NTSC/PAL/SECAM 4×10-Bit Digital Video Decoder
With Macrovision™ Detection, YPbPr Inputs, 5-Line Comb
Filter, and SCART Support
Data Manual
PRODUCTION DATA information is current as of publication date.
Products conform to specifications per the terms of the Texas
Instruments standard warranty. Production processing does not
necessarily include testing of all parameters.
With Macrovision™ Detection, YPbPr Inputs, 5-Line Comb Filter, and SCART Support
Check for Samples: TVP5146M2
1Introduction
1.1Features
1234
• Four 30-MSPS 11-bit A/D channels with
programmable gain control
• Supports NTSC (J, M, 4.43), PAL (B, D, G, H, I,
M, N, Nc, 60), SECAM (B, D, G, K, K1, L), CVBS,
and S-Video– Line-locked internal pixel sampling clock
• Supports analog SD YPbPr component and
SCART (RGB/YPbPr + CVBS) video formats
with embedded sync– Genlock output RTC format for downstream
• Ten analog video input terminals for
multisource connection• Certified Macrovision™ copy protection
• User-programmable video output formats
– 10-bit ITU-R BT.656 4:2:2 YCbCr with
embedded syncs
– 10-bit 4:2:2 YCbCr with separate syncs
– 20-bit 4:2:2 YCbCr with separate syncs
– 2× sampled raw VBI data in active video
during a vertical blanking period– Wide screen signaling (WSS)
– Sliced VBI data during a vertical blanking– Copy generation management system
period or active video period (full field mode)(CGMS)
• HSYNC/VSYNC outputs with programmable– Video program system (VPS/PDC)
position, polarity, width, and field ID (FID)
output
• Component video processing
– Gain (contrast) and offset (brightness)
adjustments
– Automatic component video detection
(525/625)
– Color space conversion from RGB to YCbCr
• Composite and S-Video processing
– Adaptive 2-D 5-line adaptive comb filter forwith power-save and power-down modes
composite video inputs; chroma-trap
available
– Automatic video standard detection
(NTSC/PAL/SECAM) and switching
– Luma-peaking with programmable gain
– Patented chroma transient improvement
(CTI) circuit
– Patented architecture for locking to weak,
noisy, or unstable signals
– Single 14.31818-MHz reference crystal for all
standards
generation with horizontal and vertical lock
signal outputs
video encoder synchronization
detection
• Available in commercial (0°C to 70°C) and
industrial (−40°C to 85°C) temperature ranges
• VBI data processor
– Teletext (NABTS, WST)
– CC and extended data service (EDS)
– Vertical interval time code (VITC)
– Gemstar™ 1×/2× mode
– V-Chip decoding
– Register readback of CC, WSS (CGMS),
VPS/PDC, VITC and Gemstar 1×/2× sliced
data
• I2C host port interface
• Reduced power consumption: 1.8-V digital
core, 3.3-V for digital I/O, and 1.8-V analog core
• 80-terminal TQFP PowerPAD™ package
• RGB Sync on Green is not currently supported,
and all references to "RGB" in this data manual
imply "SCART RGB"
1
Please be aware that an important notice concerning availability, standard warranty, and use in critical applications of Texas
Instruments semiconductor products and disclaimers thereto appears at the end of this data sheet.
2PowerPAD is a trademark of Texas Instruments.
3Gemstar is a trademark of Gemstar-TV Guide Intermational.
4Macrovision is a trademark of Macrovision Corporation.
PRODUCTION DATA information is current as of publication date.
Products conform to specifications per the terms of the Texas
Instruments standard warranty. Production processing does not
necessarily include testing of all parameters.
The TVP5146M2 device is a high-quality single-chip digital video decoder that digitizes and decodes all
popular baseband analog video formats into digital component video. The TVP5146M2 decoder supports
the analog-to-digital (A/D) conversion of component RGB and YPbPr signals, as well as the A/D
conversion and decoding of NTSC, PAL, and SECAM composite and S-Video into component YCbCr.
