits Namely a signal from the fixed band is combined in opposition with the sliding band output signal see combining circuit at right of Fig 10 This differential control modification raises the sliding band threshold at low frequencies 3 5 Low Frequency Stage Transient Control Aspects The transient control and steady state aspects of the low frequency stages are shown in Fig 11 Ina manner generally similar to that of the high frequency circuits unsmoothed rectified signals from the outputs of the variable elements are opposed by appropriate modulation control signals and are fed via diodes to the final integrator circuits Both the fixed and the sliding bands each have primary and secondary overshoot suppressors which operate at frequencies above about 100 Hz In addition each 112 PAPERS has a gentle and slow acting low frequency overshoot suppressor operating at frequencies below about 200 Hz there is a crossover effect between the two types of overshoot suppression in the 100 200 Hz region The primary overshoot suppressors provide the earliest and strongest suppression effect in simple transient sit uations With more complex signals the primary ov ershoot suppression thresholds rise and eventually the secondary overshoot suppression cir
132. nuated by the relatively slow loudness growth characteristics of the ear 13 In practice the clipper cir cuit is rarely called upon to perform its function the compressor operating linearly except with the most percussive types of program material Regarding modulation distortion it is evident that at high levels such effects are negligible because of the diminished influence of the differential component By the use of nonlinear control signal smoothing circuits distortion is minimized at low levels as well A relatively long attack time of the order of 0 1 second is used for small variations in signal amplitude the gain changes produced being slow enough that they do not gener ate audible modulation products The time constant is decreased in accordance with the size of the amplitude transition and for steps large enough to cause the com pressor output to exceed the clipper threshold the attack time is reduced to such an extent that the modulation clipping distortion produced is masked by the transient components present in the input signal Differential Network Filter Band 4 Linear Limiter 4 Non linear Limiter Non Linear Limiter 3 Linear Limiter 3 Non Linear Limiter 2 Linear Limiter 2 Non Linear Limiter 1 Filter Band 1 Linear Limiter 1 for both recording in the two modes The use of long attack time
133. o console output Repeat the test above making sure to switch to play and to check the calibration display has equally illuminated green LEDs with processing switched out before reducing the input by 24 dB am FREQUENCY ENCODE DECODE ENCODE DECODE 0 3dB 0 9dB 0 3dB 0 9dB ww m s M 39 ee ss os es is Table 1 Approximate encode decode level changes at 24 dB below Dolby level 7 3 Testers Dolby Laboratories manufactures a range of simple go no go test instruments for these modules These testers called Cat No 379 and Cat No 381 check two modules in a back to back connection and perform simple tests to give a good indication if the module is operating correctly While the test is not sophisticated enough to detect minor malfunctions it is very useful to confirm that no major problems exist 7 4 Motherboard Fault Diagnosis If the fault is isolated in the motherboard then either the unit can be returned to Dolby Laboratories offices or to the local Dolby agent Alternatively the unit may be repaired In house Circuit diagrams are provided in Section 6 and should allow simple fault finding and repair CAUTION The service instructions are for use qualified personnel only To avoid electric shock do not perform any servicing unless you are qualified to do so First remove the unit from the rack and remove the top cover four screws at the sides see Figure 7 1 at the end of the section for
134. o four frequency bands Fig 5 Dependence is placed on the masking effect for subjective noise reduction in portions of the spectrum occupied by signals having amplitudes appre ciably higher than the compression thresholds Beginning with the early studies of Wegel and Lane 5 investigations of the masking effect have been concerned almost exclusively with the masking of pure tones by tones or noise 19 The considerable body of results avail able is unfortunately not very relevant in the noise re duction application in which the masking of a band of noise by one or more tones is of interest A closer ap proach to the conditions required is the masking of one band of noise by another band of noise 17 But systematic research into the use of wideband noise as the maskee has only recently been undertaken and it would appear that it will be some time before sufficient work has been done to permit the choice of noise reduction system design parameters simply by reference to published psychoacoustic data In applying the masking phenomenon to the noise reduction system design problem the number of bands used circuit complexity must be balanced against other parameters and the overall system performance requirements It was found that for normal high quality audio channels the use of four bands yields satis factory noise reduction properties while permitting the compression thresholds to be set at a value low enough to obtain the advantages
135. o tell the difference between Tape noise and Ref noise 4 sec Tape D TT 4 sec Reference Note that due to the nature of pink noise the level indicated on any meters in the studio will depend critically on the type of meter used Do not try to read the level of Dolby noise with your studio meters The Auto Compare sequence will start whenever Ge Dolby noise is being replayed from tape AND SR processing is selected AND the Set up button is pressed LEDs on the front of each module indicate the source of the noise being heard Remote indicating LEDs can also be connected see section 2 The red LED indicates the internal reference noise and the green LED indicates Dolby noise from the tape recorder The Auto Compare circuitry may deliver spurious and unexpected noises if it receives signals other than Dolby noise including silence this is normal and harmless To avoid such anomalies it is good operating practice when playing Dolby noise to find the appropriate point on the tape before you press the Set up button and to release the button as soon as you have finished with the Auto Compare sequence 4 3 1 When the SR off A switch is in the SR position the signal paths in Set up are Model 363 Dolby noise generator Processing module To Recorder Calibration From Recorde display To Console Play out Pink noise generator The calibration display will read
136. olby level Dimensions and Weight 1 U Rackmount 44 x 483 x 285 mm 1 75 x 19 x 10 2 inches a further 65 mm 2 5 inches required for standard XLR connectors Net 6 3 kg 14 lbs including two Cat No 300 modules Power Requirements 230 V version 198 264 VAC 50 60 Hz 40 W uses one 20 mm T250 mA fuse Multi voltage version User selected 85 132 VAC 50 60 Hz 40 W uses one 1 25 inch 500 mA slow blow fuse or 187 264 VAC 50 60 Hz 40 W uses one 20 mm T250 mA fuse Operating Conditions Up to 40 C 104 F Regulatory Notices US This unit is UL listed Europe The 230 Volt unit complies with the requirements of Low Voltage Directive 73 23 EEC and EMC Directive 89 336 EEC Warranty One year limited parts and labor DISCLAIMER OF WARRANTIES Equipment manufactured by Dolby Laboratories is warranted against defects in materials and workmanship for a period of one year from the date of purchase There are no other express or implied warranties and no warranty of merchantability or fitness for a partic ular purpose LIMITATION OF LIABILITY It is understood and agreed that Dolby Laboratories liability whether in con tract in tort under any warranty in neg ligence or otherwise shall not exceed the cost of repair or replacement of the defective components and under no cir cumstances shall Dolby Laboratories be liable for incidental special direct indi rect or consequential damages includ ing but n
