HARRIS HMP8156EVAL2

Semiconductor
March 1998
RN
D FO 1 5
E
D
EN
P81
O M M Se e H M
C
E
R
NOT
E
ES
WD
S
IGN
HMP8112A
NTSC/PAL Video Decoder
Features
Description
• Supports ITU-R BT.601 (CCIR601) and Square Pixel
The HMP8112A is a high quality, digital video, color decoder
with internal A/D converters. The A/D function includes a 3:1
analog input mux, Sync Tip AGC, Black clamping and two 8bit A/D Converters. The high quality A/D converters minimize
pixel jitter and crosstalk.
• 3 Composite Analog Inputs with Sync Tip AGC, Black
Clamping and White Peak Control
• Patented Decoding Scheme with Improved 2-Line
Comb Filter, Y/C Separation
The decoder function is compatible with NTSC M, PAL B, D,
G, H, I, M, N and special combination PAL N video standards. Both composite (CVBS) and S-Video (Y/C) input formats are supported. A 2-line comb filter plus a user
selectable Chrominance trap filter provide high quality Y/C
separation. Various adjustments are available to optimize the
image such as Brightness, Contrast, Saturation, Hue and
Sharpness controls. Video synchronization is achieved with
a 4xfSC chroma burst lock PLL for color demodulation and
line lock PLL for correct pixel alignment. A chrominance subsampling 4:2:2 scheme is provided to reduce chrominance
bandwidth.
• NTSC M and PAL (B, D, G, H, I, M, N, CN) Operation
• Composite or S-Video Input
• User-Selectable Color Trap and Low Pass Video
Filters
• User Selectable Hue, Saturation, Contrast, Sharpness,
and Brightness Controls
• User Selectable Data Transfer Output Modes
• 16-Bit 4:2:2 YCbCr
• 8-Bit 4:2:2 YCbCr
The HMP8112A is ideally suited as the analog video interface to VCR’s and camera’s in any multimedia or video system. The high quality Y/C separation, user flexibility and
integrated phase locked loops are ideal for use with today’s
powerful compression processors. The HMP8112A operates
from a single 5V supply and is TTL/CMOS compatible.
• User Selectable Clock Range from 20MHz - 30MHz
• I2C Interface
• VMI Compatible Video Data Bus
Applications
Ordering Information
• Multimedia PCs
PART NUMBER
• Video Conferencing
HMP8112ACN
• Video Editing
• Video Security Systems
• Digital VCRs
TEMP.
RANGE (oC)
0 to 70
PACKAGE
PKG.NO.
80 Ld PQFP†
Q80.14x20
HMP8112EVAL2
PCI Reference Design (Includes Part)
HMP8156EVAL2
Frame Grabber Evaluation Board
(Includes Part)
† PQFP is also known as QFP and MQFP
• Related Products
- NTSC/PAL Encoders: HMP8154, HMP8156A,
HMP8170/1, HMP8172/3
- NTSC/PAL Decoders: HMP8115, HMP8130/1
Table of Contents
Page
Functional Block Diagrams . . . . . . . . . . . . . . . . . . . . . . . . 2
Functional Operation Introduction . . . . . . . . . . . . . . . . . . 5
Internal Register Description Tables. . . . . . . . . . . . . . . . 15
Pinout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Pin Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
AC and DC Electrical Specifications. . . . . . . . . . . . . . . . 25
Typical Performance Curves . . . . . . . . . . . . . . . . . . . . . . 28
Applications Information . . . . . . . . . . . . . . . . . . . . . . . . . 39
Package Dimensions . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
CAUTION: These devices are sensitive to electrostatic discharge. Users should follow proper IC Handling Procedures.
Copyright
© Harris Corporation 1998
1
File Number
4407.2
4-2
EXTERNAL
ANTIALIASING
FILTER
CCLAMP_CAP
GAIN_CTRL
LAGC_CAP
LCLAMP_CAP
LIN2/Y
LIN1
LIN0
CIN
CLAMP
LOGIC
AND
GAIN
CONTROL
AGC
AND
CLAMP
LOGIC
INPUT
MUX
+
+
-
L_ADIN
8-BIT
ADC
CLAMP
8-BIT
ADC
DIGITAL COMPARATOR
SYNC LEVEL
BLACK LEVEL
WHITE PEAK LEVEL
DIGITAL COMPARATORS
L_OUT
EXTERNAL
ANTIALIASING
FILTER
INPUT
SAMPLE
RATE
CONVERTER
VSYNC
DETECT
CHROMA
PLL
FIELD VSYNC
STD_ERR
COLOR
TRAP
LINE
LOCK
PLL
HSYNC
COLOR
ADJUST
USER
ADJUST.
HSYNC
DETECT
LOCKED
COLOR
DEMODULATION
Y/C
SEPARATION
RESET
MICROPROCESSOR
INTERFACE AND
CONTROL
OUTPUT
SAMPLE
RATE
CONVERTER
SDA
SCL
CbCr[7:0]
Y[7:0]
DVLD
ACTIVE
HMP8112A
Functional Block Diagrams
VIDEO DECODER
4-3
Y,CVBS
L[7:0]
CR[7:0]
C
M
U
X
Y DATA
C,CVBS
DATA
INPUT
SAMPLE
RATE
CONVERTER
C,CVBS
DATA
CHROMA
PLL LOOP
FILTER
U,V
ISL
SHARPNESS
ADJUST
VSYNC
DETECT
CHROMA TRAP
ENABLE
CHROMA
TRAP
U,V TO CbCr
COLOR
SPACE
CONVERTER
AND COLOR
KILLER
SATURATION
ADJUST
UV
UV
AGC SATURATION
ADJUST
AGC
ADJUST
CHROMA
PHASE
DETECTOR
HORIZONTAL
M Y DATA AND VERTICAL
U
Y DATA
SHARPNESS
X
ADJUST
Y DATA
CHROMA
DEMODULATOR
C DATA
LINE
DELAY
COMB
FILTER
CLK TO
4FSC RATIO
CHROMA
PLL NCO
HUE
ADJUST
LOW PASS
FILTER ENABLE
LP FILTER
HSYNC
DETECT
STANDARD
SELECT
OUTPUT
SAMPLE
RATE
CONVERTER
LINE LOCKED
NCO
LINE LOCKED
PLL LOOP FILTER
FIELD
VSYNC
STANDARD ERROR
CbCr
DATA
Y
SYNC
DATA
STRIPPER,
BRIGHTNESS,
& CONTRAST
ADJUST
LOCKED
HSYNC
Functional Block Diagrams
ISL
4FSC
CLOCK
CLK
(20MHz - 30MHZ)
HMP8112A
(Continued)
HMP8112A
Functional Block Diagrams
(Continued)
CONTROL
REGISTERS
....
....
....
....
ADDRESS
POINTER
ADDRESS
POINTER
OEN
0x00
0x01
.
.
.
.
CbCr[7:0]
0x1B
CONTROL
DATA BUS
R
E
G
I
S
T
E
R
8/16 OUTPUT
SELECT
CbCr[7:0]
8
32 X 16
DEEP
FIFO
SERIAL SHIFT
REGISTER
R
E
G
I
S
T
E
R
M
U
X
Y[7:0]
Y[7:0]
8
DVLD
A0
ACTIVE
SCL SDA
I2C CONTROL INTERFACE
OUTPUT INTERFACE
Schematic
U1
LUMA0
C1
1.0µF
LUMA1
R3
75
6
C2
1.0µF
LUMA2/Y
LIN0
LIN1
5 LIN2
C3
1.0µF
R2
75
7
R1
75
ANTI-ALIAS
FILTER
CHROMA
R4
75
C4
1.0µF
19
ANTI-ALIAS
FILTER
8
9
77
C5
0.01µF
C6
0.047µF
C7
0.047µF
C9
0.01µF
C10
0.1µF
L_ADIN
L_OUT
CbCr7
CbCr6
CbCr5
CbCr4
CbCr3
CbCr2
CbCr1
CbCr0
Y7
Y6
Y5
Y4
Y3
Y2
Y1
Y0
ACTIVE
DVLD
64
63
60
58
57
56
55
54
65
66
CB_CR7
CB_CR6
CB_CR5
CB_CR4
CB_CR3
CB_CR2
CB_CR1
CB_CR0
30
Y[0..7]
Y7
Y6
Y5
Y4
Y3
Y2
Y1
Y0
Y[0..7]
VCC VCC VCC VCC VCC VCC VCC
R9
10K
R10
10K
R11
10K
R12
10K
R13
10K
R14
4K
DEC_T
DEC_L
34
RESET
40
SDA
41
SCL
38
CLK
13
CLK
36
TEST
DVLD
FIELD
HDRIVE
VDRIVE
RESET
27MHz
VAA
VAA
28
GAIN_CNTL
WPE
R6
750
27
R7
10K
JP1
JUMPER
4-4
R15
4K
ACTIVE
LAGC_CAP
C11
0.01µF
R5
1K
CB_CR[0..7]
67
FIELD
71
76
LCLAMP_CAP HSYNC
70
29
VSYNC
CCLAMP_CAP
78
C8
0.1µF
CIN
CB_CR[0..7]
51
50
49
48
47
45
43
42
R8
50
C12
15pF
SDA
SCL
27MHz
HMP8112A
Introduction
The HMP8112A is designed to decode baseband composite
or s-video NTSC and PAL signals, and convert them to either
digital YCbCr or RGB data.
The digital PLLs are designed to synchronize to all NTSC
and PAL standards. A chroma PLL is used to maintain
chroma lock for demodulation of the color information; a linelocked PLL is used to maintain vertical spatial alignment.
The PLLs are designed to maintain lock even in the event of
VCR headswitches.
tip to maintain an average ADC code of 0. The DC
RESTORE circuit clamps the video signal during the back
porch to maintain an average ADC code of 64. Reference
Figure 2 for timing information and Table 5 for the recommended register values to use for different video standards.
The START and END times of the HSYNC output are also
programmable and can be used as a reference for confirming proper HAGC and DC RESTORE timing.
0HSYNC
The HMP8112A contains two 8-bit A/D converters and an
I2C port for programming internal registers
Analog Video Inputs
VIDEO INPUT
The HMP8112A supports either three composite or two
composite and one S-Video input.
Three analog video inputs (LIN0, LIN1, LIN2) are used to
select which one of three composite video sources are to be
decoded. To support S-video applications, the Y channel
drives the LIN2 analog input, and the C channel drives the
CIN analog input.
DC RESTORE
START
TIME
The analog inputs must be AC-coupled to the video signals,
as shown in the Applications section.
END
TIME
HAGC
START
TIME
Anti-Aliasing Filter
An external anti-alias filter is required to achieve optimum
performance and prevent high frequency components from
being aliased back into the video image.
