The Bt868/869 is specifically designed for video systems requiring the generation of
high-quality flicker-free composite and Y/C (S-video) signals from various YCrCb or
RGB digital streams. The Bt868/869 accepts any input format from 640 x 480 to
800 x 600 resolution.
The Bt868/869 uses Conexant’s UltraScale™ technology to provide the most
advanced vertical and horizontal scaling necessary for the display of non-interlaced
data on interlaced devices such as the TV. The UltraScale™ technology converts the
lines of input pixel data to the appropriate number of output lines for producing a
full-screen, high-quality image.
The Bt868/869 performs 5-line vertical filtering, which includes poly phase
interpolation scaling for overscan compensation and flicker filtering. Horizontal
scaling for overscan compensation is achieved by altering the encoder clock
frequency. This approach preserves all of the high frequency components of the
input signals, which are essential for the highest quality display of text intensive
images such as web pages on TVs. The amount of flicker filtering and overscan
compensation is programmable.
Worldwide video standards are supported, including NTSC-M (N. America,
Taiwan, Japan), PAL-B,D,G,H,I (Europe, Asia), PAL-M (Brazil), PAL-N (Uruguay,
Paraguay), and PAL-Nc (Argentina). Bt868 and Bt869 are functionally identical, with
the exception that Bt869 can output Macrovision Level 7.0 anticopy algorithm.
Functional Block Diagram
P[23:0]
HSYNC*
VSYNC*
BLANK*
FIELD
SIC
SID
ALTADDR
RESET*
SLEEP
SLAVE
PAL
XTALIN
XTALOUT
InputFlicker
DEMUX
Timing
Serial
Interface
XTAL
OSC
Color Space
Video
Encoder
PLL
Filter/ScalerConversion
Internal
Reference
Clock
Generation
DAC
MUX
FIFO
BIAS
GEN
FSADJUST
COMP
VREF
DACA
DACB
DACC
VBIAS
CLKO
CLKI
Technology
Distinguishing Features
• Digital RGB or YCrCb non-interlaced
input to interlaced or non-interlaced
analog TV output modes:
Information in this document is provided in connection with Conexant Systems, Inc. (“Conexant”) products. These materials are
provided by Conexant as a service to its customers and may be used for informational purposes only. Conexant assumes no
responsibility for errors or omissions in these materials. Conexant may make changes to specifications and product descriptions at
any time, without notice. Conexant makes no commitment to update the information and shall have no responsibility whatsoever for
conflicts or incompatibilities arising from future changes to its specifications and product descriptions.
° C to +70° CNo Macrovision Feature
° C to +70° C—
No license, express or implied, by estoppel or otherwise, to any intellectual property rights is granted by this document. Except as
provided in Conexant’s Terms and Conditions of Sale for such products, Conexant assumes no liability whatsoever.
THESE MATERIALS ARE PROVIDED “AS IS” WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED, RELATING
TO SALE AND/OR USE OF CONEXANT PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A
PARTICULAR PURPOSE, CONSEQUENTIAL OR INCIDENTAL DAMAGES, MERCHANTABILITY, OR INFRINGEMENT OF ANY
PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT. CONEXANT FURTHER DOES NOT WARRANT THE
ACCURACY OR COMPLETENESS OF THE INFORMATION, TEXT, GRAPHICS OR OTHER ITEMS CONTAINED WITHIN THESE
MATERIALS. CONEXANT SHALL NOT BE LIABLE FOR ANY SPECIAL, INDIRECT, INCIDENTAL, OR CONSEQUENTIAL
DAMAGES, INCLUDING WITHOUT LIMITATION, LOST REVENUES OR LOST PROFITS, WHICH MAY RESULT FROM THE USE
OF THESE MATERIALS.
Conexant products are not intended for use in medical, lifesaving or life sustaining applications. Conexant customers using or selling
Conexant products for use in such applications do so at their own risk and agree to fully indemnify Conexant for any damages
resulting from such improper use or sale.
The following are trademarks of Conexant Systems, Inc.: Conexant™, the Conexant C symbol, and “What’s Next in Communications
Technologies”™. Product names or services listed in this publication are for identification purposes only, and may be trademarks of
third parties. Third-party brands and names are the property of their respective owners.
For additional disclaimer information, please consult Conexant’s Legal Information posted at www.conexant.com, which is
incorporated by reference.
Reader Response: Conexant strives to produce quality documentation and welcomes your feedback. Please send comments and
suggestions to tech.pubs@conexant.com. For technical questions, contact your local Conexant sales office or field applications
engineer.
TEST
BLANK*
FIELD
VSYNC*
HSYNC*
P[23]
P[22]
P[21]
VSS_O
VDD_O
P[20]
P[19]
P[18]
P[17]
P[16]
P[15]
P[14]
VSS_I
VSS
NC
NC
P[0]
P[1]
P[2]
P[3]
P[4]
P[5]
P[6]
P[7]
P[8]
VSS
AGND
100123BConexant1-1
P[9]
P[10]
P[11]
VDD
P[12]
P[13]
VDD_I
Page 10
1.0 Functional DescriptionBt868/Bt869
1.1 Pin DescriptionsFlicker-Free Video Encoder with Ultrascale
Table 1-1. Pin Assignments (1 of 3)
Pin NameI/OPin #Description
XTALINI 63A crystal can be connected to these pins. The pixel clock output (CLKO) is
XTALOUTO62
VDD_X—61Crystal oscillator supply pin. This pin should be tied to the digital supply.
VSS_X—64Crystal oscillator ground pin. This pin should be tied to the digital ground
VAA_PLL—59Analog power for PLL. All VAA and VDD pins must be connected together on
AGND_PLL—58Analog ground for PLL. All AGND and VSS pins must be connected together
CLKOO56Pixel clock output (TTL compatible). This pin is three-state if the CLKI pin
VDD_CO—57Clock output supply pin. This pin should be tied to the digital supply.
VSS_CO—55Clock output ground pin. This pin should be tied to the digital ground plane.
derived from these pins with a PLL. XTALIN can be driven as a CMOS input
pin.
plane.
the same PCB plane to prevent latchup.
on the same PCB plane to prevent latchup.
provides the encoder clock.
TM
Technology
CLKII54Pixel clock input (TTL compatible). This may be used as either the encoder
clock or a delayed version of the CLKO pin synchronized with the pixel data
input.
RESET*I53Reset control input (TTL compatible). A logical 0 resets and disables video
timing (horizontal, vertical, subcarrier counters to the start of VSYNC of first
field) and resets the serial interface registers). RESET* must be a logical 1
for normal operation.
SLEEPI52Power-down control input (TTL compatible). A logical 1 configures the
device for power-down mode. A logical 0 configures the device for normal
operation.
SLAVEI51Slave/master mode select input (TTL compatible). A logical 1 configures the
device for slave video timing operation. A logical 0 configures the device for
master video timing operation.
PALI50PAL/NTSC mode select input (TTL compatible). A logical 1 configures the
device for PAL video format and Mode 1. A logical 0 configures the device
for NTSC video format and Mode 0.
VDDMAXI49Input threshold adjustment. This pin should be tied to VDD for 3.3 V input
swings and GND for 5 V input swings. This pin does not affect the serial
interface pins (SID and SIC).
ALTADDRI48Alternate slave address input (TTL compatible). A logical 0 configures the
device to respond to a serial programming address of 0x88; a logical 1
configures the device to respond to a serial programming address of
(1)
0x8A.
SICI45Serial interface clock input (TTL compatible). The maximum clock rate is
400 kHz.
SIDI/O44Serial interface data input/output (TTL compatible). Data is written to and
read from the device via this serial bus.
VDD_SI—47Serial interface input supply pin. This pin should be tied to the proper supply
voltage for the desired serial interface operating voltage (i.e., tie to 5 V for
5 V serial interface compatibility).
1-2Conexant100123B
Page 11
Bt868/Bt8691.0 Functional Description
Flicker-Free Video Encoder with Ultrascale
Table 1-1. Pin Assignments (2 of 3)
Pin NameI/OPin #Description
VSS_SI—42Serial interface input ground pin. This pin should be tied to the digital
VDD_SO—46Serial interface output supply pin. This pin should be tied to the proper
VSS_SO—43Serial interface output ground pin. This pin should be tied to the digital
TESTI39Test pin. Should be tied to VSS.
BLANK*I/O38Composite blanking control (TTL compatible). This can be generated by the
FIELDO37Field control output (TTL compatible) (Master Mode only three-state in slave
TM
Technology1.1 Pin Descriptions
ground plane.
supply voltage for the desired serial interface operating voltage (i.e., tie to
5 V for 5 V serial interface compatibility).
ground plane.
encoder or supplied from the graphics controller. If internal blanking is used,
this pin can be used to indicate the controller character clock edge.
mode). FIELD transitions after the rising edge of CLK, two clock cycles
following falling HSYNC*. It is a logical 0 during odd fields and is a logical 1
during even fields.
VSYNC*I/O36Vertical sync input/output (TTL compatible). As an output (master mode
operation), VSYNC* is output following the rising edge of CLK. As an input
(slave mode operation), VSYNC* is registered on the rising edge of CLK.
HSYNC*I/O35Horizontal sync input/output (TTL compatible). As an output (master mode
operation), HSYNC* is output following the rising edge of CLK. As an input
(slave mode operation), HSYNC* is registered on the rising edge of CLK.
P[23:0]I32–34, 23–29, 5–18Pixel inputs. See Table 1-2, “Data Pin Assignments for Multiplexed Modes,”
on page 1.05. The input data is sampled on both the rising and falling edge
of CLK for multiplexed modes, and on the rising edge of clock in
non-multiplexed modes. A higher bit index corresponds to a greater bit
significance.
VDD—20,40,60Digital power for core logic. All VAA and VDD pins must be connected
together on the same PCB plane to prevent latchup.
VDD_I—19Digital power for digital inputs. All VAA and VDD pins must be connected
together on the same PCB plane to prevent latchup. This pin should be tied
to the 5 V supply for 5 V tolerant inputs,
VDD_O—30Digital power for digital outputs. All VAA and VDD pins must be connected
together on the same PCB plane to prevent latchup.
VSS—4, 21, 41Digital ground for core logic. All AGND and VSS pins must be connected
together on the same PCB plane to prevent latchup.
VSS_I—22Digital ground for inputs. All AGND and VSS pins must be connected
together on the same PCB plane to prevent latchup.
VSS_O—31Digital ground for outputs. All AGND and VSS pins must be connected
together on the same PCB plane to prevent latchup.
VAA—80Analog power. All VAA and VDD pins must be connected together on the
same PCB plane to prevent latchup.
AGND—1, 79Analog ground. All AGND and VSS pins must be connected together on the
same PCB plane to prevent latchup.
100123BConexant1-3
Page 12
1.0 Functional DescriptionBt868/Bt869
1.1 Pin DescriptionsFlicker-Free Video Encoder with Ultrascale
Table 1-1. Pin Assignments (3 of 3)
Pin NameI/OPin #Description
FSADJUSTI78Full-scale adjust control pin. A resistor (RSET) connected between this pin
and GND controls the full-scale output current on the analog outputs.
VBIASO77DAC bias voltage. A 0.1 µF ceramic capacitor must be used to bypass this
pin to GND. The capacitor must be as close to the device as possible to keep
lead lengths to an absolute minimum.
VREFO76Voltage reference pin. A 0.1 µF ceramic capacitor must be used to decouple
this pin to GND. The decoupling capacitor must be as close to the device as
possible to keep lead lengths to an absolute minimum.
COMPO75Compensation pin. A 0.1 µF ceramic capacitor must be used to bypass this
pin to VAA. The capacitor must be as close to the device as possible to keep
lead lengths to an absolute minimum.
AGND_DAC—74Common DAC Analog ground return. All AGND and VSS pins must be
connected together on the same PCB plane to prevent latchup.
VAA_DACC—73DACC Analog power. All VAA and VDD pins must be connected together on
the same PCB plane to prevent latchup.
TM
Technology
DACCO72DACC output.
VAA_DACB—71DACB Analog power. All VAA and VDD pins must be connected together on
the same PCB plane to prevent latchup.
DACBO70DACB output.
VAA_DACA—69DACA Analog power. All VAA and VDD pins must be connected together on
the same PCB plane to prevent latchup.
DACAO68DACA output.
N/C—65, 66, 67No connect pins
NOTE(S):
(1)
Any unused inputs should not be left floating.
1-4Conexant100123B
Page 13
Bt868/Bt8691.0 Functional Description
Flicker-Free Video Encoder with Ultrascale
Table 1-2. Data Pin Assignments for Multiplexed Modes
Rising Edge of CLKIFalling Edge of CLKI
TM
Technology1.1 Pin Descriptions
Pin
P[11]G4G2Cr/Cb7Cr7R7R4Y7Y7
P[10]G3G1Cr/Cb6Cr6R6R3Y6Y6
P[9]G2G0Cr/Cb5Cr5R5R2Y5Y5
P[8]B7B4Cr/Cb4Cr4R4R1Y4Y4
P[7]B6B3Cr/Cb3Cr3R3R0Y3Y3
P[6]B5B2Cr/Cb2Cr2G7
P[5]B4B1Cr/Cb1Cr1G6G4Y1Y1
P[4]B3B0Cr/Cb0Cr0G5G3Y0Y0
P[3]G0——Cb7R2—— Cb3
P[2]B2——Cb6R1—— Cb2
P[1]B1——Cb5R0—— Cb1
P[0]B0——Cb4G1—— Cb0
NOTE(S):
(1)
G5 is ignored in 15-bit RGB mode.