137. on Dadson curves the encode characteristic resembles the subsequently derived CCIR noise weighting curve Thus the SR system is designed to reduce only those noises that can be heard The prevention of action in inaudible signal regions promotes accuracy in the au dible region 20 dB 10 20 200 2000 FREQUENCY Hz Fig 14 Low level encoding characteristic in the presence of a 200 Hz signal at the levels indicated 115 DOLBY 4 3 Treatment of Nondominant Signal Components The behavior of the system with low level nondom inant signal components in the presence of higher level dominant signal components can be simulated by the use of probe tones Such a representation is significant because it is an indicator of the noise reduction effect achieved with signals Refer to the curves shown in Figs 14 16 which were obtained by adding a swept frequency probe tone at levels between 60 dB and 80 dB into the encoder input signal and detecting the tone at the output with a tracking wave analyzer Toward the two spectrum ends nondominant signal components are boosted more than the dominant signal by high and low frequency sliding band actions If there are two dominant signals a fixed band compres sion effect prevails for the nondominant signal com ponents between the frequencies of the dominant signal components Thus nondominant signal components are boosted by an amount at least equal to th
138. on control sliding at and above the 0 dB level is greatly reduced The effect is progressively reduced for low signal levels but is observable to some extent at the 10 dB level The use of modulation control techniques also has advantages under transient conditions both from their use in the steady state circuits and also because of their use in the overshoot suppression circuits Modulation control generally prevents any further fixed band at tenuation or sliding of the variable filter than is required to respond to a given signal situation Therefore 1 signal modulation is reduced 2 the SR process is ren dered very tolerant of channel errors 3 subsequent noise modulation during decoding is reduced and 4 recovery from transient signal conditions is faster The electronic reality in both the steady state circuits and transient control circuits is that the integrator capacitors are prevented from charging to voltages as high as they normally would in the absence of modulation control With lower fully charged voltages recovery is faster 10 dB 20 200 2000 FREQUENCY Hz b Same as a with modulation control 10dB No signal 0 20000 20 200 FREQUENCY Hz d Same as c with modulation control Fig 7 Effect of modulation control on sliding band compressor circuit with a signal in the stop band a b and a signal in the pass band c d J Audio E
139. on of high and low frequencies in this path a signifi cant increase in headroom is provided further maintaining signal purity with negligible effect on low level operation Comparing Dolby spectral recording to other methods We can compare the static performance of different recording systems by superimposing their windows and the auditory window Figure 7 Any limitation that might be audible will appear as an obstruction that reduces the size of the opening in the auditory window It is important to remember that this method of comparison does not show audible dynamic effects such as modulation noise of analog systems nor does it show low level non linearities non monotonicity or effects of d c asymmetry all of which may occur to varying degrees in digital recording systems These effects all close the correspond ing system windows from the bottom when a signal is present In the figures that follow for unassisted analog tape and Dolby SR zero level is 320 nWb m for digital recording zero level is 10 dB below the absolute clipping level In all three examples the relative vertical positions of the windows have been chosen so that the maximum recorded level will be presented at an acoustic level of 110 dB during playback Analog recording no signal processing Figures 17 18 Unassisted analog tape recording shows characteristic limita tions in available headroom at low and high frequencies and a substantial level of noise a
140. ormal practice to connect program audio ground to power line ground for many reasons including safety On occasion particularly with remote location recording induced hum can sometimes be reduced by separating the two grounds Link J1101 provides this feature This link is accessible through the left hand side of the left hand front panel cut out after first removing the processing module Long nose pliers provide the easiest method of moving the jumper from its normal closed position towards the front of the unit into the open positions towards the rear see Figure 2 4 Note that there is always a 1kohm resistor across the link which remains in the open position so that the audio ground is never totally isolated from the chassis ground Also note that the chassis is always connected to the ground pin of the power line cord for safety reasons this ground should NEVER be disconnected e OA Dolby modei 253 E x NYE Ch Y Brpass Ch 2 n JL ee Peso CN A HOLDER C d LINK FOR CHASSIS GND SIGNAL GND Figure 2 4 STEP 3 If remote control of the record playback functions is required wire up a standard 9 pin female cable mounting D connector connected to J9 rec play remote as follows Figure 2 5 Each channel can be switched from the playback mode to record mode by applying a voltage in the range 4 25 V dc to pins 3 and 4 for channel 1 and pins 7 and 8 for channel 2 with
141. ot limited to damage to soft ware or recorded audio or visual mater ial or loss of use revenue or profit even if Dolby Laboratories or its agents have been advised orally or in writing of the possibility of such damages All specifications apply with input and output controls set for Dolby level equal to 4 dBr 1 23 Vpyyg balanced source and load 0 dBr is defined as 0 775 V without regard to impedance Two units back to back encode decode Weighting filter supplemented by 25 kHz 4 pole lowpass filter to ensure that only audible noise is measured Average responding or RMS meter 4 pole filters Output float is the level across a balanced load relative to an interfering signal injected at one end of the load Specifications subject to change without notice Dolby and the double D symbol are trademarks of Dolby Laboratories 2000 Dolby Laboratories Inc 500 12485 13313 2 1 SECTION 2 INSTALLATION a Rpg DC O a 2 1 Installation STEP 1 Unpack the unit from its box and check for any damage Be sure to check the packing material for the processing module s mains cord manual and other accessories The processing modules are separately packed in individual boxes STEP 2 Mount the unit appropriately in a rack or in a tape recorder Ensure that there is air flow around the unit and that it is not mounted directly above any other heat producing equipment The unit will operate within specifications
142. out The calibration display will read the level returning from the tape recorder and if the Model 363 is switched to record Dolby tone will be the output to the recorder This mode is also used to align the Model 363 to Dolby tone played back from a previously recorded tape 4 3 4 3 Level Standardization for Dolby Spectral Recording Dolby SR also has its own distinctive calibration signal called Dolby noise Dolby noise like Dolby tone not only serves as a reference level but also indicates that a recording was encoded with Dolby SR As with Dolby tone Dolby noise has been made distinctive It consists of pink noise interrupted every two seconds by 20 ms gaps as shown diagrammatically below note actually the nick is much shorter than shown and also noise is a much less regular signal than in the diagram 2 sec noise 20 ms silence Dolby noise is used in a similar way to Dolby tone for calibration and alignment Dolby noise recorded on tape should produce a reading of Dolby level on the calibration display when the Set up button is pressed Unlike Dolby tone Dolby noise is not recorded on the tape at Dolby level but at 15 dB below Dolby level Dolby level is usually around 0 VU recording Dolby noise at this level would risk saturation at both high and low frequencies especially at low tape speeds or when using thin oxide tapes Thus the noise recorded on the tape would not be a true refl
