Tabel of Content:
1. Outline
2. Recording Format
3. Data Arrangement on Program Track
4. Longitudinal Tracks
5. Head Configuration
6. Signal Processing
7. Digital Audio Interface

8. Outline of the Digital Signal Processing

9. Error Correction Coding





The Sony DVW series VTR is a 1/2-inch component digital VTR which uses a new "Digital Betacam" format. This "Digital Betacam" format has been developed as the digital version of the analog Betacam format that has been now accepted as a standard in the broadcasting, production, and Electronic News Gathering (ENG) fields of the world. That is, the Digital Betacam VTR can perform high-performance digital recording and playback while keeping the playback compatibility of the conventional analog Betacam tape, and has almost the same operation performance as in the conventional analog Betacam VTR.

To perform such a workhorse VTR, a new bit rate reduction format (i.e., coefficient recording) has been developed. The data rate is compressed into approximately 1/2, and compressed data is recorded on tape. Moreover, the introduction of a high-performance, high-precision heads and drum, and a new automatic tracking system has enabled a drastic reduction of the width of the recording track. As a result, the Digital Betacam VTR is able to use the same size 1/2-inch Digital Betacam cassette (8 and L) as in the conventional Betacam cassette. The L cassette makes continuous recording of more than 120 minutes possible.

In the DVW-A500, the analog tape (metal and oxide) recorded based on the Betacam/Betacam 8P format can also be played back. Therefore, currently-owned Betacam recorded tapes can also be used directly. Furthermore, an analog Betacam SP VTR (especially Camcorder) can also be jointly used in the Digital Betacam system.


Figure 1. Tape Pattern (525/60


Fig. 1 shows the tape pattern in the 525/60 system. To explain the operation bf the recorder (DVW-A500) with a Betacam playback function, this figure shows the tape pattern of the Digital Betacam format at right, and the tape pattern of
the Betacam SP format at left together.

In the Digital Betacam format, one-field video data is recorded on six helical tracks (program tracks). In other words, since one-field data is recorded while the drum rotates by 1.5, the number of drum rotations is approximately 90 per second. In the Betacam SP format, since one-field data is recorded while the drum rotates by 1/2, the number of drum rotations is approximately 30 per second. As shown in the figure, the tape speed of the Digital Betacam format is lower a little than that of the Betacam SP format.

The DVW-A500 detects whether the cassette tape to be played back is based on the Digital Betacam format or the analog Betacam format, by the ID hole of the cassette, and switches the control mode automatically. As the result, the tape is played back at the same drum rotation and tape speed as when it is recorded, and parameters are set so that the optimum tracking is obtained.

The drum diameter of the Digital Betacam VTR is larger about 9% than that of the Betacam SP VTR. Since the relative speed of the head to tape increases proportionally to the drum diameter even if the drum rotation is made same, the signal recorded on the helical track of the Betacam SP tape is played back as time-compressed signal about 9%. In other words, the frequency of the played back RF signal increases. However, the RF signal is processed by a signal processing circuit in the later stage so that it can obtain the same
performance as before.

The tape pattern of the Digital Betacam format has three longitudinal tracks (i.e., control track, time code track, and cue audio track). These tracks coincide with the tape pattern of the Betacam SP format. Therefore, the longitudinal tracks of the Betacam SP tape can be played back by the stationary heads of the Digital Betacam VTR. In the Betacam SP format, however, the fourth longitudinal track that records an audio CH1 signal is provided in parallel. The VTR with a Betacam playback function thus has a two-channel audio head.


Fig. 2 shows the arrangement of data recorded on helical tracks (program tracks). The size of each section in the figure is not the actual dimension ratio. For more information on the size, refer to Fig. 1.

Figure 2. Data Arrangement on Program Track

The Digital Betacam format uses an azimuth recording system to perform high-density recording. Namely, two heads with an azimuth angle of approximately 15 degrees in the opposite direction to each other are paired to record two program tracks (track O and 1) while the drum rotates by 1/2. One-field video data and audio data are recorded on six program tracks. Fig. 2 shows the arrangement of one-frame data. Field 1 data is recorded on the first-half six tracks, and field 2 data is recorded on the latter-half six tracks. Each program track is constructed to have four audio sectors in the center and to have two video sectors beyond the tracking pilot signals put before and behind the audio sectors. In other words, one-field video data is dispersed into twelve video sectors for recording, and four-channel audio data is dispersed into six audio sectors for recording for every channel. As shown in Fig. 2, each channel for the audio sectors is arranged so that the danger of error generation disperses.