24-bit
RGB
Mode
15/16-bit
RGB
Mode
16-bit
YCrCb
Mode
24-bit
YCrCb
Mode
24-bit
RGB
Mode
15/16-bit
RGB
Mode
(1)
G5
16-bit
YCrCb
Mode
Y2Y2
24-bit
YCrCb
Mode
Table 1-3. Data Pin Assignments for Non-multiplexed Modes
Pin24-bit RGB Mode24-bit YCrCb Mode
P[23:16]B[7:0]Cb[7:0]
P[15:8]G[7:0]Cr[7:0]
P[7:0]R[7:0]CY[7:0]
100123BConexant1-5
Page 14
1.0 Functional DescriptionBt868/Bt869
1.2 GUI Controller Programmability and Frequency RequirementFlicker-Free Video Encoder with Ultrascale
TM
Technology
1.2 GUI Controller Programmability and
Frequency Requirement
Programmability and frequency requirements for the GUI Controller are defined
in Tabl e 1- 4.
Table 1-4. Programmability and Frequency Requirement
Maximum TotalMaximum
Mode
PixelsLinesLine (kHz)Pixel (MHz)
640 x 480107566511739.86031.563
800 x 600107583514749.45140.000
Vsync to
Active
Maximum Frequencies
1-6Conexant100123B
Page 15
Bt868/Bt8691.0 Functional Description
Flicker-Free Video Encoder with Ultrascale
Figure 1-2 illustrates the concept of flicker filter control.
Figure 1-2. Flicker Filter Control Diagram—External Use
FIFO
YCORING[2:0]
YAT TE N UAT E [2 : 0 ]
TM
Technology 1.2 GUI Controller Programmability and Frequency Requirement
of Range
of Range
of Range
of Range
of Range
of Range
101 = 1/8
010 = 1/64
011 = 1/32
001 = 1/128
000 = Bypass
000 = 1.0 Gain
010 = 7/8 Gain
001 = 15/16 Gain
100 = 1/16
011 = 3/4 Gain
100 = 1/2 Gain
101 = 1/4 Gain
110 = 1/4
111 = Reserved
CCORING[2:0]
110 = 1/8 Gain
111 = 0.0 Gain
CATTENUATE[2:0]
of Range
of Range
of Range
of Range
of Range
of Range
011 = +/- 1/64
000 = Bypass
001 = +/-1/256
010 = +/- 1/128
000 = 1.0 Gain
010 = 7/8 Gain
001 = 15/16 Gain
100 = +/- 1/32
011 = 3/4 Gain
100 = 1/2 Gain
101 = 1/4 Gain
110 = +/- 1/8
101 = +/- 1/16
110 = 1/8 Gain
111 = 0.0 Gain
111 = Reserved
Color
Space
LPF2
LPF3
Horizontal
11 = Luma,
CLPF[1:0]
LPF1
Horizontal
00 = Bypass
01 = Luma,
Luma
Initial
0 = Enable
DIS_YFLPF
Horizontal
10 = Luma,
Filter
Low Pass
Horizontal
Flicker Filter/Scaler
LPF1
Horizontal
00 = Bypass
10 = Chroma,
01 = Chroma,
Horizontal
Horizontal
11 = Chroma,
LPF3
LPF2
Luma
Gamma
Removal
0 = Enable
Anti-Pseudo
DIS_GMUSHY
Luma
Psuedo
0 = Enable
DIS_GMSHYYLPF[1:0]
000 = 5 Line
001 = 2 Line
F_SELY[2:0]
Converter
DIS_GMUSHC
Gamma
Removal
DIS_GMSHC
010 = 3 Line
011 = 4 Line
100 = Alt. 5 Line 1
Gamma
Chroma
Removal
0 = Enable
Anti-Psuedo
Gamma
Psuedo
Chroma
Removal
0 = Enable
000 = 5 Line
001 = 2 Line
010 = 3 Line
101 = Alt. 5 Line 2
110 = Alt. 5 Line 3
F_SELC[2:0]
111 = Alt. 5 Line 4
011 = 4 Line
100 = Alt. 5 Line 1
101 = Alt. 5 Line 2
110 = Alt. 5 Line 3
111 = Alt. 5 Line 4
Input
RGB Mux
RGB Mux
000 = 24-bit
IN_MODE[2:0]
RGB Mux
010 = 15-bit
001 = 16-bit
011 = 24-bit
RGB Non-Mux
101 = 16-bit
YCrCb Mux
100 = 24-bit
YCrCb Mux
111 = 24-bit
110 = Reserved
YCrCb Non-Mux
100123BConexant1-7
Page 16
1.0 Functional DescriptionBt868/Bt869
1.2 GUI Controller Programmability and Frequency RequirementFlicker-Free Video Encoder with Ultrascale
TM
Technology
Figure 1-3 illustrates the Bt868/869 functional block diagram.
Figure 1-3. Encoder Core
COMP
VBIAS
FSADJUST
Reference
Internal Voltage
Y10
DACA
DAC
10
Out
Mode
+
Luma
Delay
CVBS 10
+
10
Out
DACB
DAC
Mux
C10
DACC
DAC
10
10
U/V
2X
Upsample
and
Mixer
Modulator
9
Video
SYNC_AMP
FIELD
Timing
RESET*
Sync
Processor
MY
SID
Control,
Registers
+
X
Closed
Captioning,
Macrovision
Y[9:0]
MCR
2X
Upsample
LPF
4:2:2
to 4:4:4
Conversion
9
X
Burst
Processor
MCB
CRCB[9:0]
RGB 24
BST_AMP
RGB
1-8Conexant100123B
Page 17
Bt868/Bt8691.0 Functional Description
Flicker-Free Video Encoder with Ultrascale
TM
Technology1.3 Circuit Description
1.3 Circuit Description
1.3.1 Overview
The Bt868/869 is a video encoder designed for TV output of non-interlaced
graphics data, such as that found in a PC or some set-top boxes. It incorporates
advanced filtering technology for flicker removal and overscan compensation
which allows high-quality display of non-interlaced images on an interlaced TV
display. The Bt868/869 accomplishes this by minimizing the flicker and providing
control of the amount of overscan so that the entire image is viewable.
The Bt868/869 consists of a Color Space Converter/Flicker Filter engine
followed by a digital video encoder. The Color Space Converter/Flicker Filter
contains the following:
•A timing converter
•Various horizontal video processing functions
•Flicker filter and vertical scaler for overscan compensation
1.3.2 Reset
The output of this engine is fed into a FIFO for synchronization with the
digital video encoder.
If the RESET* pin is held low for a minimum of two clock cycles, a timing reset
and a software reset is performed. During a timing reset, the serial interface is
held in the reset condition, the subcarrier phase is set to zero, and the horizontal
and vertical counters are held to the beginning of VSYNC of Field 1 (both
counters equal to zero). Counting resumes the next clock after rising RESET*.
The serial interface registers are reset to zero.
A software reset, which can be generated by setting the SRESET register bit,
initializes all the serial interface registers to zero (except for PLL_INT, which is
initialized to 0x0C). As a result, all output pins are three-state. The f irst 32
registers are then initialized to auto-configuration mode 0 (see the Auto
Configuration section). The EN_OUT bit must be set to enable the outputs. The
software reset can also be generated by setting the SRESET register bit.
A power-on reset is generated on power-up. The power-on reset generates both
a timing and a software reset. The power-on reset is generated by a time delay
circuit triggered after the supply voltage reaches a value sufficiently high enough
for the circuit to operate. As such, the device may not initialize to the default state
unless the power supply ramp rate is sufficiently fast enough. Therefore, a
hardware reset is recommended if the default state is required.
1.3.3 Timing Registers
After writing any registers, a timing reset is recommended by setting the T-bit.
100123BConexant1-9
Page 18
1.0 Functional DescriptionBt868/Bt869
1.3 Circuit DescriptionFlicker-Free Video Encoder with Ultrascale
1.3.4 Device Initialization
After a reset condition, the device must be programmed through the serial
interface to activate video or output one of the other video standards and to enable
the CLK0, HSYNC*, VSYNC*, and FIELD outputs.
1.3.5 Auto Configuration
The device can configure itself for one of eight combinations of video formats
and input modes with a single register write. Tab le s 1- 5 and 1-6 detail the eight
available auto configuration modes. This feature reduces the software support
required, yet allows full flexibility in generating video formats and timing. Once
the device is configured, all the registers are accessible to modify the modes. For
less common modes, the device can be configured for the closest mode, and only
those registers that differ need to be programmed. To auto-configure the device,
set the configuration bits (CONFIG[2:0]) to the desired mode. The device will
initialize the first 32 registers (registers 0x3B to 0x5A), setting the BUSY flag in
the process. When complete, the BUSY flag is cleared. The serial interface is not
available when the BUSY flag is high except for monitoring the status register.
If the mux mode is enabled, pins P[23:21] can also be used to externally
configure the device to any one of the eight configuration modes. These pins
directly emulate the CONFIG[2:0] register. In order to configure the device in
this way, the EN_PINCFG register must be set. The desired state must be present
on the P[23:21] pins for at least two clock cycles.
TM
Technology
1-10Conexant100123B
Page 19
Bt868/Bt8691.0 Functional Description
Flicker-Free Video Encoder with Ultrascale
Table 1-5. Auto-Configuration Modes 0–3—RGB Input (1 of 2)
TM
Technology1.3 Circuit Description
Mode 0
NTSC 640x480
Register Name
H_CLKO [11:0]1792700188876024649A02304900
H_ACTIVE [9:0]640280640280800320800320
HSYNC_WIDTH
[7:0]
HBURST_BEGIN
[7:0]
HBURST_END
[7:0]
H_BLANKO [10:0]38117D4491C159725552520D
V_BLANKO [9:0]3422462E32204129
V_ACTIVEO [8:0]212D4242F2216D8252FC
H_FRACT [7:0]0000000 0
H_CLKI [10:0]7843109443B08803709603C0
H_BLANKI [8:0]1267E26610A66421408C
V_BLANK_DLY00000000
V_LINESI [9:0]6002586252717352DF7502EE
V_BLANKI [7:0]754B905A8656955F
CLKO=28.195793 MHz
DECHEXDECHEXDECHEXDECHEX
132841388A182B6170AA
15096166A6206CE202CA
966010468180841549A
(1
)
Mode 1
PAL 640x480
CLKO=29.500008 MHz
Mode 2
NTSC 800x600
CLKO=38.769241 MHz
Mode 3
PAL 800x600
CLKO=36.000000 MHz
V_ACTIVEI [9:0]4801E04801E0600258600258
CLPF [1:0]0000000 0
YLPF [1:0]3333333 3
V_SCALE [13:0]526614924096100073731CCD57341666
PLL_FRACT [15:0]34830880E72821C72151243B1400
EN_XCLK00000000
BY_PLL00000000
PLL_INT [5:0]12C13D17111610
EN_SCART00000000
ECLIP00000000
PAL0011001 1
DIS_SCRESET00000000
VSYNC_DUR11001100
625LINE00110011
SETUP1100110 0
100123BConexant1-11
Page 20
1.0 Functional DescriptionBt868/Bt869
1.3 Circuit DescriptionFlicker-Free Video Encoder with Ultrascale
Table 1-5. Auto-Configuration Modes 0–3—RGB Input (2 of 2)
There are two timing generators that control the operation of the encoder. The
encoder timing block generates the signals for the proper encoding of the video
into NTSC or PAL, and extracts the processed input pixels from the internal
FIFO. The encoding timing generator can receive its clock from either an external
crystal oscillator and PLL, or from the CLKI pin. Normal operation requires that
the encoding clock be generated by the PLL. The clock source is selected by the
EN_XCLK register bit. If EN_XCLK is set to a logical 0, the internal clock
source is selected; and when the EN_OUT bit is set, the CLKO pin is enabled to
drive the derived clock.
A crystal must be present between XTALIN and XTALOUT pins if the
internal clock source is selected. The frequency of the CLK is synthesized by a
PLL such that the frequency is:
= F
F
clk
100123BConexant1-13
{PLL_INT(5:0) + [PLL_FRACT(15:0)/216]}/6
xtal *
Page 22
1.0 Functional DescriptionBt868/Bt869
TM
Technology
1.3 Circuit DescriptionFlicker-Free Video Encoder with Ultrascale
The crystal must be chosen so that the precise line rate for the video standards
required can be achieved. This is done to maintain the subcarrier relationship to
the line rate and thereby achieve the precise subcarrier frequency as required by
the standard. The crystal oscillator is designed to oscillate from 5-25 MHz. A
13.5 MHz crystal meets the requirements for both NTSC and PAL video
standards. The crystal must be within 50 ppm of the maximum desired clock rate
for NTSC operation, and 25 ppm for PAL operation, across temperature (0° to
70°C). See Appendix B for list of recommended crystal vendors.
The crystal oscillator is disabled by the SLEEP pin. Sufficient time (greater
than approximately 1 second) must be allowed after coming out of sleep mode to
allow the oscillator to stabilize.
If the external clock source is selected (EN_XCLK=1), a clock signal of the
desired pixel clock rate must be present at the CLKI pin. The CLKO pin will be
three-state, and the crystal oscillator disabled. The clock must meet the same
requirements as above. It is highly recommended that the internal clock be used in
order to ensure that the output video remain within the specifications defined by
the relevant video standard. Any aberration in the source clock is reflected in the
output video and detracts from the quality of the image.
The BY_PLL bit will bypass the PLL, and the encoder clock will be at the
crystal frequency. This bit will take precedence over the EN_XCLK bit.
The second timing generator controls the generation of the HSYNC*,
VSYNC*, BLANK*, and pixel input clocking. This is normally the same clock as
the encoding clock. The EN_ASYNC register bit, if set, will allow this clock to be
driven directly by the CLKI pin. If the DIV2 register bit is set, this internal clock
is divided by two before driving the second timing generator. This is required for
interlaced input to interlaced output mode (i.e., CCIR601 applications).