143. ow level compo nents are subtracted from the incoming signal and noise acquired in the audio channel is thereby subtracted or reduced as well The noise reduction system described is capable of performance of a high order with regard not only to noise reduction but also to signal quality compatibility of the processed signal with the audio channel and suitability of characteristics under non ideal channel conditions REFERENCES 1 H H Scott Dynamic Noise Suppressor Elec tronics 20 96 1947 2 H F Olson Audio Noise Reduction Circuits Electronics 20 118 1947 3 N J Hudak Transistorized Audio AGC Ampli fier IEEE Trans on Broadcasting BC 9 No 1 26 1963 4 A Kaiser and B B Bauer A New Automatic Level Control for Monophonic and Stereophonic Broad casting IRE Trans on Audio AU 10 No 6 171 1962 5 J G McKnight Signal To Noise Problems and A New Equalization for Magnetic Recording of Music J Audio Eng Soc 7 5 1959 6 A C Norwine Devices for Controlling Amplitude Characteristics of Telephonic Signals Bell System Tech J 17 No 4 539 1938 7 R O Carter Theory of Syllabic Compandors IEE Procs 111 No 3 503 1964 8 Several papers on compandor techniques appear in the IEE Conference Proceedings Transmission Aspects of Communication Networks Institution of Electrical Engineers London 1964 9 D Aldous NoisEx Recording
144. panel rec play switch Any remote rec play switching is disabled This mode is used to align the Model 363 to standard studio line levels 4 5 Alignment During initial or routine alignment it is simplest to send a tone at Dolby level from the console and align for unity gain through the system When aligning to a tape from another studio it is necessary to take the Dolby noise or Dolby tone recorded on the tape as the starting point and set up the unity gain structure from there For this reason there are two alignment procedures set out here the first sets Dolby level to a tone from the console Console Level Reference the second to Dolby noise or tone recorded on a tape from another studio Tape Level Reference These instructions assume that the Model 363 record play switching is being controlled manually at the front panel Should the record play switching be controlled automatically by the tape recorder the manual switches must be used when both Set up and off have been selected since under these conditions the remote record play switching is purposely disabled The various controls are shown in Figure 4 1 on page 4 11 The two procedures start on the following page 4 6 CONSOLE LEVEL REFERENCE Routine in house alignment A INITIAL STEPS The following alignment steps involve sending 1 kHz tone from the console at Dolby level In many studios this will be the normal studio line level for example O VU
145. quired to achieve the de sign goal of a full spectrally responsive system In particular both fixed and sliding bands are used in a unique combination called action substitution that draws on the best features of both types of circuits A further technique called modulation control greatly improves the performance of both the fixed and the sliding bands in resisting any modulation of signal components unless necessary The use of the new methods reduces the basic encoder to two frequency bands only high frequency and low frequency each with a fixed band circuit and a sliding band circuit this combination being referred to as a stage When the three level action staggering layout is taken into account five fixed bands and five sliding bands are employed in the spectral recording process 2 2 Action Substitution A new type of compression and expansion action that is highly responsive to spectral changes can be achieved by superposing or overlaying the individual characteristics of different types of dynamic action cir cuits One circuit may provide a quiescent characteristic or defining umbrella A further characteristic is hidden until signal components appear that cause the hidden characteristic to be revealed and become active For discussion purposes let the gains in a compressor system be arranged such that subthreshold signals pass without attenuation That is the maximum possible action is that of providing a certain gain
146. r in the play mode Test signals at various levels and frequencies and program material can be sent through the two channels in series and a comparison of input to final output will very soon show up any problems In the field it is sometimes difficult to measure levels precisely due to the lack of accurate instruments This procedure therefore has been chosen to allow a quick operational rule of thumb check to be made Accurate tests on the modules must be made using specialized test instruments and are of necessity too time consuming for an operational procedure 7 2 Measurements given and their expected ranges assume that the input signal level is only accurate to within 0 5 dB Thus the range of values given does not represent the tolerance of the Dolby system with a defined input the accuracy is much better Tests are made by first switching out the signal processing and selecting the record mode Next send a tone into the From console input at Dolby level so that the two green LEDs on the calibration display are equally illuminated Measure the output at the To recorder output which in most installations should be at the same level as the input Reduce the level from the console by 24 dB note the reading at the output and select either A type or SR The measured signal should increase as listed in the table below To check the play mode transfer the input test signal to the From recorder input and the meter to the T
147. r under certain conditions The primary overshoot suppression signal is derived from the same rectifier used in the steady state control circuit with the consequence not only of economy but of a favorable phase relationship between the overshoot suppression impulse and the signal to be controlled this results in low transient distortion However the control circuit frequency weighting re sponsible for this situation also causes a reduction of J Audio Eng Soc Vol 35 No 3 1987 March PAPERS control signal amplitude with falling frequency A de bias is used in the overshoot suppression circuit to create the required suppression threshold when the signal amplitude in the overshoot suppression circuit decreases the bias results in the overshoot suppression effect falling away faster than the signal amplitude For frequencies below about 400 Hz a reduced overshoot suppression effect is appropriate because of the atten uation and phase shift of the stage input filter see Fig 5 However in the 400 800 Hz region there is an overshoot suppression deficiency this is compensated by feeding an appropriate amount of overshoot sup pression signal from the fixed band circuit into the sliding band circuit This supplemental signal is called the secondary overshoot suppression signal Regarding recovery times the use of action substi tution and modulation control both contribute to rapid action as already mentioned Nonetheless reve
148. rgy probability distribution with frequency for nor mally encountered sounds it is evident that any practical noise reduction system should not interfere unduly with this distribution The compression of comparatively high level signal components thus must be avoided the transfer characteristic of the present system satisfies this condition see Fig 4a Dynamic Properties Overshoots arising because of control circuit time lag normally have maximum amplitudes equal in value to the degree of compression Such overshoots waste some of the dynamic range of the audio channel if they are passed linearly Furthermore if they are clipped by the channel various undesirable side effects can be created for example blocking of amplifiers break through from groove to groove with discs and inter ference with other channels if modulated carriers are used Controlled clipping of the output signal in the compressor itself is a method of avoiding these difficul ties but it has the disadvantage of reducing the overload margin The usual solution is to make the attack time as short as possible and either to clip within the device or to depend upon the shortness of the overshoot to minimize side effects with clipping in the channel Unfortunately the use of short attack times results in side effects in the signal Rapid changes in gain cause significant modulation Differential Network Record Processor Output input