To identify the beginning and end of a sector, preamble and post amble data are added before and behind all the sectors. Also, edit gaps are installed between the sectors so that each channel can be edited independently. Tracking pilot signals are used in an automatic tracking system that secures high-speed and high-precision tracking during editing.


As shown in Fig. 1, a control track, time code track, and cue audio track are provided in the longitudinal direction of the tape. Fig. 3 shows the CTL signal and time code signal in the 525/60 system, and the timing relationship between these signals and video reference signals. The CTL signal is recorded by a 60-Hz rectangular pulse in the 525/60 system and controls the start timing of each field data at the rising edge of the pulse. The CTL signal is also modulated the duty cycle of the rectangular wave to 65 : 35 in the first field of a color frame so as to identify the field number and control the color framing. Moreover, the duty cycle is modulated to 35 : 65 in the first field of an audio 5-field sequence (excepting the overlapped field with the first field of a color frame) so as to use for the 5-field sequence control.

Figure 3. Digital Betacam Timing Chart (525/60)

The time code signal uses a conventional longitudinal time code based on the SMPTE/EBU standard. One frame consists of 80 bits. Frame number, second, minute, hour, and sync word data items are coded by a bi phase mark coding method and recorded for every frame. A user bit is written in the time data as a binary group. The cue audio signal is recorded by an ordinary AC bias method.


Figure 4. shows the arrangement of the stationary and rotary heads in the DVW-A500, and its top view. However, the rotary heads can be actually viewed only their tips from the side of the drum.

Figure 4. Arrangement of Stationary and Rotary Heads

In the illustration that shows the rotary heads, ADVANCE heads and V/C heads installed in pairs on the square plates like the bi morph plates are dynamic tracking heads. These heads are used to play back a digital tape and an analog tape respectively. Therefore, the model (DVW-500 for a recorder) without a Betacam playback function has no V/C heads.

Both ADVANCE heads and CONFIDENCE heads are used to play back a digital tape. In usual PB mode, the ADVANCE heads are used to play back. In the REC and INSERT modes, the ADVANCE heads play back the signal before erasure as pre read, and the CONFIDENCE heads play back a recorded signal to confirm that the signal has been recorded. For that purpose, the ADVANCE heads and the CONFIDENCE heads are installed in the specified height on the rotary drum so that they can trace the before track and after track respectively toward the REC head.


The interface of the Sony Digital Betacam VTR is based on the component serial digital interface format SMPTE 259M/GCIR-656 Part III. The digital audio interface is based on the AES/EBU format.

6.1 Serial video interface

Each data of a digital component video signal and four-channel digital audio signal is serialized and can be transmitted over a long distance of more than 200 m using one coaxial cable. The transmission rate is 270M bits per second. This serial interface uses an S-NAZI (Scrambled Non-return to Zero Inversion) channel code system for digital data coding. The NAZI transmission has a merit that the contents of data do not change even if a signal is inverted in polarity. Therefore, this system has the advantage that a spurious radiation can be reduced by using the positive/negative polarity simultaneously for serial signal transmission or that the IC can be constituted by a stable, low-power consumption differential amplifier. Scramble processing also enables a clock to be reproduced on the receiving side even if data"0" continues.

Video data is processed in conformity with the SMPTE 125M/GGIA-601 standard. The luminance signal (V) is sampled in 13.5 MHz, and the color-difference signals (GA and GB) are sampled in 6.75 MHz. The luminance signal and color difference signals are then multiplexed in the order of "CB, Y, CA, Y, CB, Y, CA, Y, ..." and transmitted at the rate of 27M words per second.

Figure 5. Relationship between video signal levels and quantizing levels


Fig. 5 shows the relationship between the video signal level and the quantizing level. Fig. 6 shows the sampling format in a horizontal line, and Fig. 7 shows the sampling format in a field.