This decoder includes four 11-bit 30-MSPS A/D converters (ADCs). Preceding each ADC in the device,
the corresponding analog channel contains an analog circuit that clamps the input to a reference voltage
and applies a programmable gain and offset. A total of ten video input terminals can be configured to a
combination of RGB, YPbPr, CVBS, or S-Video video inputs.
Component, composite, or S-Video signals are sampled at 2× the ITU-R BT.601 clock frequency, line
locked, and are then decimated to the 1× pixel rate. CVBS decoding utilizes 5-line adaptive comb filtering
for both the luma and chroma data paths to reduce both cross-luma and cross-chroma artifacts. A chroma
trap filter is also available. On CVBS and S-Video inputs, the user can control video characteristics, such
as contrast, brightness, saturation, and hue via an I2C host port interface. Furthermore, luma peaking
(sharpness) with programmable gain is included, as well as a patented chroma transient improvement
(CTI) circuit.
A built-in color space converter is applied to decoded component RGB data.
Two output formats can be selected: 20-bit 4:2:2 YCbCr or 10-bit 4:2:2 YCbCr.
The TVP5146M2 decoder generates synchronization, blanking, field, active video window, horizontal and
vertical syncs, clock, genlock (for downstream video encoder synchronization), host CPU interrupt, and
programmable logic I/O signals, in addition to digital video outputs.
www.ti.com
The TVP5146M2 decoder includes methods for advanced vertical blanking interval (VBI) data retrieval.
The VBI data processor (VDP) slices, parses, and performs error checking on teletext, closed caption
(CC), and other VBI data. A built-in FIFO stores up to 11 lines of teletext data and, with proper host port
synchronization, full-screen teletext retrieval is possible. The TVP5146M2 decoder can pass through the
output formatter 2× the sampled raw luma data for host-based VBI processing.
The main blocks of the TVP5146M2 decoder include:
•Robust sync detection for weak and noisy signals as well as VCR trick modes
•Y/C separation by 2-D 5-line adaptive comb or chroma trap filter
•Fast-switch input for pixel-by-pixel switching between CVBS and YPbPr/RGB component video inputs
(SCART support)
•Four 11-bit 30-MSPS ADCs with analog preprocessors [clamp and automatic gain control (AGC)]
•Luminance processor
•Chrominance processor
•Component processor
•Clock/timing processor and power-down control
•Software-controlled power-saving standby mode
•Output formatter
•I2C host port interface
•VBI data processor
•Macrovision™ copy protection detection circuit (Type 1, 2, 3, and separate color stripe detection)
Throughout this data manual, several conventions are used to convey information. These conventions are
listed below:
•To identify a binary number or field, a lower case b follows the numbers. For example: 000b is a 3-bit
binary field.
•To identify a hexadecimal number or field, a lower case h follows the numbers. For example: 8AFh is a
12-bit hexadecimal field.
•All other numbers that appear in this document that do not have either a b or h following the number
are assumed to be decimal format.
•If the signal or terminal name has a bar above the name (for example, RESETB), then this indicates
the logical NOT function. When asserted, this signal is a logic low, 0, or 0b.
•RSVD indicates that the referenced item is reserved.
1.6Ordering Information
T
A
0°C to 70°C
-40°C to 85°C
(1) For the most current package and ordering information, see the Package Option Addendum at the end
of this document, or see the TI web site at www.ti.com.
(2) Package drawings, standard packing quantities, thermal data, symbolization, and PCB design
DATACLK40OLine-locked data output clock
XTAL174I
XTAL275OExternal clock reference. Not connected if XTAL1 is driven by an external single-ended oscillator.
Digital Video
57, 58,
59, 60,Digital video output of CbCr, C_9 is MSB and C_0 is LSB. C_0 and C_[9-2] can be used as
C_[9:0]/GPIO63, 64,Oprogrammable general purpose I/O. C_1 (pin 69) requires an external pulldown resistor and should
65, 66,not be used for general purpose I/O.
69, 70
43, 44,
45, 46,
Y_[9:0]47, 50,O
51, 52,
53, 54
Miscellaneous Signals
FSS/GPIO35I/O(YPbPr/RGB) and the composite video input.