149. rse biased recovery speed up diodes are used in a fairly gentle way series resistors to provide a further increase in speed 3 4 Low Frequency Stage Steady State Aspects Fig 10 shows only the steady state layout of the low frequency stages As with the high frequency stages only the basic parameter determining elements are shown The high level and mid level low frequency stages have the same block diagrams and circuits but the ac and dc gains are increased for the mid level stage there are also some other minor differences Sliding Band 0 Fixed Band Maximum Selector SPECTRAL RECORDING PROCESS Referring to the block diagram certain similarities and differences may be noted with respect to the high frequency diagrams The dual layer arrangement of the fixed band on the bottom and the sliding band on the top is similar However the sliding band acts down ward using a simulated inductance gyrator circuit As with the high frequency stages the fixed and slid ing band circuits are fed in parallel and the output signal is taken from the sliding band circuit The fixed band output is coupled to the bottom of the sliding band to provide the action substitution operation de scribed previously A notable difference from the high frequency circuit is that the fixed 800 Hz band determining filter follows rather than precedes the variable filter This arrange ment has several advantages 1 overshoot suppression
150. rshoot suppressor LF O S is used for very low frequency signals This is a very gentle slow acting circuit which reduces low frequency transient distortion 106 PAPERS 2 5 Staggered Action Multilevel Format The principles discussed above are incorporated into each stage of the multilevel staggered action encoder and decoder See 2 for a detailed discussion of stag gered action circuits In the SR system two stages are employed at low frequencies three at high frequencies The thresholds used are approximately 30 dB 48 dB and 62 dB below reference level about 20 dB below SR peak signal level In the series connected staggered action format there is a compounding of the actions of the individual stages the transfer functions of the several stages are multiplied whereby the dB characteristics add In this way a large total dynamic action can be achieved with low modulation distortion low overshoot and good manufacturing reproducibility An important additional result is that there is an overall steepness enhancement of the frequency discrimination abilities of the circuit further inhibiting signal mod ulation and noise modulation effects 2 6 Spectral Skewing The spectral skewing networks employed in the SR process comprise both high frequency and low fre quency sections with the same rationale and mode of operation as discussed in 2 The spectral distributions of the signals processed by the encoder are al
151. s and with tape recording improves the long term stability of the re cording The high frequency improvements are espe cially significant with 35 us CCIR recordings The antisaturation effect at low frequencies usefully coun teracts tape overload particularly with 3180 us NAB recordings 4 2 Quiescent Subthreshold Signal Characteristic The very low level or subthreshold characteristics of the SR process are shown in Fig 13 The general shape of this characteristic was determined in a way that takes good advantage of the properties of human hearing First there is less of a noise generation and perception problem at moderately low frequencies such as 200 Hz than at moderately high frequencies such as 3 kHz Therefore two low frequency stages are employed but three high frequency stages are used Second at very low and very high frequencies even less noise reduction is needed below 50 Hz and above 10 kHz Strong spectral skewing actions can therefore be used in these regions resulting in accurate decoding even when the recording medium has response irreg ularities In addition the spectral skewing networks provide for good immunity to high and low frequency interference supersonic audio components tape re corder bias subsonic noise components arising from wind traffic or other rumble sources Note that the overall shape of the low level SR decode characteristic resembles the low level Fletcher Munson and Robins
152. s as described not only reduces modulation distortion but tends to improve the noise reduction of the system Since the amount of noise reduction depends upon the amplitude of the differeritial component in relation to that of the signal in the main path it is an advantage if short transients of moderate amplitude are prevented from causing unnecessary com pression of the differential component Overshoots of several dB may be produced under these conditions but they are of such low amplitude compared with peak level that they are handled linearly in all respects While the attack behavior is undoubtedly the most important dynamic aspect of the system particularly in relation to ensuring compatibility of the recording send ing processor output with the audio channel the decay or recovery time is of equal significance when the noise reduction properties are considered The problem in this regard is to reduce the recovery time to such a value that noise reduction following cessation of high ampli tude signals is provided adequately by the residual mask ing phenomenon by which the sensitivity of the ear is momentarily reduced The noise reduction action of the system must thus be restored in an interval of the order of 0 1 sec during which residual masking prevails The use of short recovery times in normal compressors results in high distortion at low frequencies Further more modulation distortion which was discussed pre viously fro
153. s every 750 ms This easily recognized signal indicates that the tape has been recorded with Dolby A type noise reduction and also represents the tape fluxivity corresponding to Dolby level used on the recording For correct operation Dolby tone recorded on tape should produce a reading of Dolby level on the calibration display when the Set up button is pressed Many different parts of the audio industry have standardized their Dolby levels to ease interchange of material The list below gives some examples Typical Dolby levels Application Dolby level Meter reading Level Recording Studio Europe 0 VU 320 nWb m Recording Studio USA 0 VU 250 nWb m Most video formats 100 nWb m 35mm Magnetic film 185 nWb m 35mm Optical film 50 If no standard is listed for your application please contact Dolby Laboratories for discussions on an appropriate Dolby level The choice of Dolby level is influenced by system noise system headroom type of metering systems and type of transmission or recording method If you have to make a quick decision check with other possible local users or choose a level which is easy to read on your studio meters 4 2 1 When the SR off A switch is in the A position the signal paths in Set up are _Model 363 Dolby tone generator Processing module Only in record mode To Recorder From Console Calibration display From Recorde To Console P ay
