SAA7165 [NXP]

Digital Video Encoder EURO-DENC2; 数字视频编码器EURO- DENC2
SAA7165
型号: SAA7165
厂家: NXP    NXP
描述:

Digital Video Encoder EURO-DENC2
数字视频编码器EURO- DENC2

编码器
文件: 总45页 (文件大小:201K)
中文:  中文翻译
下载:  下载PDF数据表文档文件
INTEGRATED CIRCUITS  
DATA SHEET  
SAA7182A; SAA7183A  
Digital Video Encoder  
(EURO-DENC2)  
1996 Oct 02  
Preliminary specification  
Supersedes data of 1996 Sep 11  
File under Integrated Circuits, IC22  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
FEATURES  
Monolithic CMOS 3.3 V device with 5 V input stages  
Digital PAL/NTSC/SECAM encoder  
System pixel frequency 13.5 MHz  
Accepts MPEG decoded data on 8-bit wide input port.  
Input data format Cb, Y, Cr etc. “(CCIR 656)” or  
Y and Cb, Cr on 16 lines  
This applies to SAA7183A only. The device is protected  
by USA patent numbers 4631603, 4577216 and  
4819098 and other intellectual property rights. Use of  
the Macrovision anti-copy process in the device is  
licensed for non-commercial home use only.  
Reverse engineering or disassembly is prohibited.  
Please contact your nearest Philips Semiconductor  
sales office for more information  
Three DACs for CVBS, Y and C operating at 27 MHz  
with 10 bit resolution  
Three DACs for RGB operating at 27 MHz with 9 bit  
resolution, RGB sync on CVBS and Y  
Analog multiplexing between internal RGB and external  
RGB on-chip  
Controlled rise/fall times of output syncs and blanking  
Down-mode of DACs  
CVBS, Y, C and RGB output simultaneously  
Closed captioning and teletext encoding including  
sequencer and filter  
PQFP80 or PLCC84 package.  
Line 23 wide screen signalling encoding  
On-chip Cr, Y, Cb to RGB dematrix, including gain  
adjustment for Y and Cr, Cb, optionally to be by-passed  
for Cr, Y, Cb output on RGB DACs  
GENERAL DESCRIPTION  
The SAA7182A; SAA7183A encodes digital YUV video  
data to an NTSC, PAL, SECAM CVBS or S-Video signal  
and also RGB.  
Fast I2C-bus control port (400 kHz)  
Encoder can be master or slave  
Optionally, the YUV to RGB dematrix can be by-passed  
providing the digital-to-analog converted Cb, Y, Cr signals  
instead of RGB.  
Programmable horizontal and vertical input  
synchronization phase  
Programmable horizontal sync output phase  
Internal Colour Bar Generator (CBG)  
The circuit accepts CCIR compatible YUV data with  
720 active pixels per line in 4 : 2 : 2 multiplexed formats,  
for example MPEG decoded data. It includes a sync/clock  
generator and on-chip Digital-to-Analog Converters  
(DACs).  
Overlay with Look-Up Tables (LUTs) 8 × 3 bytes  
Macrovision Pay-per-View copy protection system as  
option, also used for RGB output.  
The circuit is compatible to the DIG-TV2 chip family.  
ORDERING INFORMATION  
TYPE  
PACKAGE  
NUMBER  
NAME  
DESCRIPTION  
plastic leaded chip carrier; 84 leads  
VERSION  
SAA7182AWP;  
SAA7183AWP  
PLCC84  
QFP80  
SOT189-2  
SOT318-2  
plastic quad flat package; 80 leads (lead length 1.95 mm);  
body 14 × 20 × 2.8 mm  
1996 Oct 02  
2
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
QUICK REFERENCE DATA  
SYMBOL  
PARAMETER  
3.3 V analog supply voltage  
MIN.  
3.1  
TYP.  
3.3  
MAX.  