GLCO/I2CA37I/O
INTREQ30OInterrupt request
PWDN33I1 = Power down
RESETB34IReset, active low
Host Interface
SCL28I/OI2C clock
SDA29I/OI2C data bus
I/ODESCRIPTION
VI_1_x: Analog video input for CVBS/Pb/B/C
VI_2_x: Analog video input for CVBS/Y/G
VI_3_x: Analog video input for CVBS/Pr/R/C
VI_4_A: Analog video input for CVBS/Y
IUp to 10 composite, 4 S-video, and 2 composite or 3 component video inputs (or a combination
thereof) can be supported.
The inputs must be ac coupled. The recommended coupling capacitor is 0.1 µF.
The possible input configurations are listed in the input select register at I2C subaddress 00h (see
Table 2-12).
External clock reference. It can be connected to an external oscillator with a 1.8-V compatible clock
signal or to a 14.31818-MHz crystal oscillator.
Digital video output of Y/YCbCr; Y_9 is MSB and Y_0 is LSB.
For the 8-bit mode, the two LSBs are ignored. Unused outputs can be left unconnected.
Fast-switch (blanking) input. Switching signal between the synchronous component video
Programmable general-purpose I/O
Genlock control output (GLCO)
During reset, this terminal is an input used to program the I2C address LSB.
Figure 2-1 shows a functional diagram of the analog processors and ADCs. This block provides the
analog interface to all video inputs. It accepts up to ten inputs and performs source selection, video
clamping, video amplification, A/D conversion, and gain and offset adjustments to center the digitized
video signal.
The TVP5146M2 decoder has four analog channels that accept up to ten video inputs. The user can
configure the internal analog video switches via the I2C interface. The ten analog video inputs can be used
for different input configurations, some of which are:
•Up to ten selectable individual composite video inputs, as well as other combinations of YPbPr,
S-Video, and SCART can be supported (see Table 2-12)
The input selection is performed by the input select register at I2C subaddress 00h (see Table 2-12).
2.1.2Analog Input Clamping
An internal clamping circuit restores the ac-coupled video signal to a fixed dc level. The clamping circuit
provides line-by-line restoration of the video sync level to a fixed dc reference voltage. The selection
between bottom and mid clamp is performed automatically by the TVP5146M2 decoder.
2.1.3Automatic Gain Control
The TVP5146M2 decoder uses four programmable gain amplifiers (PGAs), one per channel. The PGA
can scale a signal with a voltage-input compliance of 0.5 VPPto 2 VPPto a full-scale 10-bit A/D output
code range. A 4-bit code sets the coarse gain with individual adjustment per channel. Minimum gain
corresponds to a code 0x0 (2-VPP full-scale input, –6-dB gain) while maximum gain corresponds to code
0xF (0.5-VPP full scale, +6-dB gain). The TVP5146M2 decoder also has 12-bit fine gain controls for each
channel and applies them independently to coarse gain controls. For composite video, the input video
signal amplitude can vary significantly from the nominal level of 1 VPP. The TVP5146M2 decoder can
adjust its PGA setting automatically: an AGC can be enabled and can adjust the signal amplitude such
that the maximum range of the ADC is reached without clipping. Some nonstandard video signals contain
peak white levels that saturate the ADC. In these cases, the AGC automatically cuts back gain to avoid
clipping. If the AGC is on, then the TVP5146M2 decoder can read the gain currently being used.
SLES141F–JULY 2005–REVISED NOVEMBER 2010
The TVP5146M2 AGC comprises the front-end AGC before Y/C separation and the back-end AGC after
Y/C separation. The back-end AGC restores the optimum system gain whenever an amplitude reference,
such as the composite peak (which is only relevant before Y/C separation), forces the front-end AGC to
set the gain too low. The front-end and back-end AGC algorithms can use up to four amplitude references:
sync height, color burst amplitude, composite peak, and luma peak.
The specific amplitude references being used by the front-end and back-end AGC algorithms can be
independently controlled using the AGC white peak processing register located at subaddress 74h. The
TVP5146M2 gain increment speed and gain increment delay can be controlled using the AGC increment
speed register located at subaddress 78h and the AGC increment delay register located at subaddress
79h, respectively.