154. sary attenuation has two effects 1 sub stantial noise reduction action is lost during expansion and 2 when the amplitude of the 100 Hz signal varies it can modulate low level signal components at higher frequencies resulting in possible incorrect restoration of the signal by the expander if the recording channel has an irregular frequency response in the vicinity of 100 Hz Fig 6 b shows the operation of the same circuit with modulation control A greatly reduced attenuation occurs when the 100 Hz signal is varied over the same 10 dB No signal o 20 200 2000 20000 FREQUENCY Hz a Frequency response curves with 100 Hz signal at the levels indicated no modulation control 10 dB 5 dB and below 0 20 200 2000 20000 FREQUENCY Hz b Same as 6 a with modulation control Fig 6 Effect of modulation control on fixed band compressor circuit J Audio Eng Soc Vol 35 No 3 1987 March PAPERS range of levels as in Fig 6 a Thus a significant im munity to strong signals in the stop band frequency region is achieved the effect decreasing as the dominant signal frequency approaches the pass band frequency region of the circuit In Fig 7 a the operation of a sliding band circuit under comparable conditions is shown As with the fixed band circuit ideally there should be no sliding in response to a strong 100 Hz signal Nonetheless as the 100 Hz signal
155. se checking these factors is a time consuming and boring process A faster and more interesting method of accomplishing these checks would be more likely to produce reliable and consistent results For this reason practical em bodiments of the SR process include pink noise gen erators which are used for both level and frequency response calibration instead of single tone sine wave oscillators For identification the pink noise is inter rupted with 20 ms nicks every 2 s During recording this signal is fed to the tape at a level of 15 dB below reference level a level low enough not to cause satu ration problems with low speed tape recording or highly equalized transmission channels During playback the tape signal is automatically al ternated with internally generated reference pink noise uninterrupted in 4 s segments 8 s total cycle time and passed tothe monitoroutput Anaudiblecomparison can thus be made between the reference pink noise and the calibration noise coming from the tape Any dis crepancies in level and or spectral balance are imme diately noticeable and can be corrected or at least taken note of If desired the signal can also be fed to a spectrum analyzer In using the new calibration method it is important to be able to tell when the 4 s tape segments are being passed to the monitor and when the signal heard is from the reference pink noise generator Differentiation of the tape segments from the reference se
156. shown in Fig 9 In the high frequency circuits a general feature is that unsmoothed rectified signals from the control circuit rectifiers are opposed by appropriate modulation control signals and are fed via diodes to the final integrator circuits The low frequency ar rangements follow the same pattern with some mod ifications Referring to the middle of the diagram in the high frequency fixed band circuit the overshoot suppression signal is derived from the rectifier of the main control circuit As with the steady state control signal the rectified signal is opposed by MC3 so that the overshoot suppression threshold is appropriate for conditions in the steady state regime The resultant overshoot sup pression signal is coupled by a diade to the final in tegrator circuit In the sliding band circuit top of Fig 9 two ov ershoot suppression signals are used primary and sec ondary The primary overshoot suppression signal is derived from the control circuit rectifier and opposed by MC2 a smoothed version of MC1 MCI controls the steady state characteristics The smoothing is nec essary because unlike the situation in the fixed band circuit there is no constant and favorable phase rela tionship between the signal in the contro circuit and MCI because of the sliding band the smoothing en ables reliable bucking action to take place The secondary overshoot suppressor supplements the action of the primary overshoot suppresso
157. signal is bucked by modulation control signal MC6 smoothed by a 15 ms integrator and fed to one input of the max imum selector circuit The maximum selector circuit has the same purpose and mode of operation as in the high frequency circuits The 800 Hz and 1 6 kHz frequency weighted output of the fixed band circuit is also fed to the pass band control circuit bottom of Fig 10 Here the control signal is further weighted by a 400 Hz single pole low pass filter rectified smoothed by a 15 ms integrator and fed to the other imput of the maximum selector As in the high frequency stages the larger of the two signals is passed to the final integrator 300 ms to become the fixed band control signal applied to RV In this way both simple and complex signals are ac commodated The sliding band control signal as in the high fre quency circuits is derived from the stage output that is from a point following both the fixed 800 Hz band determining filter and the variable filter The signal is frequency weighted by an 80 Hz single pole low pass filter rectified and bucked by modulation control signal MC4 which also has a single pole low pass characteristic with the same type of sliding band end stop effect as in the high frequency circuits The result is smoothed by a 7 5 ms integrator and finally smoothed by a 150 ms integrator to become the sliding band control signal applied to RV As in the high frequency stages a single co
158. sured dynamic range of a recording system may seem to be increased greatly However a simple compander meets its dynamic range specifications only when no signal is recorded When a real program is recorded the compander is often at rest when it is needed most When it works on the other hand critical listeners can hear artifacts like pumping modulation of the background noise or signal as well as limited transient performance Dolby A type noise reduction No professional signal processor has ever been as widely used as Dolby A type noise reduction This is still true now 25 years after its introduction Dolby A type processing substantially improves the efficiency of magnetic and optical recording media and audio communication systems Dolby A type signal processing relies on compression and expan sion but only at low signal levels and separately in four frequency bands Figure 6a The signal components in each band Figure 6b are integrated if this level is below a fixed threshold it is boosted during recording locations marked A in Figure 6c and attenuated during playback Figure d The boost used in Dolby A type noise reduction is 10 dB across most of the audio band increasing to 15 dB at very high frequencies To improve recording further it is not enough simply to increase these figures the boost must conform more closely to the signal spectrum than is possible in a four band system The au
159. t No 300 modules which contain both Dolby SR and Dolby A type processing A Dolby SR only version using Cat No 350 modules is also available Note Cat No 22 and Cat No 280 modules cannot be used in the Model 363 For multichannel applications the Dolby SRP Series is available to provide up to 24 channels of Dolby SR noise reduction 100 Potrero Avenue San Francisco CA 94103 4813 Telephone 415 558 0200 Fax 415 863 1373 info dolby com Wootton Bassett Wiltshire SN4 8QJ England Telephone 44 1793 842100 Fax 44 1793 842101 info dolby co uk WWW d olby com DOLBY NOISE REDUCTION Dolby noise reduction is a family of signal processes that reduce the noise inherent in analog recording media without affecting the sound being recorded While they differ in perfor mance and details of operation all Dolby NR systems are complementary processes that first encode the music when it is recorded then decode it when it is played back They also treat soft signals separately from loud ones and vary the NR action with frequency to avoid the side effects typical of other systems Dolby A type noise reduction intro duced in 1965 was originally intended for use by professional recording studios to make quieter master tape recordings In the early 1970s its use was extended to film studios and motion picture release prints to make films sound better Dolby SR spectral recording intro duced in 1986