HSYNC
For the LIN0-2 inputs, a single filter is connected to L_OUT
and L_ADIN. For CIN the anti-aliasing filter should be connected to the CIN input. A recommended filter is shown
below in Figure 1.
START
TIME
END
TIME
END
TIME
FIGURE 2. DC RESTORE AND HAGC TIMING
White Peak Enable
R1
L1
332
8.2µH
C1
33pF
C2
82pF
The white peak enable input, (WPE) enables or disables the
white peak control of the luminance input. If enabled, the
AGC will reduce the gain of the video amplifier when the digital outputs exceed code 248 to prevent over-ranging the
A/D. If disabled, the AGC operates normally, keeping the horizontal sync tip at code 0 and allowing the A/D’s range to go
to 255 at the maximum peak input.
R2
4.02K
FIGURE 1. RECOMMENDED ANTI-ALIASING FILTER
Luminance AGC And DC RESTORE Circuits
Chrominance Input
After a RESET, a change of the video standard, or a PLL
Chrominance Subcarrier Ratio Register load, the decoder
enters Acquisition Mode by attempting to lock to a new video
source. During this mode, the HAGC and DC RESTORE circuits perform continuous gain and bias adjustments until the
PLL is LOCKED onto the video signal. Once LOCKED, the
HAGC and DC RESTORE functions are performed during
programmable window periods for each horizontal video line.
The digital PLL zeroes a 10-bit pixel clock counter during
each horizontal sync tip and increments the count for each
pixel of the entire video line. The AGC amplifier attenuates or
amplifies the analog video signal during the horizontal sync
The chrominance amplifier gain control is manually set by a
voltage applied to the GAIN_CNTL pin. Refer to Figure 3
below for gain characteristics. The chrominance channel
also has a digital AGC which can drive the color reference
burst to a nominal +-20 IRE. This function is enabled by
default on reset, but can be disabled using the Video Input
Control register. The chrominance input is clamped during
the DC RESTORE window to maintain an average ADC
code of 128.
4-5
HMP8112A
7
6
The input sample rate converter will interpolate between
existing CLK samples to create the chroma locked (4xfSC)
samples needed for the color decoder. An interpolation is
done to create the 4xfSC pixel and a correction factor is then
applied.
TEMPERATURE = 25oC
VCC = 5V
LINEAR GAIN
5
INCOMING VIDEO SAMPLES
4
3
2
TIME
1
0
1.6
RESAMPLED VIDEO
1.8
2.0
2.2
2.4
2.6
2.8
3.0
3.2
3.4
GAIN CONTROL VOLTAGE
FIGURE 3. CHROMINANCE AMPLIFIER GAIN
TIME
Reset
4xfSC
The RESET pin is used to return the decoder to an
initialization state. This pin should be used after a power-up
to set the part into a known state. The internal registers are
returned to their RESET state and the Serial I2C port is
returned to inactive state. The RESET pin is an active low
signal and should be asserted for minimum of 1 CLK cycle.
After a RESET or a software reset has occurred all output
pins are three-stated. The VSYNC, HSYNC, DVLD, ACTIVE
and FIELD output pins must be pulled high to ensure proper
operation. A 10K or smaller pullup resistor to VCC is
recommended.
NTSC/PAL Decoder
The NTSC/PAL decoder is designed to convert incoming
Composite or Separated (SVHS, Y/C) video into it’s YCbCr
component parts. The digital phase locked loops are
designed to synchronize to the various NTSC/PAL standards. They provide a stable internal 4xfSC (Frequency of
the Color Sub-Carrier) video clock for color demodulation,
and a line locked clock for vertical spatial pixel alignment.
The decoder uses the CLK to run the A/D converters and the
phase locked loops. This asynchronous master clock for the
decoder eliminates the need for a unique clock source in a
Multimedia application. CLK can run from 20MHz to 30MHz
when using the 16-bit Synchronous Data output Mode. The
user must program the CLK to Color Sub-Carrier Ratio to
match the CLK frequency used (see Internal Phase Locked
Loops discussion). When using the 8-bit Burst Data Output
Mode the CLK should be a 24.54MHz, 27MHz or 29.5MHz
depending on the output video standard chosen. The crystal
oscillator must have a ±50ppm accuracy and a 60/40% duty
cycle symmetry to ensure proper operation. Since the video
data from the external A/D’s are sampled at the CLK frequency a sample rate converter is employed to convert the
data from the CLK rate to the internal decoding frequency of
4xfSC.
FIGURE 4. SAMPLE RATE CONVERSION
The decoder can be used with the following video sources:
Analog Composite - NTSC M, - PAL B, D, G, H, I, N
And Special Combination PAL N
Analog S - VHS (Y/C) - NTSC M, PAL B, D, G, H, I, N
And Special Combination PAL N
Color Separation, and Demodulation
To separate the chrominance modulated color information
from the baseband luminance signal, a 2-Line comb filter is
employed. In NTSC signals the color information changes
phase 180o from one line to the next. This interleaves the
chrominance information at half line intervals throughout the
NTSC video spectrum. Therefore, NTSC has 227.5 cycles of
chrominance per NTSC line. The half of a cycle causes the
next reference burst to be 180o out of phase with the previous line’s burst. The 2-Line comb efficiently removes the
chrominance information from the baseband luminance signal. When decoding NTSC, the decoder maintains full luminance bandwidth horizontally throughout the chrominance
carrier frequency range. Unlike most 2 line comb filter separation techniques, vertical bandwidth is maintained by
means of a proprietary transform technique.
Reset
The RESET pin is used to return the decoder to an initialization state. This pin should be used after a power-up to set the
part into a known state. The internal registers are returned to
their RESET state and the Serial I2C port is returned to inactive state. The RESET pin is an active low signal and should
be asserted for minimum of 1 CLK cycle. After a RESET or a
software reset has occurred all output pins are three-stated.
The VSYNC, HSYNC, DVLD, ACTIVE and FIELD output pins
must be pulled high to ensure proper operation. A 10K or
smaller pullup resistor to VCC is recommended.
4-6
HMP8112A
Y
255
248
Y
I, Q
WHITE
100%
255
240
BLUE
100%
BLUE
75%
212
255
240
RED
100%
RED
75%
212
AMPLITUDE
128
128
44
fH/2
16
fH/2
fH
FREQUENCY
BLACK
Y DATA
RANGE
16
0
128
YELLOW
75%
YELLOW
100%
CYAN
75%
CYAN
100%
44
16
0
Cb DATA
RANGE
Cr DATA
RANGE
FIGURE 7. YCbCr DATA RANGES
Y
Y
I, Q
The decoder is compatible with all NTSC and PAL video formats available throughout the world. Table 2 shows the compatible video standards.
AMPLITUDE
Horizontal Sync Detection
FREQUENCY
FIGURE 5. COMPOSITE NTSC INTERLEAVE SCHEME
For PAL systems there are 283.75 cycles of chrominance
per line. Chrominance information is spaced at quarter line
intervals with a reference phase of 135o. The reference
phase alternates from line to line by 90o. To fully separate
the PAL chrominance and luminance signals the user selectable filters should be enabled. The chroma notch filter built
into the luminance channel should be enabled for PAL systems to reduce cross luminance effects. The low pass filter in
the chrominance processing chain helps to reduce cross
color products.
Horizontal sync is detected in the Output Sample Rate converter (OSR). The OSR spatially aligns the pixels in the vertical direction by using the horizontal sync information
embedded in the digital video data stream. The HSYNC
sync pulse out of the decoder is a video synchronous output
pin. This signal follows the horizontal sync of an input video
source. If there is no source the HSYNC pin will continue to
run at video rates due to the Line Locked PLL free-running.
HSYNC can be moved throughout the video line using the
HSYNC Start and End time registers. This 10-bit register
allows the HSYNC to be moved in OSR clock increments
(12.27MHZ, 13.5MHz or 14.75MHz).
Vertical Sync and Field Detection
Y
I, Q
I, Q
Y
The vertical sync and field detect circuit of the decoder uses
a low time counter to detect the vertical sync sequence in
the video data stream. The low time counter accumulates
the low time encounted after the horizontal sync edge or at
the start of each line. When the low time count exceeds the
vertical sync detect threshold, VSYNC is asserted immediately. VSYNC will remain asserted for a minimum of 1 line.
The FIELD flag is updated at the same time as the VSYNC
line. The FIELD pin is a ‘0’ for ODD fields and a ‘1’ for even
fields.
AMPLITUDE
fH/4
fH/4
fH
FREQUENCY
Y
I, Q
I, Q
In the case of lost vertical sync or excessive noise that would
prevent the detection of vertical sync, the FIELD flag will
continue to toggle. Lost vertical sync is declared if after 337
lines a vertical sync period was not detected for 3 successive
lines. When this occurs the phase locked loops are initialized
to the acquisition state.
Y
AMPLITUDE
FREQUENCY
FIGURE 6. COMPOSITE PAL INTERLEAVE SCHEME
The demodulator in the decoder decodes the color components into U and V. The U and V components are converted
to Cb and Cr components after the decoding process.
YCbCr has a usable data range as shown in Figure 7. The
data range for Y is limited to a minimum of 16.
The VSYNC pulse out of the decoder follows the vertical
sync detection and is typically 6.5 lines long. The VSYNC
will run at the field rate of the selected video standard
selected. For NTSC the field rate is 60Hz and for PAL the
field rate is 50Hz. This signal will continue to run even in the
event of no incoming video signal.
4-7
HMP8112A
LINE #
524
525
1
2
3
4
5
6
7
8
9
10
VIDEO
INPUT
OV
VSYNC DETECT THRESHOLD
LOW TIME
COUNTER
HSYNC
5
VSYNC
4
3
2
1
APPROX. 5.75 LINES
FIELD
‘ODD’ FIELD
‘EVEN’ FIELD
FIGURE 8. VSYNC TIMING AND THE EVEN TO ODD FIELD TRANSITION
LINE #
262
263
264
265
266
267
268
269
270
271
272
5
4
3
2
1
273
VIDEO
INPUT
LOW TIME
COUNTER
VSYNC DETECT THRESHOLD
OV
HSYNC
6
VSYNC
FIELD
APPROX. 6.25 LINES
‘ODD’ FIELD
‘EVEN’ FIELD
FIGURE 9. VSYNC TIMING AND THE ODD TO EVEN FIELD TRANSITION
Internal Phase Locked Loops
The HMP8112A has two independent digital phase locked
loops on chip. A chroma phase-locked loop is implemented
to maintain chroma lock for demodulation of the color channel, and a line locked phase lock loop is implemented to
maintain vertical spatial alignment. The phase locked loops
are designed to maintain lock even in the event of VCR
headswitches.
The HMP8112A can use a main crystal (CLK) of 20MHz to
30MHz. The crystal is used as a reference frequency for the
internal phase locked loops. The ratio of the crystal frequency to the video standard is programmed into an internal
register for the PLLs to correctly decode video.