2.1.4ADCs
All ADCs have a resolution of 11 bits and can operate up to 30 MSPS. All A/D channels receive an
identical clock from the on-chip phase-locked loop (PLL) at a frequency between 24 MHz and 30 MHz. All
ADC reference voltages are generated internally.
2.2Digital Video Processing
Figure 2-2 is a block diagram of the TVP5146M2 digital video decoder processor. This processor receives
digitized video signals from the ADCs and performs composite processing for CVBS and S-Video inputs,
YCbCr signal enhancements for CVBS and S-Video inputs, and YPbPr/RGB processing for component
video inputs. It also generates horizontal and vertical syncs and other output control signals, such as
genlock for CVBS and S-Video inputs. Additionally, it can provide field identification, horizontal and vertical
lock, vertical blanking, and active video window indication signals. The digital data output can be
programmed to two formats: 20-bit 4:2:2 with external syncs or 10-bit 4:2:2 with embedded/separate
syncs. The circuit detects pseudosync pulses, AGC pulses, and color striping in Macrovision-encoded
copy-protected material. Information present in the VBI interval can be retrieved and either inserted in the
ITU-R BT.656 output as ancillary data or stored in internal FIFO and/or registers for retrieval via the host
port interface.
www.ti.com
Figure 2-2. Digital Video Processing Block Diagram
2.2.12x Decimation Filter
All input signals are oversampled by a factor of two (27 MHz). The A/D outputs first pass through
decimation filters that reduce the data rate to 1× the pixel rate. The decimation filter is a half-band filter.
Oversampling and decimation filtering can effectively increase the overall signal-to-noise ratio by 3 dB.
2.2.2Composite Processor
Figure 2-3 is a block diagram of the TVP5146M2 digital composite video processing circuit. This circuit
receives a digitized composite or S-Video signal from the ADCs and performs Y/C separation (bypassed
for S-Video input), chroma demodulation for PAL/NTSC and SECAM, and YUV signal enhancements.
The 10-bit composite video is multiplied by the subcarrier signals in the quadrature demodulator to
generate color difference signals U and V. The U and V signals are then sent to low-pass filters to achieve
the desired bandwidth. An adaptive 5-line comb filter separates UV from Y based on the unique property
of color phase shifts from line to line. The chroma is remodulated through a quadrature modulator and
subtracted from line-delayed composite video to generate luma. This form of Y/C separation is completely
complementary, thus there is no loss of information. However, in some applications, it is desirable to limit
the U/V bandwidth to avoid crosstalk. In that case, notch filters can be turned on. To accommodate some
viewing preferences, a peaking filter is also available in the luma path. Contrast, brightness, sharpness,
hue, and saturation controls are programmable through the host port.
High filter bandwidth preserves sharp color transitions and produces crisp color boundaries. However, for
video sources that have asymmetrical U and V side bands, it is desirable to limit the filter bandwidth to
avoid UV crosstalk. The color low-pass filter bandwidth is programmable to enable one of the three notch
filters. Figure 2-4 and Figure 2-5 represent the frequency responses of the wideband color low-pass filters.
www.ti.com
Figure 2-4. Color Low-Pass Filter FrequencyFigure 2-5. Color Low-Pass Filter With Filter
Y/C separation can be done using adaptive 5-line (5-H delay) comb filters or a chroma trap filter. The
comb filter can be selectively bypassed in the luma or chroma path. If the comb filter is bypassed in the
luma path, chroma trap filters are used which are shown in Figure 2-6 and Figure 2-7. The TI patented
adaptive comb filter algorithm reduces artifacts such as hanging dots at color boundaries. It detects and
properly handles false colors in high frequency luminance images, such as a multiburst pattern or circle
pattern. Adaptive comb filtering is the recommended mode of operation.
SLES141F–JULY 2005–REVISED NOVEMBER 2010
Figure 2-6. Chroma Trap Filter Frequency Response, Figure 2-7. Chroma Trap Filter Frequency Response,
The digitized composite video signal passes through either a luminance comb filter or a chroma trap filter,
either of which removes chrominance information from the composite signal to generate a luminance
signal. The luminance signal is then fed into the input of a peaking circuit. Figure 2-8 illustrates the basic
functions of the luminance data path. In the case of S-Video, the luminance signal bypasses the comb
filter or chroma trap filter and is fed directly to the circuit. High-frequency components of the luminance
signal are enhanced by a peaking filter (sharpness). Figure 2-9 shows the characteristics of the peaking
filter at four different gain settings that are programmable via the host port.