Figure 6. Relationship between H sync and sampling (525/60)
Figure 7. Sampling Format

As shown in Fig. 6, the four-word timing reference signals (TRS) are added before and behind the digital active line. The top TAS signal defines the start of the active line and is called an SAV (Start of Active Video). The last TAS signal defines the end of the active line and is called an EA V (End of Active Video). Fig. 8 shows the format of a timing reference signal corresponding to 10 bits. Since the first three words of four words constitute a unique pattern of 3FF, 000, and 000 (HEX), they can be detected as a sync word. For serial transmission, when this sync word is detected, the bit to be sent next is judged to be an LSB, and becomes a reference for converting the serial bits into every 10 bits of parallel words. The forth word (XVZ) defines the field status, and vertical and horizontal blanking timing.

Figure 8. Timing reference signal (525/60)

7. Digital Audio Interface

The digital audio interface is classified into the serial digital interface combined with a video signal and the AES/EBU digital interface employing an XLR connector. In both interfaces, the digital audio data is based on the AES/EBU format, and corresponds with 48 kHz sampling and 20-bit linear quantizing. For the serial digital interface, the digital audio data is inserted into an auxiliary data area during horizontal and vertical blanking periods. For the AES/EBU digital interface, two channels of audio data are multiplexed and stored in a data frame for every audio-sampling period. One frame consists of two sub frames. The left channel data in the stereo mode enters into sub frame A, and the right channel data enters into sub frame B. Fig. 9 shows the format of frame and sub frames. The
sub frames contains various data items (shown in the figure)
as well as audio data and consist of 32 bits.


Figure 9. Frame and sub frame formats

8. Outline of the Digital Signal Processing

The digital signal processing of the Sony Digital Betacam VTR conforms to the D-1 format except that the bit rate reduction processing is added. Fig. 10 shows the block diagram of the recording system, and Fig. 11 shows the block diagram of the playback system.


Figure 10. Digital signal recording system
Figure 11. Digital signal playback system


8.1 Outline of the recording system

(a) Input interface
For the analog or digital video/audio input signals, the video signal is converted into parallel video data, and the audio signal into serial audio data.
(b) Bit rate reduction encoder
The video data is blocked after it is field-shuffled, then compressed into a data rate of approximate 1/2 according to the newly developed bit rate reduction format.
(c) Audio data processor
The audio data processor performs the audio data selection, gain control, REC verification, and other data processing.
(d) ECC encoder
The ECC encoder performs the error correction coding for error correction in a playback system. The video data that has been sent after bit rate reduction is added the outer error collection code (ECC) and track-interleaved. The audio data sent from the processor is also added the outer ECC and field-shuffled. Then, the video data and audio data are multiplexed, added the ID data, and inner ECC encoded. Sync data is also added to the resultant data.
(e) REC driver
The REC driver converts parallel REC data into serial data and sends a channel-encoded REC current to the head.

8.2 Outline of the playback system

In the playback system, signals are processed in the procedure opposite to the recording system. The audio data error that couldn't be corrected using inner and outer error correction codes is concealed before the audio data is output from an ECC decoder. The video data error that couldn't be corrected is concealed after the video data is returned to the former video data.

9. Error Correction Coding

In the Sony Digital Betacam format, error correction takes place by means of a two-dimensional product code of the inner and outer codes using the Reed-Solomon code. The structure of the inner code block that constitutes an ECC block is common in video and audio systems. However, the structure of the outer code block differs in video and audio systems. A data block of 180 bytes obtained when sync data is added to one inner code block data is called a sync block, which is the basic unit of a video sector and audio sector. Fig. 12 shows the structure of the ECC blocks and sync block in the 525/60 system.One video sector consists of 106 sync blocks. In other words, one video ECC block constitutes one video sector. Therefore, the two ECC blocks shown in the figure below are recorded on one track, and one field corresponds to 12 ECC blocks. One audio sector consists of five sync blocks. For the audio data in one channel, one field consists of six sectors (i.e., 30 sync blocks). Therefore, the three ECC blocks shown in the figure correspond to one field in one audio channel.

Figure 12. Structure of ECC blocks