The HMP8112A decoder contains 2 sample rate converters
and 2 phase locked loops that lock to the incoming video.
The input sample rate converter synchronizes the digitized
video from the CLK rate to a 4xfSC rate. The chrominance is
separated from the luminance and then demodulated.
The Chroma PLL uses the CLK source as a reference frequency. To initialize the Chroma PLL, the CLK to 4xfSC ratio
value must be loaded into the Chroma PLL Ratio Register
pair. A default 16-Bit Fractional Chroma PLL Ratio Value of
0x87C1 is used after a system RESET is applied. Refer to
Table 1 for example PLL Ratio values to use with the sup-
ported video standards 27MHz or 24.54MHz clocks. Using a
different CLK will require different values to be calculated per
the method shown below. The default assumes a CLK of
27MHz and NTSC as the video standard, and is calculated
as follows:
Ratio =
=
=
Register Data:
Hex Conversion:
(4 x fSC) / CLK
(4 x 3.579545MHz) / 27MHz
0.530303
Ratio * 65536
0.530303 * 65536 = 34753.94
0x87C1
The Output Sample Rate converter is locked to the horizontal line frequency and is used to spatially align pixels in a
field. The LOCKED flag signals when the phase locked loop
is within a ±4 pixel range of the horizontal sync edge. When
line errors exceed that range the LOCKED flag is cleared.
In cases where VCRs are used in Pause, Fast Forward or
Fast Reverse, lines are typically dropped or added by the
VCR. In a worst case scenario a VCR line tolerance will vary
by ±8%. The standard detect logic checks the line count
against the given standard to determine an error. VCRs in
trick mode cannot cause a standard error. With an NTSC
standard VCR the number of lines in a field should not
4-8
HMP8112A
TABLE 1. COMPATIBLE VIDEO INPUT STANDARDS
COLOR
SUBCARRIER
fSC
27MHz
PLL
Ratio
24.54MHz
PLL
Ratio
FIELDS/
SECOND
VERTICAL
LINES
LINE
FREQUENCY
NOMINAL
BANDWIDTH
BLACK
SETUP TO
BLANK
NTSC
(M)
3.579545MHz
0x87C1
0x955D
60Hz
525
15,734
(± 0.0003%)
4.2MHz
7.5 IRE
PAL
(B, D, G, H, I)
4.43361875MHz
0xA826
0xB901
50Hz
625
15,625
(± 0.02%)
5.0MHz
0 IRE
PAL
(M)
3.57561149MHz
0x879B
0x9533
60Hz
525
15,750
(± 0.0003%)
4.2MHz
7.5 IRE
PAL
(N)
4.43361875MHz
0xA826
0xB901
50Hz
625
15,625
(± 0.15%)
4.2MHz
7.5 IRE
PAL Special
Combination N
3.58205625MHz
0x97DA
0x9578
50Hz
625
15,750
(± 0.15%)
4.2MHz
7.5 IRE
STANDARD
exceed 285. Greater than 285 lines in a field is interpreted
as a PAL video source. An ideal NTSC source should have
262.5 lines per field and a PAL source should have 312.5
lines per field.
The HMP8112A can detect a STANDARD ERROR that signals when the video received does not match the standard
that was programmed into the Video Input Control Register.
This flag, when asserted, tells the user that the video standard that was expected was not found and a different standard should be selected in the Video Input Control register.
The error flag is cleared after a RESET or after the Chroma
PLL Clock Ratio register has been loaded via the I2C bus.
After the flag is cleared the standard error logic verifies the
video standard. The error flag is set after 2 vertical sync periods have passed and the line count did not match the
expected line count.
Video Adjustments
The HMP8112A allows the user to vary such video parameters as Contrast, Brightness, Sharpness, Hue and Color Saturation. These adjustments can be made via the I2C
interface. Contrast, brightness and sharpness are luminance
controls. The full dynamic range of the luminance channel
can be used by selecting the IRE setup cancellation mode.
This mode will remove the IRE setup and blanking level offset to take advantage of the full dynamic range of the luminance processing path. The sharpening filters allow the
enhancement of low, mid and high frequency components of
the luminance signal to compensate for low amplitude video.
Vertical sharpness is also controlled via the I2C interface.
Hue and Color saturation controls enhance the CbCr components of the incoming video, all under user control.
Luminance Adjustments
YOUT = (Y - IRE Setup + BRIGHTNESS) x CONTRAST
BRIGHTNESS
(-64 TO +63)
Y DATA
FROM
DECODER
+
+
X
8
IRE BLACK SETUP
(NTSC = 73, PAL = 64)
Y’
CONTRAST
(0 TO 1.999)
FIGURE 10. LUMINANCE CONTROL SETTINGS PATH
Brightness
The user can control the brightness of the incoming video by
programming the Brightness register. The brightness adjustment will offset the Y component. The brightness register is
an 8-bit register where the bottom 7 bits are brightness control and the top bit enables NTSC 7.5 IRE black setup cancellation.
When the IRE bit is set (1) for NTSC, then 73 is subtracted
from the Y data. If the IRE bit is cleared (0) for PAL, then 64
is subtracted. The brightness control bits BR[6-0] will
brighten the picture as the value is increased. BR = -64 is the
darkest and BR = +63 is the brightest. The default value of
the register after a RESET is 0 (80H).
Contrast
The contrast adjustment will allow the user to increase and
decrease the gain of the Y data. The contrast factor is an 8bit number (as shown below) that ranges from 0 to 1.992.
X.XXXXXXX
The Luminance data can be adjusted in the HMP8112A. The
user can adjust brightness and contrast of the Y or luminance data. The user can also set the IRE or setup subtraction value to eliminate the black pedestal offset from NTSC
signals. The Contrast adjustment range can exceed a value
of one so as to take full advantage of the 8-bit dynamic range
for Y. The user control settings executes the equation
The default register value of 1.4766 (0xBD) is calculated as
follows:
4-9
Register = Factor * 128 = 1.4766 * 128 = 189 = 0xBD
HMP8112A
Hue or Tint Adjust
The Hue adjustment is applied to the U and the V color difference signal. The Hue adjusts the phase of the given UV
data. The Hue can be adjusted by ±30 degrees in 1/4 degree
increments. This is achieved by changing the Burst Phase
Locked reference point. Figure 11 shows the block diagram
for the color adjustment section. This default value for this
register is 0.
VIDEO
DATA
COLOR
DECODER
DEMODULATED
UV DATA
CHROMA
AGC AND
USER
SETTINGS
UV DATA
+
HUE
ADJUST
HUE OFFSET
TO INPUT SAMPLE
RATE
CONVERTER
The Color Killer
(AGC Hysteresis and Loop Limits)
The color killer will disable the color difference path and set
the U and V components to zero. The automatic color killer
circuitry uses the AGC threshold to determine the maximum
and minimum gain factor limits. The loop filter determines
how much the AGC gain factor can be changed within one
line. The maximum gain factor (Max = 8) and the minimum
gain factor (Min = 0.5) will limit the range of the AGC. When
the gain factor exceeds the maximum gain factor of 8, the
gain factor is limited to 8. Once the signal has an amplitude
of 1/16th, the nominal video the color killer is enabled and
the chroma phase locked loop holds it’s last phase reference. While the color killer is enabled, the U and V components are forced to zero. Once the input video signal reaches
1/7th the optimum amplitude the color killer is disabled and
the color is returned.
UV
DATA
CHROMA
PHASE LOCKED
LOOP
÷4096
FIGURE 11. HUE ADJUST BLOCK DIAGRAM
I2C
MAX
GAIN
FACTOR
MIN
GAIN
FACTOR
COLOR
KILLER
AGC
GAIN
FACTOR
AGC
ENABLE
LINE
COUNT
Horizontal/Vertical Sharpness
The frequency characteristics of the video waveform can be
altered to enhance the sharpness of the picture. The Horizontal Sharpness register acts as a 4 band equalizer where
the amplitude of specific frequency ranges can be
enhanced or diminished. The Sharpness Control Register
allows the Low (LF), Mid (MF) and High Frequency (HF)
bands of the luminance signal to be enhanced. Vertical
Sharpness can be adjusted to 1 or a factor of 0. The
RESET default is a factor of 1.0
The 2-bit values allow 4 choices of scaling factors. The
sharpness control helps to compensate for losses in the
scaling interpolators that can reduce the amplitude of high
frequency components.
FIGURE 12. LOOP FILTER BLOCK DIAGRAM (HYSTERESIS)
The dynamic range of the AGC allows it to compensate for
video that is 1/8 to 2 times the specified nominal of 1VP-P.
Saturation
The color saturation component is controlled via the Color
Saturation Registers. The color saturation is applied to the
UV components after the AGC function. The saturation value
is multiplied by the UV data to increase the color intensity.
This is an 8-bit number (as shown below) that ranges from 0
to 1.992.
X.XXXXXXX
TABLE 2. SHARPNESS GAIN FACTOR SELECTS
XF1
XF0
GAIN FACTOR
0
0
Scaled By 1.0
0
1
Scaled By 2.0
1
0
Scaled By 4.0
1
1
Scaled By 0
The default register value of 1.2266 (0x9D) is calculated as
follows:
Register = Factor * 128 = 1.2266 * 128 = 157 = 0x9D
4-10
HMP8112A
NTSC M, PAL M
PAL B, D, G, H, I, N, COMB N
LINES 1 - 22 NOT ACTIVE
LINES 1 - 22 NOT ACTIVE
ODD FIELD
SYNC AND
BACK
PORCH
240 ACTIVE LINES
PER FIELD
(LINES 23-262)
288 ACTIVE LINES
PER FIELD
(LINES 23 - 310)
VERTICAL
BLANKING
480 ACTIVE
LINES/FRAME
(NTSC, PAL M)
EVEN FIELD
LINES 263 - 284 NOT ACTIVE
240 ACTIVE LINES
PER FIELD
(LINES 285 - 524)
LINE 525
NOT ACTIVE
LINES 311 - 335 NOT ACTIVE
FRONT
PORCH
576 ACTIVE
LINES/FRAME
(PAL)
288 ACTIVE LINES
PER FIELD
(LINES 336 - 623)
NUMBER OF PIXELS
RECTANGULAR (SQUARE)
NTSC
LINES 624-625
NOT ACTIVE
PAL
TOTAL PIXELS
858 (780)
864 (944)
TOTAL PIXELS
ACTIVE PIXELS
720 (640)
720 (768)
ACTIVE PIXELS
FIGURE 13. ACTIVE VIDEO REGIONS
Output Data Port Modes
8-Bit Pixel Transfer Mode
The HMP8112A can output data in 2 formats, an 8-bit Pixel
Transfer Mode and a 16-bit Pixel Transfer mode.