The CLKI pin is the clock used for synchronizing the pixel inputs (P[23:0])
and any timing input signals (HSYNC*, VSYNC*, and BLANK*) and normally
must be a delayed version of the CLKO pin. It can be directly connected to CLKO
if desired. Data is registered with this input and re-synchronized to the internal
clock. Normally, in muxed input mode, both edges of the CLKI input are used. If
the MODE2X register bit is set, the internal clock is divided by two, allowing a 2x
external clock, and the data to be provided on the rising edge only.
1.3.7 Master and Slave Modes
The device can operate as either a timing master or a slave. In master mode, the
device will generate and output HSYNC*, VSYNC*, and BLANK*. In slave
mode, these must be provided externally. The desired mode is selected by the
SLAVE pin and SLAVER bit.
It is highly recommended that the device operate as a master, to ensure that
the input and output video streams remain synchronized. If the device supplying
the HSYNC* and VSYNC* inputs in slave mode is not correctly programmed, or
the timing varies from that which is required, the output image will lose lock with
the input. By running the device in master mode, any timing errors that occur can
be absorbed to some extent by the on-board FIFO.
1-14Conexant100123B
Page 23
Bt868/Bt8691.0 Functional Description
Flicker-Free Video Encoder with Ultrascale
1.3.8 Input Formats
The device can convert a wide range of input formats to television video formats.
The input format can be either non-interlaced computer video in 640 x 480 or
800 x 600 formats, or interlaced formats such as CCIR601 formats as well as
most other formats which might be encountered. For detailed information on the
CCIR601 mode, please refer to the “DVD Movie Playback Architecture and
Solutions Application Note”. This application note may be obtained from your
local Conexant Semiconductor sales office.
1.3.9 Pixel Input Timing
The device can accept the input in data either RGB or YCrCb color spaces. Data
can be input either a full pixel at a time, clocked in on the rising edge of CLKI, or
in various multiplexed modes, using both edges of CLKI.
In YCrCb mode, either 24-bit 4:4:4 data or 16-bit 4:2:2 data can be input. In
RGB mode, either 15 bit 5:5:5, 16 bit 5:6:5, or 24-bit RGB can be input. In 16-bit
4:2:2 YCrCb input mode, multiplexed Y, Cr, and Cb data is input through the
P[11:4] inputs. The Y data is input on the falling edge of CLK. The Cr/Cb data is
input on the rising edge of CLK. The Cb/Y/Cr/Y sequence begins at the first
active pixel. In 24-bit 4:4:4 YCrCb input mode, multiplexed Y, Cr, and Cb data is
input through the P[11:0] inputs. The input data is sampled on both the rising and
falling edge of CLK. In RGB input mode, input data is sampled as 12 bits in
24-bit RGB mode or 8 bits in 15/16 bit RGB mode on both the rising and falling
edge of CLK. Tab le 1 -2 shows the assignments of input P[11:0] data on rising
edge and falling edge of CLK.
In addition, all 24-bit modes can utilize a non-multiplexed mode. See Tabl e
1-3 on page 1-5.
TM
Technology1.3 Circuit Description
1.3.10 Output Modes
The encoder can generate the video as Composite/Y-C, as YUV component, or as
VGA-style RGB. These modes are selected by the OUT_MODE[1:0] register
bits.
When outputting RGB, the device will output VGA/SVGA analog RGB. In
this mode, the R, G, and B input data is fed to the DACs after the addition of sync
and, if the SETUP bit is set, setup. The output currents are scaled so that the
DACs output the proper 1 V full-scale levels for driving a monitor. The graphics
controller provides all the timing control for the monitor, and the device operates
as a slave. Only the P[23:0], BLANK*, HSYNC*, and VSYNC* input pins and
the RGB analog output pins are active. The BLANK*, HSYNC*, and VSYNC*
pins are automatically enabled as inputs in this mode.
Each of the three video signals generated by the OUT_MODE bits can be
multiplexed to any DAC using the OUT_MUXA[1:0], OUT_MUXB[1:0], and
OUT_MUXC[1:0] register bits.
100123BConexant1-15
Page 24
1.0 Functional DescriptionBt868/Bt869
1.3 Circuit DescriptionFlicker-Free Video Encoder with Ultrascale
TM
Technology
1.3.11 YCrCb Inputs
Y has a nominal range of 16–235; Cb and Cr have a nominal range of 16–240,
with 128 equal to zero. Values of 0 and 255 are interpreted as 1 and 254,
respectively. Y values of 1–15 and 236–254, and CrCb values of 1–15 and
241–254, are interpreted as valid linear values.
Figure 1-4 shows the frequency response of the sub-sampling process. If 4:4:4
data is input, it is sub-sampled to 4:2:2 prior to overscan compensation and flicker
filtering.
Figure 1-4. Decimation Filter at Fs=27 MHz
Chroma Decimation Filter
Decibels (dB)
Freq (Fs=27MHz
The resulting 4:2:2 output must then be converted to YUV values and then
scaled for the output range of the DACs. The MY, MCR, and MCB registers must
be programmed to perform this conversion. The scaling equations are as follows:
MY = (int) [V100/(219.0
MCR = (int)[(128.0/127.0)
0.5]
MCB = (int)[(128.0/127.0)
where:V
= Full scale output voltage (1.28 V)
V
FS
SINX = SIN (2
= 100% white voltage (0.661 V for NTSC, 0.7 V for PAL)
With IN_MODE set to 24, 16, or 15-bit RGB mode, digital, gamma-corrected
RGB data with a 0-255 range is input via the P[11:0] inputs in 24-bit RGB mode
or P[11:4] inputs in 15/16-bit RGB mode on both the rising and falling edge of
CLK. The RGB data is converted to Y/R-Y/B-Y as follows:
Y[9:0] = [INT(.299
G[7:0] + INT(.114
The Y/R-Y/B-Y values are then sub-sampled to 4:2:2 data prior to overscan
compensation and flicker filtering.
The resulting 4:2:2 output must then be converted to YUV values and then
scaled for the output range of the DACs. The MY, MCR, and MCB registers must
be programmed to perform this conversion. The scaling equations are as follows:
MY = (int)[V100/(255
MCR = (int)[(128.0/127.0)
MCB = (int)[(128.0/127.0)
where:V
V
SINX = SIN (2
100
= Full scale output voltage (1.28 V)
FS
1.3.13 Video Amplitude Scaling
Both the luminance and chrominance video amplitudes can be scaled by the
MCR, MCB, and MY registers. This allows various colormetry standards to be
achieved, and can also be used to boost the chroma to compensate for the sinX/X
loss of the DACs. Tables 1-7 and 1-8 show the range of values achievable and
values for various video formats.
210) * R[7:0] + INT(.587 * 210) *
*
210)*B[7:0] = 27] * 2–8, 0 to 1024
*
VFS)*26 + 0.5]
*
V100 * 0.877/(127 * VFS * sinx) * 25 + 0.5]
*
V100 * 0.493/(127 * VFS * sinx) * 25 + 0.5]
*
= 100% white voltage (0.661 V for NTSC, 0.7 V for PAL)
πF
SC/FCLK
)/(2πFSC/F
CLK
)
Table 1-7. Video Modes
ModeNTSCNTSC-JapanPAL-BDGHIPAL-NPAL-NcPAL-MPAL-60
VSYNC_DUR1101011
625LINE0011100
SETUP1001010
PAL0011111
100123BConexant1-17
Page 26
1.0 Functional DescriptionBt868/Bt869
1.3 Circuit DescriptionFlicker-Free Video Encoder with Ultrascale
TM
Technology
Table 1-8. Video Levels
ModeRangeN T S C
100% White Amp (V)0.6610.7140.70.6610.70.6610.7
Sync Amp (V)0.2860.2860.3010.2860.3010.2860.301
Subcarrier Amp (V)0.2860.2860.30.30.30.3060.306
YCrCb InputMY0–255153158158153158153158
MCR0-255187207207187207187207
MCB0-255133149149133149133149
RGB InputMY0–255133143141133141133141
MCR0-255117127124117124117124
MCB0-25566717066706670
SYNC_AMP0–255225225238225238225238
BST_AMP0–2551141149090909292
NTSC-
Japan
PALBDGHI PAL-N PAL-Nc PAL-M PAL-60
1.3.14 Input Pixel Horizontal Sync
The HSYNC* pin provides the pixel synchronization for the pixel input data. It is
an output in master mode, and an input in slave mode. In master mode, it is a
pulse two CLK cycles in duration whose leading edge indicates the beginning of a
new line of pixel data. The period of the pulses is H_CLKI CLK cycles. The first
pixel should be presented to the device H_BLANKI minus the internal pipelined
clock (in CLK cycles) after a leading edge of HSYNC*. The next H_ACTIVE
pixels will be accepted as active pixels and used in the construction of the output
video. In slave mode, the period must be exactly the number of clocks required
for the desired overscan mode. Only the leading edge is used, and the high and
low times must be at least two CLK cycles in duration. HSYNC* is clocked by
the rising edge of CLKI. HSYNCI is clocked by the rising edge of CLKI.
The polarity of the HSYNC* pin can be programmed by the HSYNCI register
bit. The default convention is active low.
1-18Conexant100123B
Page 27
Bt868/Bt8691.0 Functional Description
Flicker-Free Video Encoder with Ultrascale
TM
Technology1.3 Circuit Description
1.3.15 Input Pixel Vertical Sync
The VSYNC* pin provides the line synchronization for the pixel input data. It is
an output in master mode, and an input in slave mode.
For non-interlaced input timing in master mode, VSYNC* is a pulse one
horizontal line time in duration whose leading edge indicates the beginning of a
frame of input pixel data. The leading edge coincides with the leading edge of
HSYNC*. The period of the pulses is V_LINESI horizontal lines. The first line of
data should be presented to the device V_BLANKI lines after the leading edge of
VSYNC*. The next V_ACTIVEI lines are accepted as active lines and used in the
construction of the output video. In slave mode, the period must be exactly the
frame rate of the desired video format. Only the leading edge is used, and the high
and low duration must be at least two CLK cycles. The beginning of the frame of
data is indicated by the next leading edge of HSYNC* coincident with or after the
leading edge of VSYNC*.
For interlaced input timing, only slave mode is supported. The period must be
exactly the frame rate of the desired video format. If the leading edge of
HSYNC* and VSYNC* are coincident, which indicates the input is in odd field,
the internal line counter is reset to line 1 at the leading edge of VSYNC*. If the
leading edges of HSYNC* and VSYNC* are not coincident, which indicates the
input is in even field, the internal line counter will be reset to line 2 at the
beginning of the next line. Only the leading edge of VSYNC* is used, and the
high and low duration must be at least two CLK cycles. VSYNC* is clocked by
the rising edge of CLKI.
The polarity of the VSYNC* output can be programmed by the VSYNCI
register bit. The default convention is active low.
1.3.16 Input Pixel Blanking
The input pixel blanking can be controlled by either the BLANK* pin or by the
internal registers. It can be programmed independently of master/slave mode
using the EN_BLANKO register bit. In output mode (EN_BLANKO=1), the
pixel blanking is generated based on the active area defined by the H_BLANKI,
H_ACTIVE, V_BLANKI, and V_ACTIVEI registers, and the BLANK* pin will
be output in the proper relationship to the syncs to indicate the active pixels. In
input mode (EN_BLANKO=0), when the BLANK* pin goes high, it will indicate
start of active pixels at the pixel input pins. The duration of active pixel is still
determined by the H_ACTIVE register. BLANK* is clocked by the rising edge of
CLKI.
An additional function for the BLANK* pin is used if the EN_DOT register
bit is set. In this mode, the internally-generated blanking is used. The BLANK*
pin becomes an input whose rising edge defines the graphics controller character
clock boundary. This is used internally by the encoder to keep track of the exact
pixel count for controllers that cannot operate at pixel clock rates but instead
operate at VGA character clock rates.
100123BConexant1-19
Page 28
1.0 Functional DescriptionBt868/Bt869
1.3 Circuit DescriptionFlicker-Free Video Encoder with Ultrascale
TM
Technology
1.3.17 Overscan Compensation and Flicker Filtering
The resulting subsampled and optionally color-space-converted pixel data is
processed by the overscan compensation and flicker filtering logic. This process
converts the lines of input pixel data to the appropriate number of output lines for
producing a full-screen image on the television receiver. The image, which is
100% within the viewable area of the screen (overscan compensated), can
perform vertical filtering to reduce the effects of picture flicker due to the
interlacing of the output image. The amount of flicker filtering is programmable,
because this process trades off vertical resolution in order to reduce the flicker,
and allows the process to be optimized for the image. Horizontal scaling is
achieved by adjusting the encoder clock rate. No additional horizontal processing
is performed on the input pixels. This allows the full bandwidth of the input
image to be output, limited only by the 2x upsampling filter response, which is
nominally greater than 6 MHz.
The device can accept a wide variety of input image formats, from 640x480 to
800x600, and can output all NTSC and PAL video formats.
Figures A-1 through A-4 in Appendix A show the possible ranges of overscan
compensation for 640x480 and 800x600 NTSC and PAL formats, for graphics
controllers with synchronization resolutions of 1, 8, and 9 pixel clocks, using a
horizontal blanking interval of 20 pixel clocks. Tables A- 3 through A-10 show
representative values for the following:
•Input picture and frame and output picture and field sizes for 640x480 and
800x600 input picture size
•NTSC and PAL outputs using a horizontal blanking interval of 2.5
The DIS_FFILT register bit disables the flicker filter. The vertical scaling
should also be disabled by setting the VSCALE register to 4096 for
non-interlaced input, or 0 for interlaced input.
CONFIG[2:0]This field determines the configuration for the
automatic configuration process.