3.5  
UNIT  
VDDA3  
VDDD3  
VDDD5  
IDDA  
V
3.3 V digital supply voltage  
5 V digital supply voltage  
analog supply current  
3.0  
4.75  
3.3  
5.0  
3.6  
5.25  
110  
80  
V
V
mA  
mA  
mA  
IDDD3  
IDDD5  
Vi  
3.3 V digital supply current  
5 V digital supply current  
input signal voltage levels  
10  
TTL compatible  
Vo(p-p)  
analog output signal voltages Y, C, CVBS and RGB without load  
(peak-to-peak value)  
1.4  
V
RL  
load resistance  
75  
300  
±2  
ILE  
LF integral linearity error  
LF differential linearity error  
operating ambient temperature  
LSB  
LSB  
°C  
DLE  
Tamb  
±1  
0
+70  
1996 Oct 02  
3
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
BLOCK DIAGRAM  
RTCI  
RCV1 TTXRQ XTALO LLC  
RCV2  
V
to V  
DDA9  
DDA4  
TESTB  
RESET SDA SCL SA  
CREF XTALI  
CDIR  
1
84 83  
4
75  
44 48  
37 50 35 36 20 47 45  
63, 64,  
68, 70,  
72, 74  
2
I C-bus  
8
control  
2
I C-BUS  
SECAM  
SYNC  
INTERFACE  
PROCESSOR  
CLOCK  
2
I C-bus  
control  
2
8
I C-bus  
8
clock  
and timing  
10 to 13  
16 to 19  
control  
DP0  
to  
DP7  
DbDr  
8
Y
Y
73  
71  
69  
25 to 28  
31 to 34  
8
CVBS  
Y
MP7  
to  
MP0  
D
OUTPUT  
INTERFACE  
DATA  
ENCODER  
CbCr  
C
MANAGER  
A
CHROMA  
OVL2  
to  
OVL0  
6 to 8  
3
V
9
SSA1  
2
2
52, 67, 76  
I C-bus  
I C-bus  
control  
KEY  
to  
8
8
internal  
control bus  
control  
2
V
SSA3  
I C-bus  
control  
53  
65  
62  
TESTC  
SELI  
2
21  
I C-bus  
8
TTX  
RI  
SAA7182A  
SAA7183A  
control  
3
61  
58  
55  
RED  
D
Y
RGB  
PROCESSOR  
GREEN  
BLUE  
CbCr  
A
3, 15, 24,  
30, 39, 42,  
51, 79, 81  
5, 14, 22,  
29, 38, 46,  
49, 80, 82  
54,  
57, 60  
2, 23, 40, 41,  
43, 66  
78  
77  
59 56  
BI  
MGD668  
GI  
n.c.  
V
SP AP  
V
V
V
DDA1  
DDD1  
SSD1  
to  
to  
to  
V
V
DDA3  
DDD9  
SSD9  
Fig.1 Block diagram; PLCC84.  
1996 Oct 02  
4
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
RTCI  
RCV1 TTXRQ XTALO LLC  
RCV2  
V
to V  
DDA9  
DDA4  
TESTB  
RESET SDA SCL SA  
CREF XTALI  
h
CDIR  
73 72 71 75  
63  
32 36  
27 38 25 26 11 35 33  
52, 53,  
56, 58,  
60, 62  
2
I C-bus  
8
control  
2
I C-BUS  
SECAM  
SYNC  
INTERFACE  
PROCESSOR  
CLOCK  
2
I C-bus  
control  
2
8
I C-bus  
8
clock  
and timing  
1 to 4  
control  
DP0  
to  
DP7  
DbDr  
7 to 10  
8
Y
Y
61  
59  
57  
15 to 18  
21 to 24  
8
CVBS  
Y
MP7  
to  
MP0  
D
OUTPUT  
INTERFACE  
DATA  
ENCODER  
CbCr  
C
MANAGER  
A
CHROMA  
77 to 79  
3
OVL2  
to  
OVL0  
V
SSA1  
80  
2
2
41, 55, 64  
I C-bus  
I C-bus  
control  
KEY  
to  
8
8
internal  
control bus  
control  
V
2
SSA3  
I C-bus  
control  
42  
54  
51  
TESTC  
SELI  
RI  
2
12  
I C-bus  
8
TTX  
SAA7182A  
SAA7183A  
control  
3
50  
47  
44  
RED  
D
Y
RGB  
PROCESSOR  
GREEN  
BLUE  
CbCr  
A
6, 14, 20,  
29, 31, 39,  
67, 69, 74  
5, 13, 19,  
28, 34, 37,  
68, 70, 76  
43,  
46, 49  
66  
65  
30, 40  
n.c.  
48 45  
BI  
MGD670  
GI  
V
SP AP  
V
V
V
DDA1  
DDD1  
SSD1  
to  
to  
to  
V
V
DDA3  
DDD9  
SSD9  
Fig.2 Block diagram; QFP80.  
1996 Oct 02  
5
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
PINNING  
PIN  
SYMBOL  
DESCRIPTION  
PLCC84 QFP80  
RESET  
1
73  
Reset input, active LOW. After reset is applied, all digital I/Os are in input mode.  
The I2C-bus receiver waits for the START condition.  
n.c.  
2
3
4
6
not connected  
digital ground 1  
VSSD1  
SA  
75  
The I2C-bus slave address select input pin. LOW: slave address = 88H,  
HIGH = 8CH.  
VDDD1  
OVL2  
OVL1  
OVL0  
KEY  
5
13  
77  
78  
79  
80  
1
digital supply voltage 1 (3.3 V)  
6
7
3-bit overlay data input. This is the index for the internal look-up table.  
Key input for OVL. When HIGH it selects OVL input.  
8
9
DP0  
10  
11  
12  
13  
14  
15  
16  
17  
18  
19  
20  
21  
22  
23  
24  
25  
26  
27  
28  
29  
30  
31  
32  
33  
34  
35  
36  
DP1  
2
Lower 4 bits of the data port. Input for multiplexed Cb, Cr data if 16 line input mode  
is used.  
DP2  
3
DP3  
4
VDDD2  
VSSD2  
DP4  
5
digital supply voltage 2 (5 V)  
digital ground 2  
14  
7
DP5  
8
Upper 4 bits of the data port. Input for multiplexed Cb, Cr data if 16 line input mode  
is used.  