160. t mid frequencies Digital recording Figures 19 20 A typical digital recording provides performance that is better than unassisted analog tape in several obvious ways The main drawbacks are the hard clipping barrier of digital recording and the disadvantageous spectral distribution of noise Although the measured dynamic range of a digital recording system may exceed 90 dB the noise level is not uniform with frequency The noise level is extremely low at very low frequencies much lower than the noise of analog tape which is already more than adequate at low frequencies However digital system noise crosses the auditory threshold precisely in the spectral region where the ear is most sensitive The usable improvement in noise level especially in the presence of a signal is not as great as theory predicts Because the noise generated by a digital system is not random and is therefore especially noticeable to the ear it is normally masked by the addition of dither noise elevating the final noise level Dolby SR Figures 21 22 This data was obtained using standard tape and a widely used professional recorder operating at 15 ips The noise at the very bottom of the window could not be heard in a recording studio or control room unless the playback gain were increased considerably under those conditions maximum peak levels would approach or surpass the auditory threshold of pain Played back at very high levels for test purposes t
161. tered or skewed well within the pass band such that the encoder action is significantly less susceptible to the influence of signals beyond the abrupt roll off frequencies of the spectral skewing networks The high frequency network is a low pass filter with an attenuation characteristic similar to that of a 12 kHz two pole Butterworth filter but with a limiting attenuation of about 35 dB that is a shelf The low frequency network is a 40 Hz high pass filter connected in series with the high frequency network also witha two pole Butterworth like characteristic but with about a 25 dB limiting attenuation These shelves do not in terfere with the attenuation within the audio band but provide phase characteristics that are essential in the decoding mode 2 7 Antisaturation The general principle of antisaturation was described in 2 Briefly by placing a fixed attenuation network usually a shelf in the main path of a dual path com pressor it is possible to create an effective antisaturation characteristic at the extremes of the audio band without undue adverse effects on the noise reduction achieved during decoding In the SR process high and low frequency networks are operative above about 5 kHz and below about 100 Hz respectively In addition the spectral skewing networks have a secondary but very useful antisaturation effect especially at very low and very high frequencies 3 BLOCK DIAGRAMS 3 1 Basic Block Diagram
162. teristic Each stage above has a low level gain of somewhat over 8 dB whereby a total dynamic effect of about 16 dB is obtained at low frequencies 24 dB at high fre quencies A further dynamic action of about 1 dB takes place above the reference level The spectral skewing network has the same purpose and function as in the C type system except that a spectral skewing action is provided at low frequencies as well The spectral skewing networks desensitize the SR process to the influence of signal components at the extreme ends of the audio frequency band This effect is particularly helpful if the recording or trans mission system has an uncertain frequency response in these regions The networks are also important in attenuating subsonic and supersonic interferences of all kinds The spectral skewing action is compensated in the decoder resulting in an overall flat frequency response Both high frequency and low frequency antisaturation networks are provided in the main signal path again operating in substantially the same way as in the C type system There is an effective compounding of the antisaturation effects produced by the antisaturation networks and the spectral skewing networks In this way the SR process achieves a significant increase in high and low frequency headroom 2 GENERAL PRINCIPLES 2 1 Least Treatment Principle A design philosophy used in the development of the new system is that the best treatment of the sign
163. the return signal from the tape recorder 9 Replay the 1 kHz alignment tone from the incoming tape and adjust the play out trimmers on the Model 363 for the correct reading e g 0 VU on the console meters As a quick check you may switch the Model 363 in and out of Bypass at this point and make sure the console meters don t change 10 Reset the SR off A switches to the proper NR type for the incoming tape For playback only purposes the alignment is now complete Should you be required to drop mto record overdub continue with the following steps C RECORD ALIGNMENT 11 Set the rec play toggle switches on each channel to the rec position red LEDs on 12 Set the SR off A to SR or A for the NR type being used 13 Place the recorder in record line in or E to E on an appropriate section of blank tape Press the Set up button Dolby noise SR or Dolby tone A type is now being sent to the tape recorder 14 15 16 17 18 4 9 Adjust the rec out trimmer on each channel on the Model 363 for equal brightness of the green LEDs Set the SR off A switches to the off position Send a 1 kHz test tone at a convenient level e g 0 VU from the console to the Model 363 Adjust the rec in trimmers on the Model 363 for the correct reading e g 0 VU on the tape recorder meters As a quick check you may switch the Model 363 in and out of Bypass at this point and make sure the tape recorder meters don t change
164. thresholds MC2 must be smoothed because the phase relationships of MC1 and the signals in the stages vary because of the sliding band action throughout the audio band being a function of frequency and level MC3 controls the high frequency fixed band circuits The signal from the takeoff point is weighted by cas caded 400 Hz and 800 Hz single pole low pass filters rectified and fed in opposition to both the steady state and the transient control circuits of the high frequency fixed band circuits There is no need to provide a smoothed MC signal for the overshoot suppressors of the high frequency fixed band stages because a fixed phase relationship exists between the stage signals and the control signals throughout the audio band MC4 controls the sliding band circuits of the low frequency stages The signal from the takeoff point is fed through a 200 Hz single pole low pass filter rec tified and fed in opposition to the sliding band control signals generated in the stages The phase of the mod ulation contro signal is optimized by the use of an all pass phase shifter as with MCI low frequency control signal ripple is thereby reduced MC4 is smoothed by a two stage 2 ms integrator to form MCS this signal is used to control the low frequency sliding band ov ershoot suppressors MC6 controls the low frequency fixed band circuits The signal from the takeoff point is weighted by cas caded 800 Hz and 1 6 kHz single pole high pass fi
165. uent interactions between these bands the noise reduction properties of the system under signal conditions are not altogether simple These properties are however amen able to investigation and measurement by the use of low level probe tones The overall noise reduction action of the system may be summarized as follows Band 1 provides noise reduc tion in the hum and rumble frequency range Band 2 in the mid audio range broadband noise crosstalk print through Bands 3 and 4 in the hiss range With aver age orchestral music Band 1 is compressed fairly often Band 2 almost all the time Band 3 fairly often and Band 4 rarely The noise reduction action thus arises most of the time from low and high frequency pre emphasis and complementary de emphasis The high frequency de emphasis not only attenuates hiss but in magnetic tape recording it reduces high frequency modu lation noise High frequency sidebands of lower frequency signals suffering frequency modulation due to scrape flutter are treated similarly SUMMARY The general principles of a noise reduction system suitable for high quality use have been described Low level signal components are amplified in four independent frequency bands prior to recording sending which is accomplished by adding the outputs of four filter and low level compressor channels to the main signal Dur ing reproduction the filter and compressor network is connected in a complementary way L