000 = NTSC 640 x 480 RGB input
001 = PAL 640 x 480 RGB input
010 = NTSC 800 x 600 RGB input
011 = PAL 800 x 600 RGB input
100 = NTSC 640 x 480 YCrCbYCrCb input
101 = PAL 640 x 480 YCrCb input
110 = NTSC 800 x 600 YCrCb input
111 = PAL 800 x 600 YCrCb input
LUMADLY[1:0] This 2-bit value can be used to program the luminance
delay in pixels for the CVBS_DLY and Y_DLY
output modes.
00 = no delay
01 = 1 pixel
10 = 2 pixels
11 = 3 pixels
µs.
1-20Conexant100123B
Page 29
Bt868/Bt8691.0 Functional Description
Flicker-Free Video Encoder with Ultrascale
1.3.18 VGA Compatibility
To achieve VGA compatibility, the controller must manipulate the VGA register
settings in order to produce a consistent output timing for all VGA modes. The
encoder has no way of knowing that a different VGA mode has been selected, and
therefore cannot make any adjustments to the timing. The extent of VGA
compatibility is entirely the controller’s responsibility.
1.3.19 Analog Horizontal Sync
The duration of the horizontal sync pulse is determined by the horizontal sync
width register (HSYNC_WIDTH[7:0]). The beginning of the horizontal sync
pulse corresponds to the reset of the internal horizontal pixel counter. The
horizontal line rate is determined by H_CLKI[11:0]. The internal horizontal
counter is reset to 1 at the beginning of the horizontal sync and counts up to
H_CLKI.
The sync rise and fall times are automatically controlled. The sync amplitude
is programmable over a range of values by SYNC_AMP[7:0]. Tab le 1 -8 lists the
range of sync values obtainable and the preferred values for various video
formats.
TM
Technology1.3 Circuit Description
1.3.20 Analog Vertical Sync
The duration of the vertical sync is selectable as either 2.5 or 3 lines by register bit
VSYNC_DUR. If VSYNC_DUR = 0, 3 lines are selected; if VSYNC_DUR = 1,
2.5 lines are selected. The duration of the serration and equalization pulses are 1/2
the duration of the horizontal sync duration.
1.3.21 Analog Video Blanking
Analog video blanking is controlled by the H_BLANKO, V_BLANKO, and
V_ACTIVEO registers. Together they define an active region where pixels will be
displayed. V_BLANKO defines the number of lines from the leading edge of the
analog vertical sync to the first active output lines, per field; V_ACTIVEO
defines the number of active output lines. H_BLANKO defines the number of
output pixels from the leading edge of horizontal sync to the first active output
pixel; H_BLANKO defines the number of active output pixels.
The device will automatically blank the video from the start of the horizontal
sync interval through the end of the burst, as well as the vertical sync to prevent
erroneous video timing generation.
1.3.22 Video Standards
There are several bits (625LINE, SETUP, and VSYNC_DUR) and a PAL pin that
control the generation of various video standards. (These are summarized in
Table A -1 .) They allow the generation of all the NTSC and PAL video standards.
These bits control the specific encoding process parameter, and other registers
may also need to be modified to meet all the video parameters of the particular
video standard. Video timing diagrams are illustrated in Figures 1-5 through 1-13,
which summarize all the common video standards and the required register values
for typical input formats.
100123BConexant1-21
Page 30
1.0 Functional DescriptionBt868/Bt869
1.3 Circuit DescriptionFlicker-Free Video Encoder with Ultrascale
Figure 1-5. Interlaced 525-Line (NTSC) Video Timing
RESET*
Start
Analog
FIELD 1
of
VSYNC
TM
Technology
525524523
Analog
FIELD 2
Analog
FIELD 3
525524523
3
21
BURST PHASE
321
BURST PHASE
89
272271270269268267263262261266265264
89
10
10
227654
285
227654
Analog
FIELD 4
272271270269268267263262261266265264285
Burst Begins with Positive Half-Cycle
Burst Phase = Reference Phase = 180° Relative to B–Y
Burst Begins with Negative Half-Cycle
Burst Phase = Reference Phase = 180° Relative to B–Y
NOTE(S): SMPTE line numbering convention is used rather than CCIR624.
1-22Conexant100123B
Page 31
Bt868/Bt8691.0 Functional Description
Flicker-Free Video Encoder with Ultrascale
Figure 1-6. Interlaced 525-Line (PAL-M) Video Timing
Analog
FIELD 1
TM
Technology1.3 Circuit Description
RESET*
Start
of
VSYNC*
525524523
Analog
FIELD 2
Analog
FIELD 3
525524523
Analog
FIELD 4
3
21
Burst Phase
321
Burst Phase
7654
89
272271270269268267263262261266265264
89
272271270269268267263262261266265264
10
10
273
273
11
11
12
274
12
274
227654
285
22
285
Burst Phase = Reference Phase = 135° Relative to U
PAL Switch = 0, +V Component
Burst Phase = Reference Phase + 90° = 225° Relative to U
PAL Switch = 1, –V Component
100123BConexant1-23
Page 32
1.0 Functional DescriptionBt868/Bt869
1.3 Circuit DescriptionFlicker-Free Video Encoder with Ultrascale
Figure 1-7. Interlaced 625-Line (PAL–B, D, G, H, I, Nc) Video Timing (Fields 1–4)
RESET*
Start
of
VSYNC
Analog
FIELD 1
TM
Technology
Burst
Blanking
Intervals
310313
309318310
62562462323
21622621620
– U PHASE
Analog
FIELD 2
Analog
FIELD 3
Analog
FIELD 4
319
FIELD One
FIELD Two
FIELD Three
2476543
337320318317316315314309308312311336
24765432162262162062562462323
337320317316315314308313312311336319
FIELD Four
Burst Phase = Reference Phase = 135° Relative to U
PAL Switch = 0, +V Component
Burst Phase = Reference Phase + 90° = 225° Relative to U
PAL Switch = 1, –V Component
1-24Conexant100123B
Page 33
Bt868/Bt8691.0 Functional Description
Flicker-Free Video Encoder with Ultrascale
Figure 1-8. Interlaced 625-Line (PAL–B, D, G, H, I, Nc) Video Timing (Fields 5–8)
TM
Technology1.3 Circuit Description
RESET*
Start
of
VSYNC
Analog
FIELD 5
Burst
Blanking
Intervals
310313
309318310
62562462323
21622621620
– U PHASE
Analog
FIELD 6
Analog
FIELD 7
Analog
FIELD 8
FIELD Five
FIELD Six
FIELD Seven
2476543
337320318317316315314309308312311336319
24765432162262162062562462323
337320317316315314308313312311336319
FIELD Eight
Burst Phase = Reference Phase = 135° Relative to U
PAL Switch = 0, +V Component
Burst Phase = Reference Phase + 90° = 225° Relative to U
PAL Switch = 1, –V Component
100123BConexant1-25
Page 34
1.0 Functional DescriptionBt868/Bt869
1.3 Circuit DescriptionFlicker-Free Video Encoder with Ultrascale
Figure 1-9. Interlaced 625-Line (PAL–N) Video Timing (Fields 1–4)
VSYNC*
Analog
FIELD 1
RESET*
– U PHASE
Analog
FIELD 2
TM
Technology
24765432162262162062562462323
Burst
Blanking
Intervals
310313319
Analog
FIELD 3
622
Analog
FIELD 4
310309
FIELD One
FIELD Two
FIELD Three
FIELD Four
Burst Phase = Reference Phase = 135° Relative to U
PAL Switch = 0, +V Component
Burst Phase = Reference Phase + 90° = 225° Relative to U
PAL Switch = 1, –V Component
318319
337320318317316315314309308312311336
6
2475432162162062562462323
337320317316315314308313312311336
1-26Conexant100123B
Page 35
Bt868/Bt8691.0 Functional Description
Flicker-Free Video Encoder with Ultrascale
Figure 1-10. Interlaced 625-Line (PAL–N) Video Timing (Fields 5–8)
TM
Technology1.3 Circuit Description
VSYNC*
Analog
FIELD 5
Burst
Blanking
Intervals
310313
309318310
2162262162062562462323
– U PHASE
Analog
FIELD 6
Analog
FIELD 7
Analog
FIELD 8
FIELD Five
FIELD Six
FIELD Seven
2476543
337320318317316315314309308312311336319
24765432162262162062562462323
337320317316315314308313312311336319
FIELD Eight
Burst Phase = Reference Phase = 135° Relative to U
PAL Switch = 0, +V Component
Burst Phase = Reference Phase + 90° = 225° Relative to U
PAL Switch = 1, –V Component
100123BConexant1-27
Page 36
1.0 Functional DescriptionBt868/Bt869
1.3 Circuit DescriptionFlicker-Free Video Encoder with Ultrascale
Figure 1-11. Noninterlaced 262-Line (NTSC) Video Timing
START
of
VSYNC
TM
Technology
21262261
FIELD 1
START
of
VSYNC
263524523268267266265264272271270269
FIELD 2
Burst Begins with Positive Half-Cycle
Burst Phase = Reference Phase = 180° Relative to B-Y
Burst Begins with Negative Half-Cycle
Burst Phase = Reference Phase = 180° Relative to B-Y
Figure 1-12. Noninterlaced 262-Line (PAL–M) Video Timing
START
of
VSYNC
10219876543
285
21525524
FIELD 1
START
of
VSYNC
266265264263262274
FIELD 2
Burst Begins with Positive Half-Cycle
Burst Phase = Reference Phase = 180° Relative to B-Y
Burst Begins with Negative Half-Cycle
Burst Phase = Reference Phase = 180° Relative to B-Y
1-28Conexant100123B
10219876543
273
11
12
285272271270269268267
Page 37
Bt868/Bt8691.0 Functional Description
Start
of
VSYNC
TM
Technology1.3 Circuit Description
RESET*
Flicker-Free Video Encoder with Ultrascale
Figure 1-13. Noninterlaced 312-Line (PAL–B, D, G, H, I, N, Nc) Video Timing
Burst Phase = Reference Phase = 135° Relative to U
PAL Switch = 0, +V Component
Burst Phase = Reference Phase + 90° = 225° Relative to U
PAL Switch = 1, –V Component
1.3.23 Subcarrier Generation
The device uses a 32-bit-word to synthesize the subcarrier. The value of the
subcarrier increment required to generate the desired subcarrier frequency is
found with the following equation:
MSC[31:0] = (int) (2
or more directly, for NTSC:
MSC[31:0] = 2
and for PAL:
MSC[31:0] = 2
23 4
32
F
/ F
sc
*
32
* [455 * / (2 * H_CLKO)]
32
* [(1135/4 + 1/625) / (H_CLKO)]
+ 0.5)
clk
237651312311310309308
24
23765131231131030930823 424
where F
subcarrier to enable the generation of any desired video standard. The 32-bit
subcarrier increment MSC[31:0] must be loaded by the serial interface before the
subcarrier can be enabled. The device is reset to disable chroma until the last byte
of the 32-bit increment is loaded, at which time the chroma will be enabled, unless
the DCHROMA bit is set.
(Discrete Time Oscillator) is reset every four fields for NTSC formats and every
eight fields for PAL formats.
100123BConexant1-29
is the encoder clock rate. This allows the generation of any desired
clk
In order to prevent any residual errors from accumulating, the subcarrier DTO
Page 38
1.0 Functional DescriptionBt868/Bt869
1.3 Circuit DescriptionFlicker-Free Video Encoder with Ultrascale
TM
Technology
1.3.24 Burst Generation
The subcarrier burst generation is a function of the video standard (e.g. NTSC or
PAL), the subcarrier frequency increment (MSC[31:03]), and the burst horizontal
begin and end register settings (HBURST_BEGIN[7:0] and
HBURST_END[7:0]). The value of HBURST_BEGIN[7:0] and
HBURST_END[7:0] is the desired pixel minus a value of 128. The burst will
automatically be blanked during the horizontal sync to prevent invalid sync pulses
from being generated. The burst blanking is automatically controlled by the
selected video format. The burst rise and fall times are automatically generated by
the device.
The burst amplitude can be programmed by BST_AMP[5:0]. Tabl e 1-8 shows
the ranges of burst values obtainable and the preferred values for various video
formats.
1.3.25 Chrominance Disable
The chrominance subcarrier can be turned off by setting the DCHROMA bit to a
logical 1. This kills burst as well, providing luminance-only signals on the CVBS
output and a static blank level on the C/R output.
1.3.26 Digital Processing
Once the input data is converted into internal YUV format, the UV components
are low-pass filtered with a filter response illustrated in Figure 1-14 (linearly
scalable by clock frequency). The Y and filtered UV components are upsampled
to CLK frequency by a digital filter whose response is illustrated in Figure 1-15.
Figure 1-14. Three-Stage Chroma Filter
5
0
– 5
– 10
– 15
– 20
Attenuation dB
– 25
– 30
– 35
– 40
0
1-30Conexant100123B
0.511.52
Frequency MHz
CLK = 27 MHz
Page 39
Bt868/Bt8691.0 Functional Description
Flicker-Free Video Encoder with Ultrascale
Figure 1-15. Luminance Upsampling Filter Response
TM
Technology1.3 Circuit Description
5
0
– 5
– 10
– 15
– 20
Attenuation dB
– 25
– 30
– 35
– 40
061224810
Frequency MHz
(Internal Encoder CLK = 27 MHz)
1.3.27 Subcarrier Phasing
In order to maintain correct SC-H phasing, subcarrier phase is set to 0 on the
leading edge of the analog vertical sync every four (NTSC) or eight (PAL) fields,
unless the DIS_SCRESET bit is set to a logical 1. This is true for both interlaced
and non-interlaced outputs. The subcarrier phase can be adjusted from the
nominal 0 phase by the PHASE_OFF[7:0] register, where each LSB change
corresponds to a 360/256 degree change in the phase.
Setting DIS_SCRESET to 1 may be useful in situations where the ratio of
CLK/2 to HSYNC* edges in a color frame is noninteger, which could produce a
significant phase impulse by resetting to 0.