For 8-Bit Pixel Transfer Mode the Y[7:0] output bus is used to
transfer all YCbCr data. The data is 4:2:2 subsampled but
will only contain the active video portion of the line. See Figure 15 for 8-Bit Pixel Transfer Mode timing. In this mode, the
data is clocked out at the CLK rate and only clock frequencies of 24.54MHz, 27MHz and 29.5MHz can be used. In 8bit Mode, the data is sequenced on the Y[7:0] bus in Cb, Y,
Cr, Y format. ACTIVE is asserted as soon as the mode is
selected. DVLD when asserted, indicates a valid active pixel
is available. Pixels during the horizontal and vertical blanking
are not available. Only the active portions of the video line
are output.
16-Bit Pixel Transfer Mode
In 16-bit Pixel Transfer Mode pixel data is output at the CLK
frequency and Table 3 shows the number of data points per
video line to expect for a given standard. Data is output as
4:2:2 subsampled data in a Y-Cb/Y-Cr 16-bit sequence. The
Data Valid (DVLD) flag is asserted when video data is
present on the 16-bit output port (Y[7:0], CbCr[7:0]). The
luminance data is output on Y[7:0] bus. Chrominance data is
sequenced on the CbCr[7:0] bus, starting with Cb and then
Cr. Per Figure 13, the ACTIVE flag is asserted when the
active video portion of the horizontal scan line is present on
the data output port. See Figure 14 for 16-Bit Pixel Transfer
Mode timing. DVLD is asserted every time the output sample
rate converter has a valid output. When DVLD and ACTIVE
are used together the visual portion of the image can be
captured. When DVLD is used alone all valid data during the
Horizontal, Vertical and Reference Burst Timing are
available.
The CLK can be run on a 20MHz - 30MHz clock source.
Data will be output (on average) at the Output Data Rate
shown in Table 3 for a given standard. Data is clocked out
synchronous to CLK and will come in bursts. To smooth out
the data output to a regular rate, a CLK of 2X the average
output data rate can be used.
TABLE 3. OUTPUT MODE STANDARDS
OUTPUT
DATA
RATE
TOTAL
PIXELS
(WITH SYNCS)
ACTIVE
PIXELS
NTSC Square Pixel
12.27MHz
780 x 525
640 x 480
NTSC CCIR 601
13.5MHz
858 x 525
720 x 480
PAL B, D, G, H, I, N,
COMB N, CCIR601
13.5MHz
864 x 625
720 x 576
PAL M CCIR 601
13.5MHz
858 x 525
720 x 480
PAL B, D, G, H, I, N
Square Pixel
14.74MHz
944 x 625
768 x 576
PAL M Square Pixel
14.74MHz
780 x 525
640 x 480
STANDARD
4-11
HMP8112A
CLK
NOTE 3
DVLD
ACTIVE
NOTE 2
YN
Y[7-0]
Y0
Y1
Y2
Y3
Y4
Cr0
Cb2
Cr2
Cb4
NOTE 1
CbCr[7-0]
CrN
Cb0
tDVLD
NOTES:
1. Y0 is the first active luminance pixel of a line. Cb0 and Cr0 are first active chrominance pixels in a line. Cb and Cr will alternate every cycle
due to the 4:2:2 subsampling. YN the last valid pixel in the blanking period.
2. ACTIVE is asserted per Figure 13.
3. DVLD is asserted for every valid pixel during both active and blanking regions. DVLD is not a 50% duty cycle synchronous output and will
appear to jitter as the Output Sample Rate converter adjusts the output timing for various data rates and clock frequency inputs.
FIGURE 14. OUTPUT TIMING 16-BIT MODE
CLK
NOTE 6
DVLD
ACTIVE
NOTE 5
Cb0
Y[7-0]
Y0
Cr0
Y1
Cb2
Y2
Cr2
Y3
Cb4
NOTE 4
tDVLD
NOTES:
4. Y0 is the first active luminance pixel of a line. Cb0 and Cr0 are first active chrominance pixels in a line. Cb and Cr will alternate every cycle
due to the 4:2:2 subsampling. Pixel data is not output during the blanking period.
5. ACTIVE stays asserted as soon as 8-Bit mode is selected.
6. DVLD is asserted for every valid pixel during the active region only per Figure 13. DVLD may deassert briefly during the active video region
as the Output Sample Rate converter adjusts the output timing for various data rates and clock frequency inputs.
FIGURE 15. OUTPUT TIMING 8-BIT MODE
4-12
HMP8112A
I 2C Control Interface
register pointer is loaded and a series of registers can be
written. If multiple registers are written, the pointer register
will autoincrement up through the register address space. A
stop cycle is used to end the transfer after the desired number of registers are programmed.
The HMP8112A utilizes an I2C control bus interface to program the internal configuration registers. This standard
mode (up to 100 KBPS) interface consists of the bidirectional
Serial Data Line (SDA) and the Serial Clock Line (SCL). The
implementation on the HMP8112A is a simple slave interface
that will not respond to general calls and cannot initiate a
transfer. The SDA and SCL control pins should be pulled
high through external 4kΩ pullup resistors to VCC.
For a read transfer, the I2C device address is the first part of
the serial transfer. Then the internal register pointer is
loaded. At this point another start cycle is initiated to access
the individual registers. Figure 18 shows the programming
flow for read transfer of the internal registers. Multiple registers can be read and the pointer register will autoincrement
up through the pointer register address space. On the last
data read, an acknowledge should not be issued. A stop
cycle is used to end the transfer after the desired number of
registers are read.
The I2C clock/data timing is shown below in Figure 16. The
HMP8112A always uses chip address 0x88. There are 28
internal registers used to program and configure the
decoder. The I2C control port contains a pointer register that
auto-increments through the entire register space and can
be written. The autoincrement pointer will wrap after the last
register has been accessed (Product ID Register) and
should be set to the desired starting address each time an
access is started. For a write transfer, the I2C device base
address is the first part of a serial transfer. Then the internal
The HMP8112A contains a product ID register that can be
used to identify the presence of a board during a Plug ’n Play
detection software algorithm. The Product ID Code register is
at sub address 0x1B and always returns a data value of 0x12.
tSU:DATA
tBUF
SDA
tHD:DATA
SCL
tLOW
tHIGH
tR
tF
tSU:STOP
FIGURE 16. I2C TIMING DIAGRAM
SDA
SCL
1-7
S
START
CONDITION
8
ADDRESS
9
R/W
1-7
ACK
8
DATA
9
P
ACK
STOP
CONDITION
FIGURE 17. I2C SERIAL DATA FLOW
DATA WRITE
1000 1000
S
CHIP ADDR
FROM MASTER
A
SUB ADDR
A
0x88
DATA READ
S
DATA
A
REGISTER
POINTED
TO BY
SUB ADDR
DATA
A
P
FROM HMP8112A
OPTIONAL FRAME
MAY BE REPEATED
n TIMES
1000 1000 (R/W)
CHIP ADDR
0x88
A
SUB ADDR
A
S
CHIP ADDR
A
0x89
DATA
REGISTER
POINTED
TO BY
SUB ADDR
A
DATA
OPTIONAL FRAME
MAY BE REPEATED
n TIMES
FIGURE 18. REGISTER WRITE/READ FLOW
4-13
NA
P
S = START CYCLE
P = STOP CYCLE
A = ACKNOWLEDGE
NA = NO ACKNOWLEDGE
HMP8112A
TABLE 4. DEFAULT REGISTER VALUES
SUB
ADDR
(HEX)
0x00
0x01
0x02
0x03
0x04
0x05
0x06
0x07
0x08
0x09
0x0A
0x0B
0x0C
0x0D
0x0E
0x0F
0x10
0x11
0x12
0x13
0x14
0x15
0x16
0x17
0x18
0x19
0x1A
0x1B
REGISTER NAME
Video Input Control
Luminance Brightness Control
Luminance Contrast Adjust
Hue Adjust
Luminance Sharpness Control
Color Saturation Adjust
PLL Clock Frequency Ratio (LSB)
PLL Clock Frequency Ratio (MSB)
HAGC Start Time (LSB)
HAGC Start Time (MSB)
HAGC End Time (LSB)
HAGC End Time (MSB)
HSYNC Start Time (LSB)
HSYNC Start Time (MSB)
HSYNC End Time (LSB)
HSYNC End Time (MSB)
PLL Adjust
PLL Sync Detect Window
DC RESTORE Start Time (LSB)
DC RESTORE Start Time (MSB)
DC RESTORE End Time (LSB)
DC RESTORE End Time (MSB)
Output Format Control
Software Reset and Video Status
Reserved
Reserved
Reserved
Product ID
DEFAULT
VALUE
(HEX)
0xF8
0x80
0xBD
0x00
0x00
0x9D
0xC1
0x87
0x3F
0x03
0x00
0x00
0x3B
0x03
0x20
0x00
0x00
0xDD
0x37
0x00
0x52
0x00
0x00
0x00
0x00
0x00
0x00
0x12
USE
VALUE
(HEX)
Comments
Defaults to NTSC and Mux = LIN0
Table 1
Table 1
Table 5
Table 5
Table 5
Table 5
Table 5
Table 5
Table 5
Table 5
0x20
0x20
Table 5
Table 5
Table 5
Table 5
Defaults to NTSC with 27MHz Clock Ratio
Recommend PLL Adjust = 0x20
Recommend PLL Sync Detect Window = 0x20
Bits 6-2 define Output Mode.
Set Bit 7 to Reset.
/
TABLE 5. RECOMMENDED TIMING REGISTER CONFIGURATION
HAGC WINDOW
(SYNC TIP)
DC RESTORE WINDOW
(BACK PORCH)
HSYNC WINDOW
(BLANKING INTERVAL)
START
END
START
END
START
END
TOTAL
PIXELS
PER
LINE
LAST
PIXEL
COUNT
(HEX)
REGISTERS
MSB/LSB
0x09/0x08
REGISTERS
MSB/LSB
0x0B/0x0A
REGISTERS
MSB/LSB
0x13/0x12
REGISTERS
MSB/LSB
0x15/0x14
REGISTERS
MSB/LSB
0x0D/0x0C
REGISTERS
MSB/LSB
0x0F/0x0E
NTSC, CCIR601
Rectangular Pixel
(720 x 480)
858
0x0359
0x033B
0x001B
0x002D
0x0048
0x033B
0x0060
NTSC
Square Pixel
(640 x 480)
780
0x030B
0x02F0
0x0014
0x0028
0x0040
0x02F0
0x0050
PAL-B, CCIR601
Rectangular Pixel
(720 x 576)
864
0x035F
0x0345
0x001A
0x0032
0x0050
0x0345
0x0070
PAL-B
Square Pixel
(768 x 576)
944
0x03AF
0x0392
0x001C
0x0044
0x0056
0x0392
0x0070
VIDEO
STANDARD
(ACTIVE PIXELS)
4-14
HMP8112A
TABLE 6. VIDEO INPUT CONTROL
SUB ADDRESS = 0x00
BIT
NUMBER
7-6
5
FUNCTION
DESCRIPTION
Video Input
Standard
These bits select the video input standard.