166. und by a sound higher in level A similar effect takes place in vision when the daylight sky makes the stars disappear Most audio signal processing systems operate quite indepen dently of the behavior of the ear and brain and always take the same action that they would if nobody were listening to the out put of the system The extraordinary purity of sound of Dolby SR recordings is due in large measure to an elegant cooperation of adaptive signal processing and auditory masking In masking a high level signal component raises the auditory threshold above and below the signal frequency Sounds lower in level and near it in frequency disappear completely in this psychoacoustic shadow Figure 14 In the Dolby SR circuit feedback of the signal characteristics determines how each filter in the Dolby SR circuit must change to most closely envelop the masking shadow This is the way that Dolby SR applies as much gain as possible everywhere in the spectrum The only region of the spectrum that is not boosted in gain is the region that is controlled by masking where audible low level information does not exist Dolby SR electronic signal processing is silently traded for auditory signal processing in that part of the spectrum It could fairly be said that although most of the Dolby SR system is on the circuit board some of it is in the human brain Immunizing the system to errors Another feature of Dolby SR is spectral skewing which r
167. up to an ambient temperature of 40 C 104 F note that the ambient temperature inside a poorly vented rack or inside a tape recorder may be considerably higher than that in the room STEP 3 Model 363 230V Unit for multi voltage units skip this step Open the fuse compartment door in the AC mains input connector with a small flat blade screwdriver as shown below and check that the fuse has the correct rating T250 mA 20mm time lag The fuse carrier must be inserted into the compartment with the orientation as shown Do not force the carrier into the compartment damage will result Close the compartment door making sure it clicks firmly into place METAL CLIP INSTALLED ON UNFUSED SIDE OF CARRIER Sa INSTALL FUSE Figure 2 1 Model 363 Multi Voltage Unit Open the voltage selector fuse compartment door in the AC mains input connector with a small flat blade screwdriver as shown below Rotate the voltage selector drum until it reads the correct mains voltage The drum may also be removed and replaced with the correct voltage displayed It will only fit one way around The compartment has positions for two fuses and will accept carriers for either 20mm or 1 25 fuses Only the lower fuse position is electrically connected Select the appropriate fuse and carrier and insert into the lower position with the arrow in the carrier in the same direction downwards as the arrows inside the compartment door A spare fuse
168. uring repro duction network Ga passes low level components noise back to the subtractor which partially cancels these com ponents in the signal from the channel In the process of reducing noise Gz and the subtractor also partially cancel low level signal components To compensate for this cancellation the network G which has the same characteristics as Go adds an identical component prior to recording sending These operations may be expressed in the following way If the input to the recording processor is x some function of time the signal in the channel is y and the output signal from the reproducing processor is z we have y 1 G x x 1 and zc Y zGo z or z 1 1 Ge z y 2 Combining Eqs 1 and 2 z 1 G x 1 1 Ga 2 x 3 The solution of interest is G4 Go z x Thus the output signal will be equal to the input signal if the recording and reproducing differential networks Oe the operators G and Go are identical on condition that G z is not allowed to become 1 no oscillation and that the functions in Eqs 1 and 2 are continuous and single valued no tracking ambiguity The prime requirement of any high quality noise re duction system that the signal should be unchanged overall is thereby satisfied and it is necessary only to choose an operator that yields a recording sending signal which is compatible with the channel and that produces satisfactory noise reduction prop
169. urn power input module power input rubber boot regulator ve LM317T regulator ve LM337T relay resistors 0 1 0 25 watt 10k0 20k0 resistors 1 0 4 watt 22E1 237 1k74 2k61 3k01 4k75 5k49 6k49 6k98 10k0 12k4 13K0 16K9 17k4 20k5 57k6 64k9 158k 243k 511k 3M01 switch SR off A switch check rec play switch bypass switch set up transformer power line transistor BC638 transistor n fet transistor p fet transformer tray up to S N 649 tray after S N 650 voltage selector wheel white noise IC MM5437 7 6 Circuit Reference only channel 1 listed 1C901 RV101 201 301 401 IC1101 IC1102 K201 401 R108 109 203 204 208 209 210 211 217 220 308 309 403 404 408 409 410 411 417 420 R106 107 206 207 212 213 216 219 306 307 406 407 412 413 416 419 R214 215 414 415 R1103 1106 R111 201 311 401 R1104 1105 R1202 R1011 1218 1238 1241 1250 1251 R1009 R402 1220 R113 202 228 229 R140 105 304 305 R1015 1204 R1239 1242 1243 R1201 R1203 R101 301 1006 1013 1237 R1205 1206 1214 1216 1217 R1236 R1235 R1234 R102 103 302 303 1227 1228 R1226 901 903 904 902 1302 1101 T1101 Q1101 QF101 102 301 302 901 1202 QF103 201 202 303 11101 101205 Dolby Part Number Cat No 373 Cat No 386 33003 12014 70106 70112 44025 44042 51015 13022 13023 10350 10871 10455 10473 10479 10499 10505 10511 10515 10530 10539 10541 10553
170. variable gain device can be operated at higher levels than would be possible if it were called upon to pass the whole dynamic range Tracking accuracy problems between units are also reduced since the transfer characteristic is largely deter mined by two readily controlled factors the compression threshold and the addition or subtraction coefficient of the differential component At low and high levels the possibility of mistracking is minimal and in the transi tion region it is not a difficult design matter to hold the error to a small fraction of a decibel A further tracking characteristic concerns compati bility of the system with the audio channel to a first order gain variation in the channel manifests itself only as a level change at the output not as an alteration of signal dynamics For the parameters used in the present system the maximum tracking error having a decibel value approximately equal to that of the dB error in gain occurs at about 30 dB below peak operating level where its effect is unobtrusive The method is thus in practice tolerant of moderate errors in gain This tolerance is especially significant in stereo as it enables the noise re duction system to operate without control signal inter connections 4OdB INPUT o T ME OUTPUT a Compression 4 4008 40dB INPUT o OUTPUT b Expansion e y ff d 4048 oY 74043 INPUT o VA E H rd OUTPUT c Differe
171. variety is used in your unit there will only be one style in any particular unit In two of the styles Figure 7 2 a and b the bodies are clamped by means of a cam lock screw which turns at the most through an eighth of a turn These styles are made by made by either Switchcraft or Neutrik These connectors while using the same principles have the locking screws in slightly different positions and operate in opposite senses The screws must be unlocked from the rear of the unit even though there appears to be access from the inside of the unit for the Switchcraft version A more recent Switchcraft design uses a simple countersunk locking screw on the top of the body of the connector towards the rear Figure 1 2 C unscrew NEUTRIK unlock SWITCHCRAFT SWITCHCRAFT b a EM c 8a 8b Figure 7 2 In the case of the designs using a cam lock system identify the connector manufacturer by reading the name at the bottom outside of the connector If a Switchcraft component the screws will be towards the top right of the connector and are silver coloured to stand out against the black colour of the body and insulator Unlock the shells using a thin flat bladed screwdriver by turning the screws clockwise If a Neutrik connector the screws are on the centre line of the connector just above the centre A very thin bladed screwdriver must be used tip width about 1 8 mm around half that of the screw hole alternatively the