1.3.28 Noninterlaced Operation
When the Bt868/869 is programmed for noninterlaced master mode, it always
displays the odd field. FIELD will change state on the leading edge of the analog
vertical sync. A 30 Hz offset should be subtracted from the color subcarrier
frequency while in NTSC mode so that the color subcarrier phase will be inverted
from field to field. Transition from interlaced to noninterlaced in master mode
occurs during odd fields to prevent synchronization disturbance.
NOTE: Consumer VCRs can record noninterlaced video with minor noise
artifacts, but special effects (e.g., scan > 2x) may not function properly.
100123BConexant1-31
Page 40
1.0 Functional DescriptionBt868/Bt869
1.3 Circuit DescriptionFlicker-Free Video Encoder with Ultrascale
1.3.29 Closed Captioning
The Bt868/869 encodes NTSC/PAL–M closed captioning on scan line 21, and
NTSC/PAL–M extended data services on scan line 284. Four 8-bit registers
(CCF1B1, CCF1B2, CCF2B1, and CCF2B2) provide the data while bits ECCF1
and ECCF2 enable display of the data. A logical 0 corresponds to the blanking
level of 0 IRE, while a logical 1 corresponds to 50 IRE above the blanking level.
Closed captioning for PAL–B, D, G, H, I, N, Nc is similar to that for NTSC.
Closed-caption (CC) encoding is performed for 625-line systems according to the
system proposed by the National Captioning Institute; clock and data timing is
identical to that of NTSC system, except that encoding is provided on lines 22
and 335, for closed captioning and extended data services respectively.
The Bt868/869 generates the clock run-in and appropriate timing
automatically. Pixel inputs are ignored during CC encoding. See FCC Code of
Federal Regulations (CFR) 47 Section 15.119 (10/91 edition or later) for
programming information. EIA608 describes ancillary data applications for Field
2 Line 21 (line 284).
When CCF1B2 is written, CCSTAT_O is set; when CCF2B2 is written,
CCSTAT_E is set. After the CC bytes for the odd field are encoded, CCSTAT_O
is cleared; after the CC bytes for the even field are encoded, CCSTAT_E is
cleared. If the ECCGATE bit is set, no further encoding will be performed until
the appropriate registers are again written; a null will be transmitted on the
appropriate CC line in that case. If the ECCGATE bit is not set, the user must
rewrite the CC registers prior to reaching the CC line; otherwise the last bytes will
be re-encoded. The CC data bytes are double-buffered to prevent loss of data
during the encoding process.
TM
Technology
1.3.30 Internal Color Bars
The Bt868/869 can be configured to generate 100% amplitude, 75% saturation
(100/7.5/75/7.5 for NTSC/PAL-M with setup, 100/0/75/0 for PAL) color bars.
Color bars can be enabled or disabled by setting the ECBAR bit to a logical 1. The
device uses the H_BLANKO register value to determine the starting point of the
color bars, and the H_ACTIVE register value to determine the width. Eight bars
are displayed, with the colors and amplitudes being generated internally. The pixel
inputs are ignored in color bar mode. The MY, MCR, and MCB registers must be
programmed for RGB inputs prior to color bar operation.
1.3.31 Macrovision Encoding
The Bt869 device supports Version 7.xx of the Macrovision specification for
copy protection for all NTSC and PAL modes. The Bt868 does not support the
Macrovision feature.
1-32Conexant100123B
Page 41
Bt868/Bt8691.0 Functional Description
Flicker-Free Video Encoder with Ultrascale
1.3.32 Outputs
There are four modes for the analog outputs, selected by OUT_MODE[1:0]. The
first mode (OUT_MODE=0) generates Composite video (CVBS), Luma (Y),
Chroma (C), and Delayed Luma (Y_DLY). The second mode (OUT_MODE=1)
generates Luma-Delayed Composite video (CVBS_DLY), Luma (Y), Chroma
(C), and Delayed Luma (Y_DLY). The third mode (OUT_MODE=2) generates
Component YUV and Delayed Luma (Y_DLY). The fourth mode
(OUT_MODE=3) generates VGA-style RGB outputs. The LUMADLY[1:0]
register bits control the amount of delay for the delayed luma, from 0 to 3 pixel
clocks. For each mode, any of the four generated outputs can be muxed to any of
three output DACs by the register bits OUT_MUXA[1:0], OUT_MUXB[1:0],
and OUT_MUXC[1:0]. All digital-to-analog converters are designed to drive
standard video levels into a combined RLOAD of 37.5
Ω). Unused outputs should be disabled by setting the corresponding
75
DACDISX bit to minimize supply current, or connected directly to ground to
minimize supply switching currents.
1.3.33 Output Connection Status
TM
Technology1.3 Circuit Description
Ω (doubly-terminated
The device can determine whether or not the DAC output is connected to a
monitor by verifying that the output is doubly-terminated. The MONSTATx bit
for the corresponding DAC is set to a 1 if the device senses a doubly-terminated
load on a reset condition or if the CHECK_STAT register bit is set. While
CHECK_STAT is set, the output is forced to 2/3 of VREF when terminated and
4/3 of VREF if unterminated. The MONSTATx bit reflects the condition when the
DAC output is less than or equal to VREF. The CHECK_STAT bit is
automatically cleared after two clock cycles.
1.3.34 Output Filtering and SINX/X Compensation
The DAC output response is a typical sinx/x response. For the composite video
output, this results in a slightly lower than desired burst and chroma amplitude
value. This can be compensated for, to some extent, by choosing an output filter
which boosts higher frequency response slightly. Another method which can be
used effectively, and is used by default in the auto configuration modes, is to
boost the burst and chroma gain as programmed by the BST_AMP and
MCR/MCB register values by x/sinx. The amount of sinx/x amplitude reduction
is calculated by:
π
sinx/x = sin (
Fsc/Fclk) / (π * Fsc/Fclk)
*
100123BConexant1-33
Page 42
1.0 Functional DescriptionBt868/Bt869
1.3 Circuit DescriptionFlicker-Free Video Encoder with Ultrascale
1.3.35 Power-Down Modes
The device can be placed in a low-power mode by the SLEEP pin. In this mode,
the analog circuitry is shut down, disabling the output video, and the internal
clock to the device is held constant, placing both the analog and digital current
draw to a minimum. Register states are preserved, but other chip functionality
(including the serial interface) is disabled. This mode achieves the greatest
reduction in power.
In addition, the entire analog subsection can be powered-down with the
DACOFF bit, allowing digital operations to continue while reducing the power in
the analog circuitry. This will achieve a significant reduction in power while
maintaining all digital functionality. Each individual DAC can also be powered
down by its corresponding DACDISx bit. This is useful only if some of the DACs
are being used, in order to minimize the power in the system.
1.3.36 Serial Interface
The device includes a 2-wire serial interface which is used for programming the
registers in the device. The interface is designed to operate with either 3.3 V or
5 V input levels by changing the supply voltage for the input and output drivers
with the VDD_SI pin.
TM
Technology
1-34Conexant100123B
Page 43
2.0 Internal Registers
A register bit map is displayed in Tabl e 2-1 , and a read-back bit map is displayed
in Tabl e 2- 2. Bit descriptions and detailed programming information follow the
bit map. All registers are write-only and are set to 0 following a software reset. A
software reset is always performed at power-up; after power-up, a reset can be
triggered by writing the SRESET register bit.
The power-up state is defined to be black burst CCIR601 NTSC video. To enable
active video, the EN_OUT register bit must be set. This bit enables CLKO,
HSYNC*, VSYNC*, BLANK*, and FIELD outputs. If this bit is not set, then
these pins are set to a high impedance.
2.2 Writing Addresses
Following a start condition, writing 0x88 initiates access to subaddresses.
Alternative address 0x8A must be written if the ALTADDR pin is high. If the data
is written in subaddress order, only the beginning subaddress needs to be written;
the internal address counter will automatically increment after each write to a
register.
2.3 Reading Information
Following a start condition, writing 0x89 initiates the read-back sequence, during
which 8 bits of information can be read from the SID pin, MSB first. Alternative
address 0x8B is required if the ALTADDR pin is high. For the case of
ESTATUS[1:0]=00 prior to the read sequence, the first three bits indicate the part
type (Bt868 or Bt869). The lower f ive bits indicate the version number or the
status bits. The instances where ESTATUS[1:0]=01 and ESTATUS[1:0]=10, the
bits read back from the VGA Encoder will contain information as specified by
Table 2 -2.
For software detection of a connected TV monitor on each DAC output,
ESTATUS[1:0] must equal 01 and the MONSTAT x bits should be read
accordingly after writing to CHECK_STAT.
Data details are defined in Ta ble 2 -3 ; bit and register definitions are displayed
in Tabl e 2- 4.
100123BConexant2-3
Page 46
2.0 Internal RegistersBt868/Bt869
2.3 Reading InformationFlicker-Free Video Encoder with Ultrascale
Table 2-2. Read-Back Bit Map
ESTATUS[1:0]76543210
00ID[2:0]VERSION[2:0]
TM
Technology
01
10PLL_LOCKFIFO_OVERFIFO_UNDERPALBUSY
Table 2-3. Data Details Defined
Bit NamesData Definition
ID[2:0]Indicates the part number: 000 is returned from the Bt868; 001 is returned from the Bt869.
VERSION[4:0]Version number; for this revision, these bits are 00001.
MONSTAT_AMonitor connection status for DACA output, 1 denotes monitor connected to DACA.
MONSTAT_BMonitor connection status for DACB output, 1 denotes monitor connected to DACB.
MONSTAT_CMonitor connection status for DACC output, 1 denotes monitor connected to DACC.
CCSTAT_E
CCSTAT_O
FIELD[2:0]Field number, where 000 indicates the first field, 111 indicates the 8th field.
PLL_LOCKHigh when PLL is locked.
MONSTAT_
A
High if closed-caption data has been written for the even field; it is low immediately after the clock run-in on
line 21(NTSC) or 22(PAL).
High if closed-caption data has been written for the odd field; it is low immediately after the clock run-in on
line 284(NTSC) or 335(PAL).
MONSTAT_BMONSTAT_CCCSTAT_ECCSTAT_OFIELD[2:0]
FIFO_OVERSet to one if FIFO overflows. Reset on read.
FIFO_UNDERSet to one if FIFO underflows. Reset on read.
PALIndicates status of PAL pin.
BUSY
Table 2-4. Programming Detail (1 of 7)
Bit/Register NamesBit/Register Definition
H_CLKO[11:0]Number of output CLKs/line
H_ACTIVE[9:0]Number of active input and output pixels
HSYNC_WIDTH[7:0]Analog sync width in clocks
HBURST_BEGIN[7:0]Beginning of burst 50% point in number of clock cycles from analog hsync falling edge
HBURST_END[7:0]End of burst 50% point in number of clock cycles—128 from analog sync falling edge
H_BLANKO[9:0]Number of output CLKs between leading edge of horizontal sync and active video
V_BLANKO[7:0]Line number of first active line (number of blank lines + 1)
V_ACTIVEO[8:0]Number of active output lines/field
Indicates that the device is in the process of initializing the registers and that the registers cannot be written.
This bit remains high for 512 CLK-0 after an auto configuration cycle begins.
2-4Conexant100123B
Page 47
Bt868/Bt8692.0 Internal Registers
Flicker-Free Video Encoder with Ultrascale
Table 2-4. Programming Detail (2 of 7)
Bit/Register NamesBit/Register Definition
H_FRACT[7:0]Fractional number of input clocks per line
H_CLKI[10:0]Number of clocks per line between successive HSYNC* edges
H_BLANKI[8:0]Number of input pixels between HSYNC* leading edge and first active pixel
VBLANKDLYIf set, the effective vertical blanking value in the second field is V_BLANKI+1.
V_LINESI[9:0]Number of vertical input lines
V_BLANKI[7:0]Number of input lines between VSYNC* leading and first active line
Writing a 1 to this bit performs a software reset; all registers are reset to 0s unless the
CONFIG[2:0] field is non-0; in that case, the automatic configuration process is begun and the
BUSY status bit is set. This bit is automatically cleared.
Writing a 1 to this bit checks the status of the monitor connections at the DAC output. This is also
automatically performed on any reset condition, including a software reset. This bit is automatically
cleared.
0 = Normal operation
1 = Disables DAC output current and internal voltage reference. This will limit power consumption
to just the digital circuits.
0 = Normal operation
1 = Disables DAC output. Current is set to 0; output will go to 0 V.
0 = Normal operation
1 = Disables DACB output. Current is set to 0; output will go to 0 V.
DACDISA
CCF2B1[7:0]
CCF2B2[7:0]
CCF1B1[7:0]
CCF1B2[7:0]
ESTATUS[1:0]Serial read-back status bit selection. (See Tabl e 2 -2.)
2-6Conexant100123B
0 = Normal operation
1 = Disables DACA output. Current is set to 0; output will go to 0 V.
This is the first byte of closed-caption information for the even field, line 284 for NTSC or line 335
for PAL. Data is encoded LSB first.
This is the second byte of closed-caption information for the even field, line 284 for NTSC or line
335 for PAL. Data is encoded LSB first.
This is the first byte of closed-caption information for the odd field, line 21 for NTSC or line 22 for
PAL. Data is encoded LSB first.
This is the second byte of closed-caption information for the odd field, line 21 for NTSC or line 22
for PAL. Data is encoded LSB first.
Page 49
Bt868/Bt8692.0 Internal Registers
Flicker-Free Video Encoder with Ultrascale
Table 2-4. Programming Detail (4 of 7)
Bit/Register NamesBit/Register Definition
TM
Technology2.3 Reading Information
ECCF2
ECCF1
ECCGATE
ECBAR
DCHROMA
EN_OUT
EN_BLANKOEnables BLANK* pin as an output.
EN_DOTEnables dot clock synchronization on BLANK* pin.