Color Trap Filter
Disable
This bit enables the color subcarrier trap filter. The filter removes the color subcarrier information from the luminance channel. The filter should be enabled for PAL Standard systems.
RESET
STATE
11B
00 = PAL B, G, H, I, N; 4.43MHz subcarrier; 50fps; 625 lines/frame;
01 = PAL M; 3.58MHz subcarrier; 60fps; 525 lines/frame;
10 = Special PAL N; 3.58MHz subcarrier; 50fps; 625 lines/frame;
11 = NTSC M; 3.58MHz subcarrier; 60fps; 525 lines/frame (default);
1B
0 = Enabled
1 = Disabled (default)
4
Chrominance Low
Pass Filter Disable
This bit enables the chrominance low pass filter. This filter band limits the chrominance
channel to remove luminance artifacts. This filter should be enabled for PAL Standard
systems.
1B
0 = Enabled
1 = Disabled (default)
3
Automatic Color
Gain Control
This bit enables the color AGC function. When this bit is set the color AGC will automatically adjust the chrominance channel gain, to drive the color reference burst to a nominal
±20 IRE’s. When this bit is cleared the color AGC gain factor is set to 1.0 and the color
saturation must be adjusted to obtain nominal CrCb values.
1B
0 = Disabled
1 = Enabled (default)
2-1
A/D Converter
Multiplexer Selects
These bits control the A/D input select multiplexers and whether S-Video is being input as
follows:
00B
0, 0 = Select Composite Video Input = LIN0 (Pin 7), set decoder for Composite
1, 0 = Select Composite Video Input = LIN1 (Pin 6), set decoder for Composite
0, 1 = Select Composite Video Input = LIN2 (Pin 5), set decoder for Composite
1, 1 = Select S-Video Y Input = LIN2 (Pin 5) and C Input = CIN (Pin 19)
0
Not Used
Write Ignored, Read 0’s
0B
TABLE 7. LUMINANCE BRIGHTNESS CONTROL
SUB ADDRESS = 0x01
BIT
NUMBER
7
FUNCTION
DESCRIPTION
IRE Setup
This bit enables the black setup cancellation circuit for NTSC sources. When this bit is set
Cancellation Control a value of 73 is used to strip the sync information from the video signal. When this bit is
cleared a value of 64 is used to strip the sync information.
RESET
STATE
1B
0 = subtract 64 from the luminance signal
1 = subtract 73 from the luminance signal
6-0
Luminance
Brightness Control
These bits control the brightness adjustment to the luminance channel. The brightness
adjustment value is a number that ranges from +63 to -64. This register is in the two’s
complement format, where bit 6 is the sign bit.
4-15
000 0000B
HMP8112A
TABLE 8. LUMINANCE CONTRAST ADJUST REGISTER
SUB ADDRESS = 0x02
BIT
NUMBER
7-0
FUNCTION
DESCRIPTION
Luminance Contrast This register sets the contrast adjust factor which is applied after the brightness. This valAdjust Factor
ue is multiplied by the luminance data and allows the data to be scaled from 0 to a factor
of +1.996. This 8-bit number is a fractional number as shown below:
20 2-1 2-2 2-3 2-4 2-5 2-6 2-7
The default contrast factor of 1.4766 is calculated as follows:
Register Data = Factor * 128 = 1.4766 * 128 = 189 = 0xBD
RESET
STATE
1011 1101B
(0xBD)
TABLE 9. HUE ADJUST REGISTER
SUB ADDRESS = 0x03
BIT
NUMBER
7-0
FUNCTION
Hue Phase Adjust
DESCRIPTION
This register sets the hue phase offset adjustment. This 8-bit number is applied as a
phase offset to the CbCr data coming out of the demodulator. This 8-bit number is a in
the range of +127 to -128. The hue adjust has as range of 30o with each count in this
register allowing a 0.25o phase adjustment. This register is in two’s complement format,
where bit 7 is the sign bit.
RESET
STATE
0000 0000B
(0x00)
TABLE 10. LUMINANCE SHARPNESS CONTROL REGISTER
SUB ADDRESS = 0x04
BIT
NUMBER
7-6
FUNCTION
DESCRIPTION
High Frequency
These bits adjust the amplitude of high frequency components in the luminance video
Enhancement Factor signal. The attenuation or multiplication of the high frequency components is adjusted as
shown below:
RESET
STATE
00B
00 = Multiply high frequency components by 1.0
01 = Multiply high frequency components by 2.0
10 = Multiply high frequency components by 4.0
11 = Zero out high frequency components.
5-4
Middle Frequency
Enhancement
Factor
These bits adjust the amplitude of middle frequency components in the luminance video
signal. The attenuation or multiplication of the middle frequency components is adjusted
as shown below:
00B
00 = Multiply middle frequency components by 1.0
01 = Multiply middle frequency components by 2.0
10 = Multiply middle frequency components by 4.0
11 = Zero out middle frequency components.
3-2
Low Frequency
Enhancement
Factor
These bits adjust the amplitude of low frequency components in the luminance video signal. The attenuation or multiplication of the low frequency components is adjusted as
shown below:
00B
00 = Multiply low frequency components by 1.0
01 = Multiply low frequency components by 2.0
10 = Multiply low frequency components by 4.0
11 = Zero out low frequency components.
1-0
Vertical High
Frequency
Enhancement
Factor
These bits adjust the amplitude of vertical high frequency components in the luminance
video signal. The attenuation or multiplication of the vertical high frequency components
is adjusted as shown below:
00 = Multiply vertical high frequency components by 1.0
01 = Reserved.
10 = Reserved.
11 = Zero out vertical high frequency components.
4-16
00B
HMP8112A
TABLE 11. COLOR SATURATION ADJUST FACTOR
SUB ADDRESS = 0x05
BIT
NUMBER
7-0
FUNCTION
Color Saturation
Adjust Factor
DESCRIPTION
This register sets the color saturation adjust factor. This value is multiplied by the chrominance (CbCr) data and allows the data to be scaled from 0 to a factor of +1.996. This 8-bit
number is a fractional number as shown below:
RESET
STATE
1001 1101B
(0x9D)
20 2-1 2-2 2-3 2-4 2-5 2-6 2-7
The default saturation factor of 1.2266 is calculated as follows:
Register Data = Factor * 128 = 1.2266 * 128 = 157 = 0x9D
TABLE 12. PLL CLOCK FREQUENCY RATIO (LSB)
SUB ADDRESS = 0x06
BIT
NUMBER
7-0
FUNCTION
PLL Clock
Frequency Ratio
(LSB)
DESCRIPTION
These bits are used to program the ratio of the incoming video chrominance color subcarrier frequency to the input clock (CLK) used. This number serves as the reference frequency of the chrominance PLL. This is the lower byte (LSB) of the ratio and
encompasses the following range:
RESET
STATE
1100 0001B
(0xC1)
2-9 2-10 2-11 2-12 2-13 2-14 2-15 2-16
The default value is for a CLK frequency of 27MHz and a color subcarrier of 3.579545
MHz. The register data is calculated as follows:
Ratio =(4 x fSC) / CLK
=(4 x 3.579545MHz) / 27MHz
=0.530303
Register Data: Ratio * 65536
0.530303 * 65536
34753.94
Convert to Hex:0x87C1
Reg 0x06 LSB =0xC1
Reg 0x07 MSB =0x87
Refer to Table 1 for common PLL Ratio values with CLKs of 27MHz or 24.54Hz
TABLE 13. PLL CLOCK FREQUENCY RATIO (MSB)
SUB ADDRESS = 0x07
BIT
NUMBER
15 - 8
DESCRIPTION
RESET
STATE
This is the upper data byte (MSB) of the PLL Clock Freq as described in Reg 0x06 above
and encompasses the following range:
1000 0111B
(0x87)
FUNCTION
PLL Clock
Frequency Ratio
(MSB)
2-1 2-2 2-3 2-4 2-5 2-6 2-7 2-8
4-17
HMP8112A
TABLE 14. HAGC START TIME (LSB) REGISTER
SUB ADDRESS = 0x08
BIT
NUMBER
7-0
FUNCTION
HAGC
START Time (LSB)
DESCRIPTION
This register provides a programmable delay for the HAGC pulse that control the sync
tip AGC in the A/D converters. This is the lower byte of the 10-bit word.
RESET
STATE
0011 1111B
(0x3F)
TABLE 15. HAGC START TIME (MSB) REGISTER
SUB ADDRESS = 0x09
BIT
NUMBER
15 - 10
9-8
FUNCTION
DESCRIPTION
Not Used
Write Ignored, Read 0’s.
HAGC
START Time (MSB)
This register provides a programmable delay for the HAGC pulse that control the sync
tip AGC in the A/D converters. This is the upper byte of the 10-bit word.
RESET
STATE
0000 0011B
(0x03)
TABLE 16. HAGC END TIME (LSB) REGISTER
SUB ADDRESS = 0x0A
BIT
NUMBER
7-0
FUNCTION
HAGC
END Time (LSB)
DESCRIPTION
This register provides a programmable delay for the HAGC pulse that control the sync
tip AGC in the A/D converters. This is the lower byte of the 10-bit word.
RESET
STATE
0000 0000B
(0x00)
TABLE 17. HAGC END TIME (MSB) REGISTER
SUB ADDRESS = 0x0B
BIT
NUMBER
15 - 10
9-8
FUNCTION
DESCRIPTION
Not Used
Write Ignored, Read 0’s
HAGC
END Time (MSB)
This register provides a programmable delay for the HAGC pulse that controls the sync
tip AGC in the A/D converters. This is the upper byte of the 10-bit word.
RESET
STATE
0000 0000B
(0x00)
TABLE 18. HSYNC START TIME (LSB) REGISTER
SUB ADDRESS = 0x0C
BIT
NUMBER
7-0
FUNCTION
DESCRIPTION
RESET
STATE
HSYNC Pulse
START Time (LSB)
This register provides a programmable delay for the external HSYNC pulse. This is the
lower byte of the 10-bit word.
0011 1011B
(0x3B)
TABLE 19. HSYNC START TIME (MSB) REGISTER
SUB ADDRESS = 0x0D
BIT
NUMBER
15 - 10
9-8
FUNCTION
Not Used
DESCRIPTION
RESET
STATE
Write Ignored, Read 0’s
This register provides a programmable delay for the external HSYNC pulse. This is the
HSYNC Pulse
START Time (MSB) upper byte of the 10-bit word.