172. was designed not only to provide more noise reduction than A type but also to enable analog master recordings that equal or surpass 16 bit digital recordings with respect to overall dynamic range Today analog with Dolby SR is still a preferred format for some musicians producers and record ing and mastering facilities and is the standard format for the analog sound tracks of virtually all feature films 363 SPECIFICATIONS Frequency Response 20 Hz 20 kHz I dB encode decode at any level Overall Distortion T HD lt 0 2 at Dolby level Processor Headroom 21 dB above Dolby level Overall Dynamic Range SR 105 dB clipping level to CCIR ARM noise level 105 dB clipping level to NAB A weighted noise level 95 dB clipping level to unweighted noise level 20 Hz 20 kHz A type 104 dB clipping level to CCIR ARM noise level 105 dB clipping level to unweighted noise level 20 Hz 20 kHz Typical Obtainable Dynamic Range SR 90 95 dB A Type 75 80 dB typical at 38 cm s 15ips tape speed Matching Between Units dB at any level and any frequency 20 Hz 20 kHz Crosstalk lt 100 dB 20 Hz 20 kHz processor off or encode decode Signal Delay 6 us for a single channel 12 us for overall encode decode system Analog Audio Inputs 0 d r 0 775 Vous XLR connectors balanced 20 kQ common mode rejection gt 55 dB 50 Hz 10 kHz maximum input level 27 dBr balanced 21 dBr unbalanced input line
173. will yield a disappointing result which does not agree with the listening experience Most of the noise being measured will be at and beyond the extremes of the audible range this can be confirmed by examination of the noise on a spectrum analyzer On an oscilloscope the unreduced very low frequency noise will yield an unsteady base line which has in the past mistakenly been interpreted as low frequency instability it actually is the original noise waveform with the audible frequencies removed leaving the inaudible very low frequencies Weighted noise measurements yield answers which correlate more closely with audibility However depending on the amount of ultrasonic noise the A weighting curve may provide inadequate attenuation of noise above 20 kHz and therefore must be supplemented by an audio bandpass filter say 20 kHz preferably using a 4 pole design to eliminate inaudible frequencies CCIR 468 and CCIR ARM give results which agree with listening provided the test equipment works properly Unfortunately many integrated instruments meters containing switchable weighting filters have inadequate headroom in their amplifiers and or high frequency crosstalk across their Switches and as a result give grossly inaccurate results Such problems can be identified while measuring noise reduced by SR by switching the meter from wide band not audio band but 100 kHz or more to weighted measurement If it is necessary to increase the meter gain by
174. y 388 mV From here the main path continues to the Output 2 pin The signal also passes to a side chain which comprises the signal processing circuits A switch at the output of these allows the output of the selected processor to continue to the mixing stages in modules with only one type of processing this switch is not present In the record mode the side chain signal is added to the main signal to form the encoded signal which passes to Output 2 In the play mode the side chain signal is subtracted from the main signal The monitor output of the module is Output 1 in the record mode this is the input unencoded signal and in the play mode it is the decoded signal thus in all cases the signal at Output 1 is a normal signal 6 4 6 9 2 Set up In Set up Dolby noise and Dolby tone calibration signals are present at the appropriate inputs of the module The appropriate calibration signal is selected by IC802a The Cat No 350 has no connection to the Dolby tone input on pin 18 In the Cat No 450 Dolby tone on pin 18 is connected directly to the normal set up switch shown here as IC803a and there is no connection to the Dolby noise input on pin 20 Switch IC803a sends the selected calibration signal to Output 2 and then to the To recorder output of the unit Note that while the A type Dolby tone is recorded at normal level at or near O VU the SR Dolby noise is reached about 15 dB below to avoid any tape saturation Circuits
175. y behavior in the decoder Note in passing that it IS possible to use instantaneous or near instantaneous attack to eliminate overshoot but such an approach can be disastrous subjectively because of wide band modulation products The Dolby systems use a dual path configuration in which the input signal passes directly from input to output the processing consists of the addition or subtraction of a further signal whose maximum amplitude is small compared with the maximum amplitude of the input This method imposes an upper threshold above which gain reduction ceases see Figures 5 3 and 5 4 note the fixed gains at low and high levels This shape of characteristic permits overshoot suppression within the further path The result is that overshoots resulting from high level signals are much smaller than the degree of gain reduction and there is litle danger of transient overload of the recording or transmission system Variable Variable frequency frequency response and response and overshoot overshoot suppression suppression Recording or transmission Addition Subtraction Figure 5 3 Dual path configuration of all Dolby systems 5 3 40 dB Input 0 40dB Input 0 Output Output 40 dB 40 dB Compression Expansion Figure GA General form of compression and expansion characteristic of all Dolby systems 5 3 Dolby A type noise reduction A type noise reduction nr is a professional system
176. z pass in the fixed band steady state control circuit lower middle of Fig 11 To prevent interfer ence with the low frequency overshoot suppression circuit at low frequencies the signal is further filtered by a 200 Hz single pole high pass network as in the primary overshoot suppression circuits the filtered signal is then rectified Note that in Fig 10 this filter is not shown for clarity On a steady state basis the pass band contro circuit controls the circuit at very low frequencies via the maximum selector circuit this arrangement allows the main control circuit rectifier to serve a double function The de signal is opposed by MC6 in order to phase out the secondary overshoot suppression effect at high frequencies An optimal phase relationship is obtained between the rectified signal and MC6 apart from the effect of the 200 Hz filter which is negligible An ideal tracking effect 1s achieved between the steady state and secondary ov ershoot suppression thresholds The effect of the 800 Hz and 1 6 kHz frequency weighting networks is to introduce a time delay into the secondary overshoot suppression signal The ef fective delay is significantly reduced by increasing the gain used in the secondary overshoot suppressor circuit and applying limiting The resultant overshoot sup pression signal is more in the nature of a nearly fixed amplitude impulse applied in the rare circumstances when necessary than itis a proportional r

Download Pdf Manuals

Related Search

Related Contents

Fujitsu P3PC-1562-01ENZ0 Photo Scanner User Manual  Mode d`emploi Scrubmaster B140  Hampton Bay AL508-ORB Use and Care Manual  Descargar Ficha Técnica  吹田市立武道館指定管理者 管理運営基準 平成22年7月1日 吹田市  Epson EB-1723 video projector  Kensington ClickSafe® Point 2 Point Keyed Lock  MA507 Oxicapnografo  [ès DITS  Origin Storage 500GB TLC SATA  

Copyright © All rights reserved. Failed to retrieve file