FIELDI
VSYNCI
HSYNCI
0 = Disables closed-caption encoding on field 2.
1 = Enables closed-caption encoding on field 2.
0 = Disables closed-caption encoding on field 1.
1 = Enables closed-caption encoding on field 1.
0 = Normal closed-caption encoding.
1 = Enables closed-caption encoding constraints. After encoding, future encoding is disabled until
a complete pair of new data bytes is received. This prevents encoding of redundant or
incomplete data.
0 = Normal operation
1 = Enables color bars.
0 = Normal operation
1 = Blank chroma
0 = Three-states all outputs.
1 = Allows outputs to be enabled (depending upon EN_BLANKO register bit and SLAVE pin).
0 = Logical 1 on FIELD indicates an even field.
1 = Logical 1 on FIELD indicates an odd field.
000 = Bypass
001 = 1/128 of range
010 = 1/64 of range
011 = 1/32 of range
100 = 1/16 of range
101 = 1/8 of range
110 = 1/4 of range
111 = Reserved
000 = 1.0 gain (no attenuation)
001 = 15/16 gain
010 = 7/8 gain
011 = 3/4 gain
100 = 1/2 gain
101 = 1/4 gain
110 = 1/8 gain
111 = 0 gain (Force Luma to 0)
001 = 1/128 of range (+/- 1/256 of range)
010 = 1/64 of range (+/- 1/128 of range)
011 = 1/32 of range (+/- 1/64 of range)
100 = 1/16 of range (+/- 1/32 of range)
101 = 1/8 of range (+/- 1/16 of range)
110 = 1/4 of range (+/- 1/8 of range)
111 = Reserved
Page 51
Bt868/Bt8692.0 Internal Registers
Flicker-Free Video Encoder with Ultrascale
Table 2-4. Programming Detail (6 of 7)
Bit/Register NamesBit/Register Definition
CATTENUATE[2:0]Chroma Attenuation:
000 = 1.0 gain (No Attenuation)
001 = 15/16 gain
010 = 7/8 gain
011 = 3/4 gain
100 = 1/2 gain
101 = 1/4 gain
110 = 1/8 gain
111 = 0 gain (Force Chroma to 0)
00 = Output Video[0] on DACA
OUT_MUXA[1:0]
OUT_MUXB[1:0]
OUT_MUXC[1:0]
01 = Output Video[1] on DACA
10 = Output Video[2] on DACA
11 = Output Video[3] on DACA
00 = Output Video[0] on DACB
01 = Output Video[1] on DACB
10 = Output Video[2] on DACB
11 = Output Video[3] on DACB
00 = Output Video[0] on DACC
01 = Output Video[1] on DACC
10 = Output Video[2] on DACC
11 = Output Video[3] on DACC
TM
Technology2.3 Reading Information
CCR_START[8:0]Closed-captioning clock run-in start in clock cycles from leading edge of HSYNC*
CC_ADD[11:0]Closed-captioning DTO increment
DIV2Divides input pixel rate by two (for CCIR601 interlaced timing input.
MODE2XDivides selected input clock by two (allows for 2x rather than double-edge clock input).
EN_ASYNCSet to 0 for normal operation.
00 = Video[0-3] is CVBS/ Y/ C/ Y_DLY
OUT_MODE[1:0]
LUMADLY[1:0]
HSYNOFFSET[9:0]
01 = Video[0-3] is CVBS_DLY/ Y/ C/ Y_DLY
10 = Video[0-3] is V/ Y/ U/ Y_DLY
11 = Video[0-3] is R/ G/ B/ X (VGA mode)
This 2-bit value can be used to program the luminance delay in pixels for the CVBS_DLY and
Y_DLY output modes.
A 2s-complement number. The values range from –512 pixels to +511 pixels. This register manipulates the falling edge position of the digital HSYNC output from Bt868. The default value is 0 and
denotes the standard position of the HSYNC leading edge.
100123BConexant2-9
Page 52
2.0 Internal RegistersBt868/Bt869
2.3 Reading InformationFlicker-Free Video Encoder with Ultrascale
Table 2-4. Programming Detail (7 of 7)
Bit/Register NamesBit/Register Definition
A 2s-complement number. The values range from –HCLKI pixels to +HCLKI pixels. This register
VSYNOFFSET[10:0]
HSYNWIDTH[5:0]
VSYNWIDTH[2:0]
DATDLY
causes the falling edge position of the Bt868’s digital VSYNC output to occur earlier (– value) or
later (+) in time compared to the standard position for NTSC or PAL. The default value is 0 and
denotes the standard position of the VSYNC leading edge.
Controls the duration/width of the digital HSYNC output pulse. Value will be hexadecimal and its
units are in terms of pixels. A value of 0 is a disallowed condition. The acceptable range is 2 pixels
to 3F pixels (=63 decimal). The default value is 2.
Controls the width of the VSYNC output pulse. Denotes the number of lines the VSYNC digital signal remains low on field transitions. Value will be hexadecimal value and its units are in terms of
lines. A value of 0 is a disallowed condition. The acceptable range is 1 line to (2
default value is 1.
Delays the falling edge pixel data by 1 full clock period when the falling edge data precedes the rising edge data. Ensures that the Bt868 moves the falling edge data after the rising edge data. The
correct sequence of rising edge data/falling edge data/rising edge data will then be encoded by
Buteo. The default value for this bit is 0 because most graphics controllers already transmit data in
the expected rising edge data/falling edge data sequence.
TM
Technology
3
–1) lines. The
Swaps the falling edge pixel data with the rising edge pixel data at the input of the pixel port. The
DATSWP
NOTE(S): VSYNWIDTH, HSYNWIDTH, VSYNOFFSET, and HSYNOFFSET are active only when the Bt868/869 is in the master
timing mode when the encoder outputs HSYNC and VSYNC pulses. In the slave timing mode, these registers are ignored.
VSYNCWIDTH and HSYNCWIDTH should never be set to 0.
default value for this bit is 0 which tells the VGA Encoder to expect an order of rising edge data/falling edge data coming from the graphics controller.
2-10Conexant100123B
Page 53
3
3.0 PC Board Considerations
For optimum performance of the Bt868/869, proper CMOS layout techniques
should be studied in the Bt451/457/458 Evaluation Module Operation and Measurements Application Note (AN-16), before PC board layout is begun.
The layout should be optimized for lowest noise on the power and ground
planes by providing good decoupling. The trace length between groups of VAA
and GND pins should be as short as possible to minimize inductive ringing.
A well-designed power distribution network is critical to eliminating digital
switching noise. The ground plane must provide a low-impedance return path for
the digital circuits. A PC board with a minimum of four layers is recommended,
with layers 1 (top) and 4 (bottom) for signals, and layers 2 and 3 for ground and
power, respectively.
3.1 Component Placement
Components should be placed as close as possible to the associated pin in order
for traces to be connected point to point. The optimum layout enables the
Bt868/869 to be located close to the power supply connector and the video output
connector. For an illustration, see Figure 3-1.
3.2 Power and Ground Planes
For optimum performance, a common digital and analog ground plane and a
common digital and analog power plane are recommended. The power plane
should provide power to all Bt868/869 power pins, reference voltage (Vref)
circuitry, and COMP decoupling.
The Bt868/869 power plane should be connected to the graphics system power
plane (VCC) at a single point through a ferrite bead, as illustrated in Figures 3-1
and 3-2. This bead should be located within 3 inches of the Bt868/869. The bead
provides resistance to switching currents, acting as a resistor at high frequencies.
A low-resistance bead should be used, such as Ferroxcube 5659065-3B, Fair-Rite
2723021447, or TDK BF45-4001. See Tabl e 3- 1 for a typical parts list for key
passive components and Figure 3-3 for a schematic diagram of the recommended
layout.
100123BConexant3-1
Page 54
3.0 PC Board ConsiderationsBt868/Bt869
3.2 Power and Ground PlanesFlicker-Free Video Encoder with Ultrascale
Figure 3-1. Power Plane Illustration
Bracket
3V
Bt868
Ferrite Bead
TM
Technology
Analog
Bt decoder
5V
3VAA-Bt868 o
Oscillator
Clocks
PCI or AGP Connector
o VCC3
o VCC3
Data
3 V
4 layer board
plane order:
TOP
BOTTOM
Signals
GND
PWR
Signals
3-2Conexant100123B
Page 55
Bt868/Bt8693.0 PC Board Considerations
Flicker-Free Video Encoder with Ultrascale
Figure 3-2. Connection Diagram for Output Filters and Other Key Passive Components
TM
Technology3.2 Power and Ground Planes
Bt868/869
FSADJUST
1.8 µH
270 pF
VAA
COMP
VBIAS
VREF
GND
DACA
DACB
DACC
LPF
22 pF
C7
C8
DAC Output
330 pF
C9
RSET =
100 Ω,1%
VAA
P
GND
Bt868/869 Power Plane
C2–C6
Ω,75 Ω,
75
1%
Schottky Diodes
To Filter
Schottky Diodes
22 pF
1.8 µH
270 pF
1%
+
75 Ω,
1%
RF Mod/CVBS
P
P
RF Modulator/CVBS Out
75
330 pF
82
C10
LPF
LPF
L1
Buffer
TRAP
C1
RF
Modulator
ZIN = 1 K
Audio
+3.3 V (VCC)
Ground
(Power Supply
Connector)
To V i de o
Connector
CVBS
P
RF
(1)
(1)
Some modulators may require AC coupling capacitors (10 µF).
Table 3-1. Typical Parts List for Key Passive Components
For optimum performance, all capacitors should be located as close as possible to
the device, and the shortest possible leads (consistent with reliable operation)
should be used to reduce the lead inductance. Chip capacitors are recommended
for minimum lead inductance. Radial lead ceramic capacitors can be substituted
for chip capacitors and are better than axial lead capacitors for self-resonance.
Values are chosen to have self-resonance above the pixel clock.
3.3.2 Power Supply Decoupling
The best power supply performance is obtained with a 0.1 µF ceramic capacitor
decoupling each group of VAA pins and each group of VDD pins to GND. The
capacitors should be placed as close as possible to the device VAA/VDD pins and
GND pins and connected with short, wide traces.
The 47
ripple; the 0.1
When a linear regulator is used, the proper power-up sequence must be
verified to prevent latchup. A linear regulator is recommended to filter the analog
power supply if the power supply noise is greater than or equal to 200 mV. This is
especially important when a switching power supply is used, and the switching
frequency is close to the raster scan frequency. About 5% of the power supply
hum and ripple noise less than 1 MHz will couple onto the analog outputs.
µF capacitor shown in Figure 3-2 is for low-frequency power supply
µF capacitors are for high-frequency power supply noise rejection.
3.3.3 COMP Decoupling
The COMP pin must be decoupled to the closest VAA pin, typically with a 0.1 µF
ceramic capacitor. Low-frequency supply noise will require a larger value. The
COMP capacitor must be as close as possible to the COMP and VAA pins. A
surface-mount ceramic chip capacitor is preferred for minimal lead inductance.
Lead inductance degrades the noise rejection of the circuit. Short, wide traces will
also reduce lead inductance.
3.3.4 VREF Decoupling
A 0.1 µF ceramic capacitor should be used to decouple this input to GND.
3.3.5 VBIAS Decoupling
A 0.1 µf ceramic capacitor should be used to decouple this output to GND.
100123BConexant3-5
Page 58
3.0 PC Board ConsiderationsBt868/Bt869
3.4 Signal InterconnectFlicker-Free Video Encoder with Ultrascale
TM
Technology
3.4 Signal Interconnect
3.4.1 Digital Signal Interconnect
The digital inputs to the Bt868/869 should be isolated as much as possible from
the analog outputs and other analog circuitry. Also, these input signals should not
overlay the analog power plane or analog output signals.
Most of the noise on the analog outputs will be caused by excessive edge rates
(less than 3 ns), overshoot, undershoot, and ringing on the digital inputs.
The digital edge rates should not be faster than necessary, as feedthrough
noise is proportional to the digital edge rates. Lower-speed applications will
benefit from using lower-speed logic (3–5 ns edge rates) to reduce data-related
noise on the analog outputs.
Transmission lines will mismatch if the lines do not match the source and
destination impedance. This will degrade signal fidelity if the line length
reflection time is greater than one-fourth the signal edge time (refer to
Application Notes AN-11 and AN-12). Line termination or line-length reduction
is the solution. For example, logic edge rates of 2 ns require line lengths of less
than 4 inches without use of termination. Ringing can be reduced by damping the
line with a series resistor (30–300
Radiation of digital signals can also be picked up by the analog circuitry. This
is prevented by reducing the digital edge rates (rise/fall time), minimizing ringing
with damping resistors, and minimizing coupling through PC board capacitance
by routing the digital signals at a 90-degree angle to any analog signals.
The clock driver and all other digital devices must be adequately decoupled to
prevent noise generated by the digital devices from coupling into the analog
circuitry.
Ω).
3.4.2 Analog Signal Interconnect
The Bt868/869 should be located as close as possible to the output connectors to
minimize noise pickup and reflections caused by impedance mismatch.
The analog outputs are susceptible to crosstalk from digital lines; digital traces
must not be routed under or adjacent to the analog output traces.
To maximize the high-frequency power supply rejection, the video output
signals should overlay the ground plane.
For maximum performance, the analog video output impedance, cable
impedance, and load impedance should be the same. The load resistor connection
between the video outputs and GND should be as close as possible to the
Bt868/869 to minimize reflections. Unused analog outputs should be connected
to GND.
3-6Conexant100123B
Page 59
Bt868/Bt8693.0 PC Board Considerations
Flicker-Free Video Encoder with Ultrascale
TM
Technology3.5 Applications Information
3.5 Applications Information
3.5.1 Electrostatic Discharge and Latchup Considerations
Correct electrostatic discharge (ESD)-sensitive handling procedures are required
to prevent device damage. Device damage can produce symptoms of catastrophic
failure or erratic device behavior with leaky inputs.