4-18
0000 0011B
(0x03)
HMP8112A
TABLE 20. HSYNC END TIME (LSB) REGISTER
SUB ADDRESS = 0x0E
BIT
NUMBER
7-0
DESCRIPTION
RESET
STATE
This register provides a programmable delay for the external HSYNC pulse. This is the
lower byte of the 10-bit word.
0010 0000B
(0x20)
FUNCTION
HSYNC Pulse
END Time (LSB)
TABLE 21. HSYNC END TIME (MSB) REGISTER
SUB ADDRESS = 0x0F
BIT
NUMBER
15 - 10
9-8
FUNCTION
DESCRIPTION
Not Used
Write Ignored, Read 0’s
HSYNC Pulse
END Time (MSB)
This register provides a programmable delay for the external HSYNC pulse. This is the
upper byte of the 10-bit word.
RESET
STATE
0000 0000B
(0x00)
TABLE 22. PLL FILTER ADJUST REGISTER
SUB ADDRESS = 0x10
BIT
NUMBER
7-0
DESCRIPTION
RESET
STATE
The Phase Locked Loop (PLL) time constants can be changed for testing purposes. It is
recommended that the default value of 0x20 always be used. The reset state is 0x00.
0000 0000B
(0x00)
FUNCTION
PLL Filter Adjust
Register
TABLE 23. PLL SYNC DETECT WINDOW REGISTER
SUB ADDRESS = 0x11
BIT
NUMBER
7-0
FUNCTION
DESCRIPTION
PLL Horizontal Sync These bits control the PLL horizontal sync detect window. This window sets the length
Detect Window
of time that the line lock PLL will allow the detection of the HSYNC. HSYNC outside of
this window are declared missing and will cause the missing sync logic to start counting
missing syncs.
RESET
STATE
1101 1101B
(0xDD)
TABLE 24. DC RESTORE START TIME (LSB) REGISTER
SUB ADDRESS = 0x12
BIT
NUMBER
7-0
FUNCTION
DESCRIPTION
RESET
STATE
DC RESTORE
START Time (LSB)
This register provides a programmable delay for the internal DC RESTORE signal. This
is the lower byte of the 10-bit word.
0011 0111B
(0x3F)
TABLE 25. DC RESTORE START TIME (MSB) REGISTER
SUB ADDRESS = 0x13
BIT
NUMBER
15 - 10
9-8
FUNCTION
Not Used
DESCRIPTION
RESET
STATE
Write Ignored, Read 0’s
DC RESTORE
This register provides a programmable delay for the internal DC RESTORE signal. This
START Time (MSB) is the upper byte of the 10-bit word.
4-19
0000 0000B
(0x00)
HMP8112A
TABLE 26. DC RESTORE END TIME (LSB) REGISTER
SUB ADDRESS = 0x14
BIT
NUMBER
7-0
DESCRIPTION
RESET
STATE
This register provides a programmable delay for the internal DC RESTORE signal. This
is the lower byte of the 10-bit word.
0101 0010B
(0x52)
FUNCTION
DC RESTORE
END Time (LSB)
TABLE 27. DC RESTORE END TIME (MSB) REGISTER
SUB ADDRESS = 0x15
BIT
NUMBER
15 - 10
9-8
FUNCTION
DESCRIPTION
Not Used
Write Ignored, Read 0’s
DC RESTORE
END Time (MSB)
This register provides a programmable delay for the internal DC RESTORE signal. This
is the upper byte of the 10-bit word.
RESET
STATE
0000 0000B
(0x00)
TABLE 28. OUTPUT FORMAT CONTROL REGISTER
SUB ADDRESS = 0x16
BIT
NUMBER
7
6, 5, 4
FUNCTION
DESCRIPTION
RESET
STATE
Square Pixel/ITU-R
BT601 Select
When “1”, Square pixel output is selected, when “0” ITU-R BT601 output rate is selected.
0B
Output Field Control These bits control the field capture rate of the HMP8112A. The user can select every 4th
“FLD_CONT(2-0)”
field, every other field or every field of video to be output to the data port.
000B
000 = No Capture Enabled
001 = Capture every 4th field
010 = Capture every 2nd field
011 = Capture every 2nd odd field
100 = Capture every 2nd even field
101 = Capture every odd field
110 = Capture every even field
111 = Capture all fields
3
8/16 output Select
When “1”, the 8-bit Burst Transfer output mode is selected. When “0”, the 16-bit Synchronous Pixel Transfer output mode is selected.
0B
2
OEN
This bit enables the Y(7-0), CbCr(7-0), ACTIVE, FIELD, HSYNC, VSYNC and DVLD outputs. 1 = Outputs enabled; 0 = three-stated.
0B
1
Vertical Pixel Siting
When this bit is cleared (‘0’) the chrominance pixels have a 1/2 line pixel offset from their
associated luminance pixel in a 4:2:2 subsampled scheme. When this bit is set (‘1’) the
pixel siting is line aligned with the luminance pixels in a 4:2:2 subsampled scheme. The
bit is cleared by a RESET.
0B
0
Not Used
Write Ignored, Read 0’s
0B
4-20
HMP8112A
TABLE 29. SOFTWARE RESET AND VIDEO STATUS REGISTER
SUB ADDRESS = 0x17
BIT
NUMBER
FUNCTION
DESCRIPTION
RESET
STATE
7
Software Reset
When this bit is set to 1, the entire device except the I2C bus is reset to a known state
exactly like the RESET input. The software reset will initialize all register bits to their reset
state. Once set this bit is self clearing after only 4 CLK periods. This bit is cleared on power-up by the external RESET pin.
0B
6
Black Screen
This flag when set (‘1’) will set the output video to black when a lost vertical sync has
been detect. This flag is cleared after a RESET.
0B
5
Line LOCKED Flag
This flag when set (‘1’) indicates that the Line Locked-Phase Locked Loop has locked to
the video data. This flag is read only and cleared after a RESET or Software Reset.
0B
4
Standard Error Flag This flag when set (‘1’) indicates that the Standard detected does not match the one selected in the Video Input Control Register. The standard is checked against a line count
and if the line count is significantly different than the expected value then this flag is triggered. This flag is read only and cleared after a RESET or Software Reset.
0B
3-0
Not Used
Write ignored, Read 0’s.
0000B
TABLE 30. RESERVED
SUB ADDRESS = 0x18
BIT
NUMBER
7-0
FUNCTION
Reserved
DESCRIPTION
This register is reserved. This register will read all zero’s and is write ignored.
RESET
STATE
0000 0000B
TABLE 31. RESERVED
SUB ADDRESS = 0x19
BIT
NUMBER
7-6
5
4-0
FUNCTION
DESCRIPTION
RESET
STATE
Reserved
This register is reserved. This register will read all zero’s and is write ignored.
00B
Lost HSYNC
Control (SNAP Bit)
This bit controls when the PLL will declare lost horizontal sync, leave track mode and return to acquisition to acquire a new HSYNC reference. When this bit is cleared, lost line
lock is declared after 12 missing horizontal syncs. When this bit is set, lost line lock is
declared after one missing horizontal sync. This bit is cleared by RESET.
0B
Reserved
This register is reserved. This register will read all zero’s and is write ignored.
0 0000B
TABLE 32. RESERVED
SUB ADDRESS = 0x1A
BIT
NUMBER
7-0
FUNCTION
Reserved
DESCRIPTION
This register is reserved. This register will read all zero’s and is write ignored.
RESET
STATE
0000 0000B
TABLE 33. PRODUCT ID REGISTER
SUB ADDRESS = 0x1B
BIT
NUMBER
7-0
DESCRIPTION
RESET
STATE
This register contains the last two digits of the product part number for use as a software
ID. These bits are read only and always read 0x12.
0001 0010B
(0x12)
FUNCTION
Product ID Code
4-21
HMP8112A
Pinout
GND
VCC
DEC_T
LAGC_CAP
LCLAMP_CAP
VCC
NC
NC
GND
HSYNC
VSYNC
GND
VCC
FIELD
DVLD
ACTIVE
80 LEAD PQFP
TOP VIEW
80 79 78 77 76 75 74 73 72 71 70 69 68 67 66 65
AGND
VAA
AGND
NC
LIN2
LIN1
LIN0
L_ADIN
L_OUT
AGND
AGND
VAA
CLK
VAA
AGND
AGND
A/D_TEST
NC
CIN
NC
AGND
AGND
AGND
AGND
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
64
63
62
61
60
59
58
57
56
55
54
53
52
51
50
49
48
47
46
45
44
43
42
41
WPE
GAIN_CNTL
CCLAMP_CAP
DEC_L
VCC
NC
GND
RESET
GND
TEST
VCC
CLK
GND
SDA
GND
VCC
25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
NOTE:
7. Refer to Pin Description for this pin.
4-22
Y7
Y6
GND
NC (NOTE 7)
Y5
VCC
Y4
Y3
Y2
Y1
Y0
GND
VCC
CbCr7
CbCr6
CbCr5
CbCr4
CbCr3
GND
CbCr2
NC (NOTE 7)
CbCr1
CbCr0
SCL
HMP8112A
Pin Description
NAME
PQFP PIN
NUMBER
INPUT/
OUTPUT
LIN0
7
Input
Composite video input. This input must be AC-coupled to the video signal using a
1.0 µF capacitor and terminated with a 75-ohm resistor. These components should
be as close to this pin as possible for best performance. If not used, this pin should
be connected to AGND thru a 0.1 µF capacitor.
LIN1
6
Input
Composite video input. This input must be AC-coupled to the video signal using a
1.0 µF capacitor and terminated with a 75-ohm resistor. These components should
be as close to this pin as possible for best performance. If not used, this pin should
be connected to AGND thru a 0.1 µF capacitor.
LIN2
5
Input
Composite video or Luminance (Y) video input. This input must be AC-coupled to the
video signal using a 1.0 µF capacitor and terminated with a 75-ohm resistor. These
components should be as close to this pin as possible for best performance. If not
used, this pin should be connected to AGND thru a 0.1 µF capacitor.
CIN
19
Input
Chrominance (C) video Input. This input must be AC-coupled to the video signal using a 1.0 µF capacitor and terminated with a 75-ohm resistor. These components,
and corresponding anti-aliasing low-pass filter, should be as close to this pin as possible for best performance. If not used, this pin should be connected to AGND thru a
0.1 µF capacitor.
WPE
27
Input
White Peak Enable. When enabled (‘1’), the video amplifiers gain is reduced when
the digital output code exceeds 248. When disabled (‘0’) the video amplifier will clip
when the A/D reaches code 255.
GAIN_CTRL
28
Input
Gain Control Input. DC voltage to set the S-Video CIN chrominance video amplifier’s
gain. Reference Figure 3 for gain control curve.