CTI enhances horizontal color transients by delay modulation for both color difference signals. The
operation must be performed only on YCbCr-formatted data. The color difference signal transition points
are maintained, but the edges are enhanced for signals that have bandwidth-limited color components (for
example, CVBS and S-Video).
The component video processing block supports a user-selectable contrast, brightness, and saturation
adjustment in YCbCr output formats. For YCbCr output formats, gain and offset values are applied to the
luma data path to map the pixel values to the correct output range (for 10-bit Y
and to provide a means of adjusting contrast and brightness. For Y, digital contrast (gain) and brightness
(offset) factors can vary from 0 to 255. The contrast control adjusts the amplitude range of the Y output
centered at the midpoint of the output code range. The limit block limits the output to the ITU-R BT.601
range (Y
For CbCr components, a saturation (gain) factor is applied to the CbCr inputs to map them to the CbCr
output code range and provide saturation control. Similarly, the limit block can limit CbCr outputs to a valid
range:
Cb,Cr
to Y
min
= 64 / Cb,Cr
min
) or an extended range, depending on a user setting.
max
Figure 2-10. Y Component Gain, Offset, Limit
= 960
max
SLES141F–JULY 2005–REVISED NOVEMBER 2010
= 64 and Y
min
max
= 940),
Figure 2-11. CbCr Component Gain, Offset, Limit
2.2.6Color Space Conversion
The formulas for RGB to YCbCr conversion are given as:
Y = 0.299 × R + 0.587 × G + 0.114 × B
Cb = –0.172 × R – 0.339 × G + 0.511 × B + 512
Cr = 0.511 × R – 0.428 × G – 0.083 × B + 512
2.3Clock Circuits
An internal line-locked PLL generates the system and pixel clocks. A 14.31818-MHz clock is required to
drive the PLL. This can be input to the TVP5146M2 decoder at the 1.8-V level on terminal 74 (XTAL1), or
a crystal of 14.31818-MHz fundamental resonant frequency can be connected across terminals 74 and 75
(XTAL2). If a parallel resonant circuit is used as shown in Figure 2-12, then the external capacitors must
have the following relationship:
CL1= CL2= 2CL− C
Where,
C
is the terminal capacitance with respect to ground
STRAY
CLis the crystal load capacitance specified by the crystal manufacturer
Figure 2-12 shows the reference clock configurations. The TVP5146M2 decoder generates the DATACLK
Although the TVP5146M2 decoder is a line-locked system, the color-burst information is used to
determine accurately the color subcarrier frequency and phase. This ensures proper operation with
nonstandard video signals that do not follow exactly the required frequency multiple between color
subcarrier frequency and video line frequency. The frequency control word of the internal color subcarrier
PLL and the subcarrier reset bit are transmitted via terminal 37 (GLCO) for optional use in an end system
(for example, by a video encoder). The frequency control word is a 23-bit binary number. The
instantaneous frequency of the color subcarrier can be calculated from the following equation:
The TVP5146M2 decoder supports the SCART interface used in European audio/video end equipment to
carry composite video, S-Video, and RGB video on the same cable. If composite video and RGB video are
present simultaneously on the video terminals assigned to a SCART interface, the TVP5146M2 decoder
assumes they are pixel synchronous to each other. The timing for both composite video and RGB video is
obtained from the composite source, and its derived clock is used to sample RGB video as well. The
fast-switch input terminal allows switching between these two input video sources on a pixel-by-pixel
basis. The fast switch is a hard switch; there is no alpha blending between both sources.
2.5.2Separate Syncs
VS, HS, and VBLK are independently software programmable to a 1× pixel count. This allows any
possible alignment to the internal pixel count and line count. The default settings for 525-line and 625-line
video outputs are given in Figure 2-14 and Figure 2-15. FID changes at the same transient time when the
trailing edge of vertical sync occurs. The polarity of FID is programmable by an I2C interface.