All logic inputs should be held low until power to the device has settled to the
specified tolerance. DAC power decoupling networks with large time constants
should be avoided; they could delay VAA and VDD power to the device. Ferrite
beads must be used only for analog power VAA decoupling. Inductors cause a
time-constant delay that induces latchup, and should not be substituted for a
ferrite bead.
Latchup can be prevented by ensuring that all VAA and all GND pins are at
the same potential and that the VAA and VDD supply voltage is applied before
the signal pin voltages. The correct power-up sequence ensures that any signal pin
voltage will never exceed the power supply voltage.
3.5.2 Clock and Subcarrier Stability
The color subcarrier is derived directly from the CLKO (derived from
XTALIN/XTALOUT) CLKI when EN_XCLK=1 input, hence any jitter or
frequency deviation of CLKO (XTALIN/XTALOUT) or CLKI when
EN_XCLK=1 will be transferred directly to the color subcarrier. Jitter within the
valid CLKO cycle interval will result in hue noise on the color subcarrier on the
order of 0.9–1.6 degrees per nanosecond. Random hue noise can result in
degradation in AM/PM noise ratio (typically around 40 dB for consumer media
such as Videodiscs and VCRs). Periodic or coherent hue noise can result in
differential phase error (which is limited to 10 degrees by FCC cable TV
standards).
Any frequency deviation of CLKO from nominal will challenge the subcarrier
tracking capability of the destination receiver. This may range from a few
parts-per-million (ppm) for broadcast equipment to 100 ppm for industrial
equipment, to a few hundred ppm for consumer equipment. Greater subcarrier
tracking range generally results in poorer subcarrier decoding dynamic range, so
that receivers that tolerate jitter and wide subcarrier frequency deviation will
introduce more noise in the decoded image. Crystal-based clock sources with a
maximum total deviation of 50 ppm (NTSC) or 25 ppm (PAL) across the
temperature range of 0
industrial applications. In rare cases, temperature-compensated clock sources
with tighter tolerances may be warranted for broadcast or more stringent PAL
(e.g., type I) applications.
Some applications call for maintaining correct Subcarrier-Horizontal (SC-H)
phasing for correct color framing. This requires subcarrier coherence within
specified tolerances over a four-field interval for 525-line systems or 8 fields for
625-line systems. Any clock interruption (even during vertical blanking interval)
which results in mis-registration of the CLKI input or nonstandard pixel counts
per line, can result in SC-H excursions outside the NTSC limit of ±40 degrees
°C to 70°C produce the best results for consumer and
100123BConexant3-7
Page 60
3.0 PC Board ConsiderationsBt868/Bt869
3.5 Applications InformationFlicker-Free Video Encoder with Ultrascale
(reference EIA RS170A) or the PAL limit of ±20 degrees (reference EBU
D23-1984).
In slave mode, any deviation exceeding the 50 ppm (NTSC) or 25 ppm (PAL)
limits of the number CLKO cycles between HSYNC* falling edges when in slave
mode may result in a switch to Master Mode.
TM
Technology
3.5.3 Filtering Radio Frequency Modulator Connection
The Bt868/869’s internal upsampling filter alleviates external filtering
requirements by moving significant sampling alias components above 19 MHz
and reducing the sinx/x aperture loss up to the filters passband cutoff of 5.75
MHz. While typical chrominance subcarrier decoders can handle the Bt868/869
output signals without analog filtering, the higher frequency alias products pose
some EMI concerns and may create troublesome images when introduced to a
radio frequency (RF) modulator. When the video is presented to an RF modulator,
it should be free of energy in the region of the aural subcarrier (4.5 MHz for
NTSC, 5.5–6.5 MHz for PAL). Hence some additional frequency traps may be
necessary when the video signal contains fundamental or harmonic energy (as
from unfiltered character generators) in that region. Where better frequency
response flatness is required, some peaking in the analog filter is appropriate to
compensate for residual digital filter losses with sufficient margin to tolerate 10%
reactive components.
A three-pole elliptic filter (one inductor, three capacitors) with a 6.75 MHz
passband can provide at least 45 dB attenuation (including sinx/x loss) of
frequency components above 20 MHz and provide some flexibility for mild
peaking or special traps. An inductor value with a self-resonant frequency above
80 MHz is chosen so that its intrinsic capacitance contributes less than 10% of the
total effective circuit value. The inductor itself may induce 1% (0.1 dB) loss. Any
additional ferrites introduced for EMI control should have less than 5
impedance below 5 MHz to minimize additional losses. The capacitor to ground
at the Bt868/869 output pin is compensating for the parasitic capacitance of the
chip plus any protection diodes and lumped circuit traces (about 22 pF +
5 pF/diode). Some filter peaking can be accomplished by splitting the 75
source impedance across the reactive PI filter network. However, this will also
introduce some chrominance-luminance delay distortion in the range of 10–20 ns
for a maximum of 0.5 dB boost at the subcarrier frequency.
The filter network feeding an RF modulator may include the aforementioned
trap, which could take two forms depending on the depth of attenuation and type
of resonator device employed.
The trap circuitry can interact with the lowpass filter, compromising
frequency response flatness. A simple PNP buffer can preserve the benefits of an
oversampling encoder when simultaneous Composite Video Baseband Signals
(CVBS) are required for driving external cables. In addition, an active video
buffer, serves to isolate the RF modulator signal amplitude from anomalies in the
external termination. This buffer can be implemented with a transistor array or
video amplify IC which provides a gain of two (before series termination),
capable of driving 740 µA into the 75
input/output compliance range. When simultaneous Y/C (s-video) outputs are not
required, a second CVBS signal can be created (with a 600 mV sync to tip offset)
by tying these pins together with a single termination resistor (typically 75
driving the lowpass filter circuit.
Ω
Ω
Ω destination, and is biased within its
Ω) and
3-8Conexant100123B
Page 61
Bt868/Bt8693.0 PC Board Considerations
Flicker-Free Video Encoder with Ultrascale
The RF modulator typically has a high input impedance (about 1k Ω ±30%)
and loose tolerance. Consequently, the amplitude variation at the modulator input
will be greater, especially when the trap is properly terminated at the modulator
input for maximum effect. Some modulators, video or aural fidelity, degrade
dramatically when overdriven, so the value of the effective termination
(nominally 37.5
linearity (or depth of modulation margin) in the RF signal.
A two-section trap (with associated inductor) may be warranted to achieve
better than 20 dB attenuation when stereo, SAP, or AM aural carriers are
generated, or when > 40 dB audio dynamic range is desired. Some impedance
isolation (e.g., buffer) may be required before the trap to obtain the flattest
frequency response. See Figure 3-2.
TM
Technology3.6 Bt868/Bt869 Evaluation Board
Ω) may need to be adjusted downward to maintain sufficient
3.6 Bt868/Bt869 Evaluation Board
See Figure 3-4 for a schematic diagram of the Bt868 EVK evaluation card. This is
a reference design intended to facilitate implementation of Conexant’s VGA
Encoder into a graphics card. The Bt868EVK may be obtained through your local
Conexant Semiconductor sales office.
100123BConexant3-9
Page 62
3.0 PC Board ConsiderationsBt868/Bt869
3.6 Bt868/Bt869 Evaluation BoardFlicker-Free Video Encoder with Ultrascale
Figure 3-4. Bt868/Bt869 Evaluation Board (1 of 6)
TM
Technology
3-10Conexant100123B
Page 63
Bt868/Bt8693.0 PC Board Considerations
Flicker-Free Video Encoder with Ultrascale
Figure 3-4. Bt868/Bt869 Evaluation Board (2 of 6)
TM
Technology3.6 Bt868/Bt869 Evaluation Board
100123BConexant3-11
Page 64
3.0 PC Board ConsiderationsBt868/Bt869
3.6 Bt868/Bt869 Evaluation BoardFlicker-Free Video Encoder with Ultrascale
Figure 3-4. Bt868/Bt869 Evaluation Board (3 of 6)
TM
Technology
3-12Conexant100123B
Page 65
Bt868/Bt8693.0 PC Board Considerations
Flicker-Free Video Encoder with Ultrascale
Figure 3-4. Bt868/Bt869 Evaluation Board (4 of 6)
TM
Technology3.6 Bt868/Bt869 Evaluation Board
100123BConexant3-13
Page 66
3.0 PC Board ConsiderationsBt868/Bt869
3.6 Bt868/Bt869 Evaluation BoardFlicker-Free Video Encoder with Ultrascale
Figure 3-4. Bt868/Bt869 Evaluation Board (5 of 6)
TM
Technology
3-14Conexant100123B
Page 67
Bt868/Bt8693.0 PC Board Considerations
Flicker-Free Video Encoder with Ultrascale
Figure 3-4. Bt868/Bt869 Evaluation Board (6 of 6)
TM
Technology3.6 Bt868/Bt869 Evaluation Board
100123BConexant3-15
Page 68
3.0 PC Board ConsiderationsBt868/Bt869
3.7 Serial InterfaceFlicker-Free Video Encoder with Ultrascale
TM
Technology
3.7 Serial Interface
3.7.1 Data Transfer on the Serial Interface Bus
Figure 3-5 illustrates the relationship between SID (Serial Interface Data) and
SIC (Serial Interface Clock) to be used when programming the internal registers
via the Serial Interface bus. If the bus is not being used, both SID and SIC lines
must be left high.
Every byte put onto the SID line should be 8 bits long (MSB first), followed
by an acknowledge bit, which is generated by the receiving device. Each data
transfer is initiated with a start condition and ended with a stop condition. The
first byte after a start condition is always the slave address byte. If this is the
device’s own address, the device will generate an acknowledge by pulling the SID
line low during the ninth clock pulse, then accept the data in subsequent bytes
(auto-incrementing the subaddress) until another stop condition is detected.
The eighth bit of the address byte is the read/write bit (high = read from
addressed device; low = write to the addressed device) so, for the Bt868/869, the
subaddress is only considered valid if the R/W bit is low. Data bytes are always
acknowledged during the ninth clock pulse by the addressed device. Note that
during the acknowledge period, the transmitting device must leave the SID line
high.
Premature termination of the data transfer is allowed by generating a stop
condition at any time. When this happens, the Bt868/869 will remain in the state
defined by the last complete data byte transmitted and any master acknowledge
subsequent to reading the chip ID (subaddress 0x89) is ignored.
Figure 3-5. SID/SIC Diagram
SIC
1
23456789
SID
MSB
Start Condition
(1)
Acknowledge generated by Bt868/869.
Main Address
Slave
(XX)
123456789
LSB
(1)
Subaddress
(XX)
Subsequent Bytes and Acknowledge
Interpreted as Data Values for
Auto-Incrementing Subaddress Locations
1234 56 789
(1)(1)
Data
(XX)
Stop Condition
3-16Conexant100123B
Page 69
4.0 Parametric Information
4.1 DC Electrical Parameters
DC electrical parameters are defined in Tabl es 4 -1 through 4-3. AC electrical
parameters are defined in
Table 4 -4. Timing diagrams are in Figures 4-1 through 4-3.
Table 4-1. Recommended Operating Conditions
ParameterSymbolMinTypMaxUnits
4
Power SupplyVAA, VDD3.003.303.60V
Serial Input SupplyVDD_SI3.005.25V
Ambient Operating TemperatureTA070°C
DAC Output LoadRL37.5W
Nominal RSETRSET100.0W
Table 4-2. Absolute Maximum Rating (1 of 2)
ParameterSymbolMinTypMaxUnits
VAA, VDD (measured to GND)7.0V
VDD_SI (measured to GND)7.0V
100123BConexant4-1
Page 70
4.0 Parametric InformationBt868/Bt869
4.1 DC Electrical ParametersFlicker-Free Video Encoder with Ultrascale
Table 4-2. Absolute Maximum Rating (2 of 2)
ParameterSymbolMinTypMaxUnits
Voltage on Any Signal Pin
(1)GND – 0.5VDD_SI+ 0.5V
TM
Technology
Analog Output Short Circuit Duration to Any Power
Supply or Common
Storage TemperatureTS– 65+150°C
Junction TemperatureTJ+125°C
Vapor Phase Soldering (1 Minute)TVSOL220°C
(1)
This device employs high-impedance CMOS devices on all signal pins. It should be handled as an ESD-sensitive device.
Voltage on any signal pin that exceeds the power supply or ground voltage by more than 0.5 V can cause destructive latchup.
Stresses above those listed under “Absolute Maximum Ratings” can cause permanent damage to the device. This
is a stress rating only, and functional operation of the device at these or any other conditions above those listed in
the operational sections of this specification is not implied. Exposure to absolute maximum rating conditions for
extended periods can affect device reliability.
Table 4-3. DC Characteristics (1 of 2)
ParameterSymbolMinTypMaxUnits
Video D/A Resolution101010Bits
Output Current-DAC Code 1023 (Iout Full Scale)34.13mA
Output Voltage-DAC Code 10231.28V
Video Level Error (Nominal Resistors)5%
Output Capacitance (of DAC output)22pF
ISCIndefinite
Digital Inputs (Except those specified below)
Input High VoltageVIH2.0
Input Low VoltageVILGND – 0.50.8V
Input High Current (Vin = 2.4 V)IIH1µA
Input Low Current (Vin = 0.4 V)IIL– 1µA
Input Capacitance (f =1 MHz, Vin = 2.4 V)CIN7pF
SID, SDO
Input High VoltageVIH0.7
Input Low VoltageVILGND – 0.50.3 * VDD_SIV
CLKI Input
Input High VoltageVIH2.4
Input Low VoltageVILGND – 0.50.8V
VDD_SI
*
VDD_I
VDD_SI
VDD_I
+ 0.5
+ 0.5
+ 0.5
V
V
V
4-2Conexant100123B
Page 71
Bt868/Bt8694.0 Parametric Information
Flicker-Free Video Encoder with Ultrascale
Table 4-3. DC Characteristics (2 of 2)
ParameterSymbolMinTypMaxUnits
Digital Outputs
Output High Voltage (IOH = –400 µA)VOH2.4VDDV
Output Low Voltage (IOL = 3.2 mA)VOLGND0.4V
Three-State CurrentIOZ50µA
Output CapacitanceCDOUT10pF
NOTE(S): Recommended Operating Conditions, NTSC CCIR 601 operation, and internal clock frequency = 27 MHz. As the above
parameters are guaranteed over the full temperature range (0°C to 70°C), temperature coefficients are not specified or required.