DEC_T
78
Input
Decoupling for upper A/D Converter Reference. Recommend connecting 0.1 µF and
0.01µF ceramic capacitors in parallel to AGND.
DEC_L
30
Input
Decoupling for lower A/D Converter Reference. Recommend connecting 0.1 µF and
0.01µF ceramic capacitors in parallel to AGND.
LAGC_CAP
77
Input
Capacitor Connection for Luminance AGC Circuit. Controls the AGC loop time constant. Recommend connecting a 0.01 µF ceramic capacitor to AGND.
LCLAMP_CAP
76
Input
Capacitor Connection for Luminance Clamp Circuit. Controls the clamp loop time
constant. Recommend connecting a 0.047 µF ceramic capacitor to AGND.
CCLAMP_CAP
29
Input
Capacitor Connection for Chrominance Clamp Circuit. Controls the clamp loop time
constant. Recommend connecting a 0.047 µF ceramic capacitor to AGND.
L_ADIN
8
Input
Luminance A/D Converter input from external anti-alias filter. Reference Figure 1.
L_OUT
9
Output
Analog output of the video multiplexer. This output should connect to an external
anti-alias filter and return to L_ADIN input. Reference Figure 1.
SDA
40
Input/
Output
The serial I2C serial input/output data line.
SCL
41
Input
The serial I2C serial bus clock line.
CLK
13, 38
Input
Master clock for the decoder. This clock is used to run the internal logic, A/D converters, and Phase Locked Loops. All I/O pins (except the I2C) are synchronous to this
master clock. A ±50ppm crystal should be used with a waveform symmetry of
60/40% or better.
RESET
34
Input
Asynchronous Reset pin. Master Chip reset to initialize the internal states and set
the internal registers to a known state.
DESCRIPTION
4-23
HMP8112A
Pin Description
(Continued)
PQFP PIN
NUMBER
INPUT/
OUTPUT
CbCr[0:7]
42, 43, 45,
47-51
Output
CbCr Data Output Port. The chrominance data output port of the decoder. Data is in
unsigned format and can range from 0 to 255. The CbCr data is subsampled to 4:2:2
format. In 4:2:2 format the CbCr bus toggles between Cb and Cr samples with the
first sample of a line always being Cb. The port is designed to minimize external logic
needed to interface to a VRAM Serial Access Port, DRAM or FIFO.
Y[0:7]
54-58, 60, 63,
64
Output
Y Data Output Port. The luminance data output port of the decoder. Data is in unsigned format and can range from 16 to 255. The port is designed to minimize external logic needed to interface to a VRAM Serial Access Port, DRAM or FIFO.
DVLD
66
Output
Data Valid. This pin signals when valid data is available on the data output ports. This
pin is three-stated after a RESET or software reset and should be pulled high through
a 10K resistor.
HSYNC
71
Output
Horizontal Sync. This video synchronous pulse is generated by the detection of horizontal sync on the video input. In the absence of video, the HSYNC rate is set when
the internal PLL counters overflow. The HSYNC START and END time can be programmed. This pin is three-stated after a RESET or software reset and should be
pulled high through a 10K resistor.
VSYNC
70
Output
Vertical Sync. This video synchronous pulse is generated by the detection of a vertical
sync on the video input. In the absence of video the VSYNC rate is set by the over flow
of the internal line rate counter. This pin is three-stated after a RESET or software reset
and should be pulled high through a 10K resistor.
FIELD
67
Output
Field Flag. When set (‘0’) this signals that an ODD field is presently being output from
the decoder. When cleared (‘1’) this signals an EVEN field. This flag will toggle when
no vertical sync is detected and 337 lines have elapsed. This pin is three-stated after
a RESET or software reset and should be pulled high through a 10K resistor.
ACTIVE
65
Output
Active Video Flag. This flag is asserted (‘1’) when the active portion of the video line
is available on the output port. This signal is always set during Burst Output data
mode. This flag is free running and synchronous to CLK. This pin is three-stated after
a RESET or software reset and should be pulled high through a 10K resistor.
TEST
36
Input
Test input. This pin is used for production test and should be connected to digital
ground.
VCC
26, 31, 37, 52,
59, 68, 75, 79
Input
5V Logic Supply Pins
GND
25, 33, 35, 39,
46, 53, 62, 69,
72, 80
Input
Digital Ground Pins
VAA
2, 12,14
Input
5V Analog Supply Pins
AGND
1, 3, 10, 11,
15,16, 21, 22,
23, 24
Input
Analog GND
A/D TEST
17
Output
NC
44, 61
NA
Pins used as logic outputs on later decoders. Refer to HMP8115 data sheet for details.
NC
4, 18, 20, 32,
73, 74
NA
No Connect. These pins should be left open.
NAME
DESCRIPTION
Chrominance ADC Test Pin. This pin should be left open.
4-24
HMP8112A
Absolute Maximum Ratings
Thermal Information
Digital Supply Voltage (VCC to GND) . . . . . . . . . . . . . . . . . . . . 7.0V
Digital Input Voltages . . . . . . . . . . . . . . . . . GND -0.5V to VCC 0.5V
ESD Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Class 1
Thermal Resistance (Typical, See Note 8)
Operating Conditions
Temperature Range, HMP8112ACN . . . . . . . . . . . . . . 0oC to 70oC
θJA (oC/W)
PQFP Package . . . . . . . . . . . . . . . . . . . . . . . . . . . .
42
Maximum Power Dissipation
HMP8112ACN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.9W
Maximum Storage Temperature Range . . . . . . . . . .-65oC to 150oC
Maximum Junction Temperatures . . . . . . . . . . . . . . . . . . . . . 150oC
Maximum Lead Temperature (Soldering 10s) . . . . . . . . . . . . 300oC
CAUTION: Stresses above those listed in “Absolute Maximum Ratings” may cause permanent damage to the device. This is a stress only rating and operation of
the device at these or any other conditions above those indicated in the operational sections of this specification is not implied.
NOTE:
8. θJA is measured with the component mounted on an evaluation PC board in free air. Dissipation rating assumes device is mounted with
all leads soldered to printed circuit board
Electrical Specifications
VCC = 5.0V, TA = 25oC
HMP8112AC
PARAMETER
SYMBOL
TEST CONDITION
MIN
TYP
MAX
UNITS
4.75
5
5.25
V
POWER SUPPLY CHARACTERISTICS
Power Supply Voltage Range
VCC, VAA
Note 9
Digital ICCOP
fCLK = 30MHz,
VCC = 5.25V,
Outputs Not Loaded
-
45
60
mA
Analog ICAOP
fCLK = 30MHz,
VAA = 5.25V
-
190
220
mA
PTOT
fCLK = 30MHz,
VCC = VAA = 5.25,
Outputs Not Loaded
-
1.11
1.47
W
Bus Clock Frequency
CLK
Note 9
20
-
30
MHz
Clock Cycle Time
CLK
33
-
50
ns
40
-
60
%
Power Supply Current
Total Power Dissipation
DIGITAL I/O
Clock Waveform Symmetry
Clock Pulse Width High
tPWH
8
-
-
ns
Clock Pulse Width
tPWL
13
-
-
ns
Input Logic High Voltage
VIH CLK
VCC = Max
2.8
-
-
V
Input Logic Low Voltage
VIL CLK
VCC = Min
-
-
0.8
V
VCC = Max
Input = 0V or VCC
-
-
10
µA
-450
-
-
µA
Input Leakage Current
IIH
IIL
Input/Output Capacitance
CIN
CLK Frequency = 1MHz,
Note 9, All Measurements
Referenced to Ground TA = 25oC
-
-
8
pF
Rise/Fall Time
tr, tf
Note 9
-
-
2.0
ns
Input Logic High Voltage
VIH
VCC = Max
2.0
-
-
V
Input Logic Low Voltage
VIL
VCC = Min
-
-
0.8
V
VCC = Max
Input = 0V or 5V
-
-
±10
µA
Input Logic Current
IIH, IIL
4-25
HMP8112A
Electrical Specifications
VCC = 5.0V, TA = 25oC (Continued)
HMP8112AC
PARAMETER
SYMBOL
TEST CONDITION
MIN
TYP
MAX
UNITS
2.4
-
-
V
Output Logic High Voltage
VOH
IOH = -4mA, VCC = Max
Output Logic Low Voltage
VOL
IOL = 4mA, VCC = Min
-
-
0.4
V
Output Logic Current
IOH
VCC = Max, Input = 0V or 5V
-
-
4
mA
Three-State Output Current Leakage
IOZ
-
-
10
µA
I2C DIGITAL I/O
(SDA, SCL, Fast Mode)
Input Logic High Voltage
VIH
VCC = Max
0.7x
VCC
-
-
V
Input Logic Low Voltage
VIL
VCC = Min
-
-
0.3xVCC
V
VCC = Max
Input = 0V or 5V
-
-
10
µA
-
-
8
pF
3.0
-
-
V
Input Logic Current
IIH, IIL
Input/Output Capacitance
CIN
CLK Frequency = 400kHz,
Note 9, All Measurements
Referenced to GND
TA = 25oC
Output Logic High Voltage
VOH
IOH = -1mA, VCC = Max
Output Logic Low Voltage
VOL
IOL = 3mA, VCC = Min
0
-
0.4
V
SCL Clock Frequency
fSCL
Note 9
0
-
100
kHz
SCL Minimum Low Pulse Width
tLOW
4.7
-
-
µs
SCL Minimum High Pulse Width
tHIGH
4.0
-
-
µs
Data Hold Time
tHD:DATA
0
-
-
ns
Data Setup Time
tSU:DATA
250
-
-
ns
-
-
1000
ns
-
-
300
ns
10
-
-
ns
Rise Time
tR
Fall Time
tF
Note 9
TIMING CHARACTERISTICS
Data Setup Time
tSU
Data Hold Time
tHD
0
-
-
ns
tDVLD
-
-
8.0
ns
Input Termination of 75Ω and
1.0µF AC Coupling, Note 9
0.5
1.0
2.0
VP-P
Note 9
200
-
-
kΩ
-6
-
+18
dB
1VP-P Sine Wave Input to
-3dBc Reduction, Note 9
-
15
-
MHz
Note 9
-
VAA - 1.9
-
V
AIN Offset/Zero
-
VAA -3.4
-
V
BA/D
6
-
-
MHz
Clock to Out
Notes 9, 10
ANALOG PERFORMANCE
Video Input Amplifier Voltage Range
Video Input Amplifier Impedance
Color Sub-carrier AGC Range
Video Input Amplifier Analog
Bandwidth
A/D Input Range
A/D Input Bandwidth
VLIN[0:2], VCIN
RAIN
SCAGC
B
AIN Full Scale
4-26
HMP8112A
Electrical Specifications
VCC = 5.0V, TA = 25oC (Continued)
HMP8112AC
PARAMETER
SYMBOL
TEST CONDITION
MIN
TYP
MAX
UNITS
-
2
-
%
-
1
-
o
-
2
-
LSB
-
0.35
1.0
LSB
Note 9
-
49.9
-
dB
In Composite Input Mode,
Note 9
-
40
-
dB
-
40
-
dB
Time from Initial Lock
Acquisition to an Error of
1 Pixel, Note 9
-
30
-
Lines
Note 9
-
-
12
#
-
-
3
#
400
-
-
Hz
-
1/ 8
-
Pixel
-
10
-
ns
Color Saturation Adjustment Range
-
-
10
dB
Hue Accuracy
-
-
2
oC
Hue Adjustment Range
-
-
30
oC
Brightness Adjustment Range
-
-
10
dB
VIDEO PERFORMANCE
Differential Gain
AV DIFF
Differential Phase
Θ DIFF
Integral Linearity
INL
Differential Linearity
DNL
SNR
SNRL WEIGHTED
Luminance to Chrominance Crosstalk
XLUMA
Chrominance to Luminance Crosstalk
XCHROMA
Horizontal Locking and Recovery Time
tLOCK
# of Missing Horizontal Syncs Before
Lost Lock Declared
HSYNC LOST
# of Missing Vertical Syncs Before Lost
Lock Declared
VSYNC LOST
EBU 75% Color Bars, Note 9
Best Fit Linearity
Subcarrier Lock in Range
Pixel Jitter
NOTES:
9. Guaranteed by design or characterization.
10. Test performed with CL = 40pF, IOL = 4mA, IOH = -4mA. Input reference level is 1.5V for all inputs. VIH = 3.0V, VIL = 0V.
4-27
HMP8112A
Typical Performance Curves
NTSC Composite Phase
FIGURE 19. COLOR BARS NTSC 100% (EIA)
FIGURE 20. COLOR BARS VECTORSCOPE
4-28
HMP8112A
Typical Performance Curves
(Continued)
NTSC Composite Phase (Continued)
FIGURE 21. COLOR BARS VM700 TEST
FIGURE 22. DIFFERENTIAL PHASE AND GAIN
4-29
HMP8112A
Typical Performance Curves
(Continued)
NTSC Frequency Response
FIGURE 23. MULTIBURST
FIGURE 24. MULTIBURST VM700 FREQUENCY ROLL-OFF TEST
4-30
HMP8112A
Typical Performance Curves
(Continued)
NTSC Noise Measurements
FIGURE 25. SIGNAL TO NOISE RATIO - FLAT FREQUENCY RESPONSE
FIGURE 26. SIGNAL TO NOISE RATIO - 5.0MHz LOW PASS FILTERED
4-31
HMP8112A
Typical Performance Curves
(Continued)
NTSC Noise Measurements (Continued)
FIGURE 27. SIGNAL TO NOISE RATIO - 4.2MHz LOW PASS FILTERED
Pixel Jitter Test
FIGURE 28. LONG TERM JITTER - 20 PULSE BAR 2T
4-32
HMP8112A
Typical Performance Curves
(Continued)
PAL Composite Phase
FIGURE 29. COLOR BARS NTSC 100% (EIA)
FIGURE 30. COLOR BARS VECTORSCOPE
4-33
HMP8112A
Typical Performance Curves
PAL Composite Phase
(Continued)
(Continued)
FIGURE 31. COLOR BARS VM700 TEST
FIGURE 32. DIFFERENTIAL PHASE AND GAIN
4-34
HMP8112A
Typical Performance Curves
(Continued)
PAL Frequency Response
FIGURE 33. MULTIBURST
FIGURE 34. NTSC MULTI-TEST PATTERN
4-35
HMP8112A
Typical Performance Curves
(Continued)
FIGURE 35. NTSC CONVERGENCE TEST PATTERN
FIGURE 36. NTSC MULTIBURST TEST PATTERN
4-36
HMP8112A
Typical Performance Curves
(Continued)
FIGURE 37. NTSC SMPTE COLORBARS TEST PATTERN
FIGURE 38. PAL CONVERGENCE TEST PATTERN
4-37
HMP8112A
Typical Performance Curves
(Continued)
FIGURE 39. PAL MULTIBURST TEST PATTERN
FIGURE 40. PAL SMPTE COLORBARS TEST PATTERN
4-38
HMP8112A
PCB Layout Considerations
A PCB board with a minimum of 4 layers is recommended,
with layers 1 and 4 (top and bottom) for signals and layers 2
and 3 for power and ground. The PCB layout should
implement the lowest possible noise on the power and
ground planes by providing excellent decoupling. PCB trace
lengths between groups of VCC and GND pins should be as
short as possible.
If a separate linear regulator is used to provide power to the
analog power plane, the power-up sequence should be
designed to ensure latchup will not occur. A separate linear regulator is recommended if the power supply noise on the VAA
pins exceeds 200mV.
The optimum layout places the HMP8112A as close as possible to the power supply connector and the video output
connector.
Traces containing digital signals should not be routed over,
under, or adjacent to the analog output traces to minimize
crosstalk. If this is not possible, coupling can be minimized
by routing the digital signals at a 90 degree angle to the analog signals. The analog input traces should also not overlay
the VAA power plane to maximize high-frequency power supply rejection.
Component Placement
External components should be positioned as close as possible to the appropriate pin, ideally such that traces can be
connected point to point. Chip capacitors are recommended
where possible, with radial lead ceramic capacitors the second-best choice.
Power supply decoupling should be done using a 0.1µF
ceramic capacitor in parallel with a 0.01µF chip capacitor for
each group of VAA and VCC pins to ground. These capacitors should be located as close to the power and ground pins
as possible, using short, wide traces.
Digital Ground Plane
All GND pins on the HMP8112A should be connected to the
digital ground plane of the board.
Analog Ground Plane
A separate analog ground plane for the HMP8112A is recommended. All AGND pins on the HMP8112A should be
connected to the analog ground plane. This analog ground
plane should be connected to the board’s digital ground
plane at a single point.
Analog Signals
Evaluation Boards
The HMP8156EVAL2 stand-alone evaluation board allows
connecting the NTSC/PAL decoder into an IBM PC ISA slot
for evaluation. The board contains the HMP8112A
NTSC/PAL decoder, 2 Mbytes of VRAM and a encoder. The
board can accept Composite or S-Video input and display
video on a stand composite or S-Video display. The ISA bus
and Windows 95 evaluation software allows easy plug and
play of the decoder for analysis with such tools as a VM700
video test system.
Related Application Notes
Application Notes are also available on the Harris Multimedia
web site at:
http://www.semi.harris.com/datasheets/mmedia.
AN9644: Composite Video Separation Techniques
AN9716: Widescreen Signalling
Analog Power Plane
AN9717: YCbCr to RGB Considerations
The HMP8112A should have its own VAA power plane that is
isolated from the common power plane of the board, with a
gap between the two power planes of at least 1/8 inch. All
VAA pins on the HMP8112A must be connected to this
analog power plane. The analog power plane should be
connected to the board’s normal VCC power plane at a
single point though a low-resistance ferrite bead, such as a
Ferroxcube 5659065-3B, Fair-Rite 2743001111, or TDK
BF45-4001. The ferrite bead provides resistance to
switching currents, improving the performance of
HMP8112A. A single 47µF capacitor should also be used
between the analog power plane and the ground plane to
control low-frequency power supply ripple.
AN9728: BT.656 Video Interface for ICs
AN9738: VMI Video Interface for ICs
4-39
HMP8112A
Metric Plastic Quad Flatpack Packages (MQFP/PQFP)
Q80.14x20 (JEDEC MO-108CB-1 ISSUE A)
80 LEAD METRIC PLASTIC QUAD FLATPACK PACKAGE
D
D1
-D-
-B-
-AE E1
e
PIN 1
SEATING
A PLANE
-H-
0.10
0.004
0.40
0.016 MIN
-C-
5o-16o
0.20
A-B S
0.008 M C
0o MIN
A2 A1
0o-7o
L
5o-16o
INCHES
MILLIMETERS
SYMBOL
MIN
MAX
MIN
MAX
NOTES
A
-
0.134
-
3.40
-
A1
0.010
-
0.25
-
-
A2
0.100
0.120
2.55
3.05
-
B
0.012
0.018
0.30
0.45
6
B1
0.012
0.016
0.30
0.40
-
D
0.904
0.923
22.95
23.45
3
D1
0.783
0.791
19.90
20.10
4, 5
E
0.667
0.687
16.95
17.45
3
E1
0.547
0.555
13.90
14.10
4, 5
L
0.026
0.037
0.65
0.95
N
80
80
7
e
0.032 BSC
0.80 BSC
-
ND
24
24
-
NE
16
16
Rev. 0 1/94
NOTES:
D S
1. Controlling dimension: MILLIMETER. Converted inch
dimensions are not necessarily exact.
B
2. All dimensions and tolerances per ANSI Y14.5M-1982.
B1
3. Dimensions D and E to be determined at seating plane -C- .
4. Dimensions D1 and E1 to be determined at datum plane
-H- .
0.13/0.17
0.005/0.007
5. Dimensions D1 and E1 do not include mold protrusion.
Allowable protrusion is 0.25mm (0.010 inch) per side.
BASE METAL
WITH PLATING
6. Dimension B does not include dambar protrusion. Allowable
dambar protrusion shall be 0.08mm (0.003 inch) total.
0.13/0.23
0.005/0.009
7. “N” is the number of terminal positions.
All Harris Semiconductor products are manufactured, assembled and tested under ISO9000 quality systems certification.
Harris Semiconductor products are sold by description only. Harris Semiconductor reserves the right to make changes in circuit design and/or specifications at
any time without notice. Accordingly, the reader is cautioned to verify that data sheets are current before placing orders. Information furnished by Harris is
believed to be accurate and reliable. However, no responsibility is assumed by Harris or its subsidiaries for its use; nor for any infringements of patents or other
rights of third parties which may result from its use. No license is granted by implication or otherwise under any patent or patent rights of Harris or its subsidiaries.
Sales Office Headquarters
For general information regarding Harris Semiconductor and its products, call 1-800-4-HARRIS
NORTH AMERICA
Harris Semiconductor
P. O. Box 883, Mail Stop 53-210
Melbourne, FL 32902
TEL: 1-800-442-7747
(407) 729-4984
FAX: (407) 729-5321
EUROPE
Harris Semiconductor
Mercure Center
100, Rue de la Fusee
1130 Brussels, Belgium
TEL: (32) 2.724.2111
FAX: (32) 2.724.22.05
S E M I C O N D U C TO R
40
ASIA
Harris Semiconductor PTE Ltd.
No. 1 Tannery Road
Cencon 1, #09-01
Singapore 1334
TEL: (65) 748-4200
FAX: (65) 748-0400