Typical values are based on nominal temperature, i.e., room temperature, and nominal voltage, i.e., 3.3 V.
TM
Technology4.1 DC Electrical Parameters
100123BConexant4-3
Page 72
4.0 Parametric InformationBt868/Bt869
4.2 AC Electrical ParametersFlicker-Free Video Encoder with Ultrascale
TM
Technology
4.2 AC Electrical Parameters
Table 4-4. AC Characteristics (1 of 2)
Parameter
Hue Accurac(1, 2)
Color Saturation Accuracy
Chroma AM/PM Noise
Differential Gain
Differential Phase(2)
SNR (Unweighted 100 IRE Y Ramp
Tilt Correct)
RMS 6.3.1dB rms
Peak Per iodic6.3.2dB p–p
100 IRE Multiburst
(2)6.2.2.1C3.4.1.3% p–p
(2)
(1, 2)
(3)
(3)
EIA/TIA
250C Ref
1 MHz
Red Field
6.2.2.2C3.4.1.4
6.1.1
CCIR 567SymbolMinTypMaxUnits
Gain/frequencyC3.5.4.1
Chroma/Luma Gain Ineq
Chroma/Luma Delay Ineq
Short Time Distortion
100 IRE/PIXEL(3)
(3)
(3)6.1.2C3.5.3.2ns
6.1.2.2C3.5.3.1
6.1.6%
± ×
± %
dB rms
× p–p
± IRE
± IRE
Luminance Nonlinearity(2)6.2.1%
Chroma/Luma Intermod(2)
Chroma Nonlinear Gain
Chroma Nonlinear Phase
Pixel/Control Setup Time
Pixel/Control Hold Time(2)2.35ns
Control Output Delay Time(4)315ns
Control Output Hold Time(4)42ns
CLKI/O Frequency40.5MHz
CLKI/O Pulse Width Low Duty
Cycle
CLKI/O Pulse Width High Duty
Cycle
4-4Conexant100123B
(2)
(2)
(2)1312 ns
6.2.3
6.2.4.1
6.2.4.2
± IRE
± IRE
± ×
405060%
405060%
Page 73
Bt868/Bt8694.0 Parametric Information
Flicker-Free Video Encoder with Ultrascale
Table 4-4. AC Characteristics (2 of 2)
TM
Technology4.2 AC Electrical Parameters
Parameter
CLKO to CLKI Delay70.8CLKO cycles
SLAVE to HSYNC*/VSYNC*
Three-state
SLAVE to HSYNC*/VSYNC* Active62CLKI cycles
VAA Supply Current132mA
VAA Power-Down Current1mA
VDD Supply Current118mA
VDD Power-Down Current1mA
Total Supply Current250mA
(1)
5/7.5/75/7.5 Color bars normalized to burst.
(2)
Guaranteed by characterization.
(3)
Without post filter. Guaranteed by design.
(4)
Control pins are defined as: P[11:0], BLANK*, HSYNC*, VSYNC*, FIELD, CLKDIR, RESET*, PAL, and SLAVE.
5. “Recommended Operating Conditions,” NTSC CCIR 601 operation, and CLK frequency = 27 MHz. Analog output load <
pF. HSYNC*, VSYNC*, BLANK*, and FIELD output load <
temperature range, temperature coefficients are not specified or required. Typical values are based on nominal temperature,
i.e., room temperature, and nominal voltage, i.e., 3.3 V. Video input and output timing is shown in Figure 4-1.
EIA/TIA
250C Ref
CCIR 567SymbolMinTypMaxUnits
52CLKI cycles
75
75 pF. As the above parameters are guaranteed over the full
100123BConexant4-5
Page 74
4.0 Parametric InformationBt868/Bt869
4.2 AC Electrical ParametersFlicker-Free Video Encoder with Ultrascale
Figure 4-1. Interface Timing
CLKO
7
CLKI
P[11:0]
1
2
HSYNC*,VSYNC*,
BLANK* (Input)
1
CLKO/CLKI
(Internal/External
Clock Source)
2
1
2
TM
Technology
HSYNC*,VSYNC*
BLANK* (Output)
SLAVE
HSYNC*,VSYNC*
2.4
.8
4
3
6
5
4-6Conexant100123B
Page 75
Bt868/Bt8694.0 Parametric Information
Flicker-Free Video Encoder with Ultrascale
Figure 4-2. Master Mode with Flicker Filter Interface Timing
CLKI
P[11:0]
(Mux Mode)
P[23:0]
(Non-
Mux Mode)
BLANK*
(Input)
CLKO
HSYNC*
(Output)
TM
Technology4.2 AC Electrical Parameters
POL
POH
H_BLANKI - 3
P1L P1H P2L
P0P1P2Pn
P2H
PnH
Internal
Sample
Counter
VSYNC*
(Output)
Internal
Counter
BLANK*
(Output)
Line
Sample1 Sample2Sample2Sample1
Line 1Line V_BLANK1+1
Sample
H_Blank
Sample
H_Blank
Sample
H_Blank
Sample
H_Blank
100123BConexant4-7
Page 76
4.0 Parametric InformationBt868/Bt869
4.2 AC Electrical ParametersFlicker-Free Video Encoder with Ultrascale
Figure 4-3. Slave Mode with Flicker Filter Interface Timing
CLKO
CLKI
POL
POH
P1L
P1H
P2LPnL
H_BLANK
P0P1P2Pn
BLANK*
(Input)
HSYNC*
(Input)
PnH
TM
Technology
Internal
Sample
Counter
VSYNC*
(Input)
BLANK*
(Output)
Sample
HCLKI
Sample1
Sample
H_BLANKI
-2
Sample
H_BLANKI
-1
Sample
H_BLANKI
Sample
H_Blank
4-8Conexant100123B
Page 77
Bt868/Bt8694.0 Parametric Information
Flicker-Free Video Encoder with Ultrascale
TM
Technology4.3 Mechanical Drawing
4.3 Mechanical Drawing
A detailed mechanical diagram is shown in Figure 0-1.
Figure 0-1. 80 MQFP Package Diagram
80 MQFP - 1.6/0.15mm FORM
TOP VIEW
D
D2
BOTTOM VIEW
D1
E
E2
e
SIDE VIEW
DETAIL A
A2
A1
1.60
(.063)
b
A
L
REF.
S
Y
M
B
O
L
MIN. NOM.MAX.
---
A
0.05
A1
A2
16.95
D
D1
D2
16.95
E
E1
E2
L
0.73
L1
e
b
0.25
ALL DIMENSIONS IN
MILLIMETERS
---
---
2.0 REF.
---
14.0 REF.
12.35 REF.
---
14.0 REF.
12.35 REF.
0.80
16 REF.
0.65 BSC
---
2.4
0.35
17.45
17.45
1.03
0.45
E1
100123BConexant4-9
Page 78
4.0 Parametric InformationBt868/Bt869
4.3 Mechanical DrawingFlicker-Free Video Encoder with Ultrascale
TM
Technology
4-10Conexant100123B
Page 79
A
Appendix A. Scaling and I/O Timing
Register Calculations
The calculated values are used to program the registers controlling the total active
pixels and lines in the input frame and the output field, as well as the vertical
scaling register and the clock PLL registers. These calculations assume pixel
resolution for synchronizing the graphics controller, and master mode operation
unless otherwise stated, and require the following input values:
MFP—Minimum Front Porch Blanking in the Input in Clocks = max
(12, Controller_Minimum_Front_Porch_Blanking_Clocks);
MBP—Minimum Back Porch Blanking in the Input in Clocks = max
(4, Controller_Minimum_Back_Porch_Blanking_Clocks);
VOC—desired Vertical Overscan Compensation (e.g., 0.15)
HOC—desired Horizontal Overscan Compensation (e.g., 0.15)
V_ACTIVEI—Active Lines per Input Frame (e.g., 480 or 600)
H_ACTIVE—Active Pixels per Input Line (e.g., 640 or 800)
ALO—Target Active Lines per Output Field (See Table A-2)
TLO—Total Lines per Output Field (See Tab le A-2 )
AT O—Active Time per Output Line (See Tabl e A -2)
TTO—Total Time per Output Line (See Tabl e A- 2)
Table A -1 displays details of the video formats. Tabl e A- 2 details the constant
values dependent on encoding modes. Figures A-1 through A-4 diagram overscan
compensation. Tab le s A-3 through A-10 display overscan values.
Appendix A . Scaling and I/O Timing Register CalculationsBt868/Bt869
Flicker-Free Video Encoder with Ultrascale
TM
Technology
A-20Conexant100123B
Page 99
B
Appendix B. Approved Crystal
Vendors
Conexant conducted a series of internal tests and used the results to generate this
list of approved crystal vendors. Contact your local Conexant Field Applications
Engineer for additional details.
Standard Crystal (El Monte, CA)
Phone Number:(626)443-2121
FAX Number:(626)443-9049
E-mail:stdxtl@worldnet.att.net
Part Numbers for 13.500 MHz, Fundamental, 20pF Load Crystal with an HC49U Type of Package:
Full Height/50 ppm Total Tolerance:AAK13M500000GXE20A
Half-Height/50 ppm:AAL13M500000GXE20A
Full Height/25 ppm:Did Not Qualify
MMD Components (Irvine, CA)
Phone Number:(949)753-5888
FAX Number:(949)753-5889
E-mail:mmdcomp@earthlink.net
Part Numbers for 13.500 MHz, Fundamental, 20pF Load Crystal with an HC49U Type of Package:
Full Height/50ppm Total Tolerance:A20BA1-13.500 MHz.
Half-Height/50 ppm:B20BA1-13,500 MHz
Full Height/25 ppm:Did Not Qualify
General Electronics (San Marcos, CA)
Phone Number:(760)591-4170
FAX Number:(760)591-4164
E-mail:gedlm@4dcomm.com
Part Numbers for 13.500 MHz, Fundamental, 20pF Load Crystal with an HC49U Type of Package:
Full Height/50 ppm Total Tolerance:PKHC49/U-13.500-.020-.005
Half-Height/50 ppm:PKHC49/US-13.500-.020-.005
Full Height/25 ppm:PKHC49/U-13.500-.0025-15R
Fox Electronics (Fort Myers, FL)
Phone Number:(941)693-0099
FAX Number:(941)693-1554
E-mail:barbc@foxonline.com
Part Numbers for 13.500 MHz, Fundamental, 20pF Load Crystal with an HC49U Type of Package:
Full Height/50 ppm Total Tolerance:HC49U-13.500 MHz-/50/0/70/20 pF
Half Height/50 ppm:HC49S 13.500-/50/0/70/20 pF
Full Height/25 ppm:HC49U 13.500-/25/0/70/20 pF
100123BConexantB-1
Page 100
Appendix B . Approved Crystal VendorsBt868/Bt869
Flicker-Free Video Encoder with Ultrascale
Bomar (Middlesex, NJ)
Phone Number:(732)356-7787
FAX Number:(732)356-7362
E-mail:sales@bomarcrystal.com
Part Numbers for 13.500 MHz, Fundamental, 20pF Load Crystal with an HC49U Type of Package:
Full Height/50 ppm Total Tolerance:BRC1C14F-13.50000
Half Height/50 ppm:ACR-49S012025-13.50000
Full Height/25 ppm:BRCIE14F-13.50000
HY-Q (Erlanger, Kentucky)
Phone Number:(606)283-5000
FAX Number:(606)283-0883
E-mail:Cpainter@hyqusa.com
Part Numbers for 13.500 MHz, Fundamental, 20pF Load Crystal with an HC49U Type of Package:
Full Height/50 ppm Total Tolerance:HT81818/01
Half Height/50 ppm:SC30002/01
Full Height/25 ppm:
Phone Number:(425)828-4886 / (888)355-4574
FAX Number:(425)828-4878
E-mail:ilsiam@ilsiamerica.co
Part Numbers for 13.500 MHz, Fundamental, 20pF Load Crystal with an HC49U Type of Package:
Full Height/50 ppm Total Tolerance:HC49U-25/25-13.500-20
Half Height/50 ppm:HC49US-FB1F20-13.500
Full Height/25 ppm:Did Not Qualify
HT81819/01
ILSI America (Kirkland, WA)
TM
Technology
Cardinal Components (Wayne, NJ)
Phone Number:(973)785-1333
FAX Number:(973)785-0053
E-mail:dbabcock@cardinalxtal.com
Part Numbers for 13.500 MHz, Fundamental, 20pF Load Crystal with an HC49U Type of Package:
Full Height/50 ppm Total Tolerance:C49-A4BRC7-25-13.5D20
Half Height/50 ppm:CLP-A4B6C4-50-13.5D20
Full Height/25 ppm:C49-A4B6C4-25-13.5D20
Raltron Electronics Corp. (Miami, FL)
Phone Number:(305)593-6033
FAX Number:(305)594-3973
E-mail:maria@raltron.com
Part Numbers for 13.500 MHz, Fundamental, 20pF Load Crystal with an HC49U Type of Package:
Full Height/50 ppm Total Tolerance:A-13.500-20-RS1
Half Height/50 ppm:AS-13.500-20-RS1
Full Height/25 ppm:A-13.500-20-RS1