DP6  
9
DP7  
10  
11  
12  
28  
TTXRQ  
TTX  
Teletext request output, indicating when bit stream is valid.  
Teletext bit stream input.  
VDDD3  
n.c.  
digital supply voltage 3 (3.3 V)  
not connected  
VSSD3  
MP7  
20  
15  
16  
17  
18  
19  
29  
21  
22  
23  
24  
25  
26  
digital ground 1  
MP6  
Upper 4 bits of MPEG port. It is an input for “CCIR 656” style multiplexed Cb, Y, Cr  
data, or for Y data only, if 16 line input mode is used.  
MP5  
MP4  
VDDD4  
VSSD4  
MP3  
digital supply voltage 4 (5 V)  
digital ground 4  
MP2  
Lower 4 bits of MPEG port. It is an input for “CCIR 656” style multiplexed Cb, Y, Cr  
data, or for Y data only, if 16 line input mode is used.  
MP1  
MP0  
RCV1  
RCV2  
Raster Control 1 for video port. This pin receives/provides a VS/FS/FSEQ signal.  
Raster Control 2 for video port. This pin provides an HS pulse of programmable  
length or receives an HS pulse.  
1996 Oct 02  
6
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
PIN  
SYMBOL  
DESCRIPTION  
PLCC84 QFP80  
RTCI  
37  
27  
Real Time Control input. If the LLC clock is provided by an SAA7111 or SAA7151B,  
RTCI should be connected to the RTCO pin of the respective decoder to improve  
the signal quality.  
VDDD5  
VSSD5  
n.c.  
38  
39  
40  
41  
42  
43  
44  
68  
39  
40  
digital supply voltage 5 (3.3 V)  
digital ground 5  
not connected  
n.c.  
not connected  
VSSD6  
n.c.  
31  
30  
32  
digital ground 6 for oscillator  
not connected  
XTALI  
Crystal oscillator input (from crystal). If the oscillator is not used, this pin should be  
connected to ground.  
XTALO  
VDDD6  
CREF  
LLC  
45  
46  
47  
48  
33  
34  
35  
36  
Crystal oscillator output (to crystal).  
digital supply voltage 6 for oscillator (3.3 V)  
Clock Reference signal. This is the clock qualifier for DIG-TV2 compatible signals.  
Line-Locked Clock. This is the 27 MHz master clock for the encoder. The I/O  
direction is set by the CDIR pin.  
VDDD7  
CDIR  
49  
50  
37  
38  
digital supply voltage 7 (5 V)  
Clock direction. If CDIR input is HIGH, the circuit receives a clock and optional  
CREF signal, otherwise if CDIR is LOW, CREF and LLC are generated by the  
internal crystal oscillator.  
VSSD7  
VSSA1  
TESTC  
VDDA1  
BLUE  
BI  
51  
52  
53  
54  
55  
56  
57  
58  
59  
60  
61  
62  
63  
64  
65  
67  
41  
42  
43  
44  
45  
46  
47  
48  
49  
50  
51  
52  
53  
54  
digital ground 7  
Analog ground 1 for the DACs.  
Analog test pin. Leave open-circuit for normal operation.  
Analog supply voltage 1 for the RGB DACs (3.3 V).  
Analog output of the BLUE component.  
Analog input that can be switched to BLUE when SELI = HIGH.  
Analog supply voltage 2 for RGB DACs (3.3 V).  
Analog output of GREEN component.  
VDDA2  
GREEN  
GI  
Analog input that can be switched to GREEN when SELI = HIGH.  
Analog supply voltage 3 for RGB DACs (3.3 V).  
Analog output of RED component.  
VDDA3  
RED  
RI  
Analog input that can be switched to RED when SELI = HIGH.  
Analog supply voltage 4 for DACs (3.3 V).  
Analog supply voltage 5 for DACs (3.3 V).  
VDDA4  
VDDA5  
SELI  
Select analog input. Digital-to-analog converted RGB output when SELI = LOW;  
RI, GI and BI output when SELI = HIGH.  
n.c.  
66  
67  
68  
69  
70  
not connected  
VSSA2  
VDDA6  
CHROMA  
VDDA7  
55  
56  
57  
58  
Analog ground 2 for the DACs.  
Analog supply voltage 6 for DACs (3.3 V).  
Analog output of the chrominance signal.  
Analog supply voltage 7 for the Y/C/CVBS DACs (3.3 V).  
1996 Oct 02  
7
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
PIN  
SYMBOL  
DESCRIPTION  
PLCC84 QFP80  
Y
71  
72  
73  
74  
75  
76  
77  
78  
79  
80  
81  
82  
83  
84  
59  
60  
61  
62  
63  
64  
65  
66  
69  
76  
74  
70  
71  
72  
Analog output of VBS signal.  
VDDA8  
CVBS  
VDDA9  
TESTB  
VSSA3  
AP  
Analog supply voltage 8 for the Y/C/CVBS DACs.  
Analog output of the CVBS signal.  
Analog supply voltage 9 for the Y/C/CVBS DACs.  
Analog test pin. Leave open-circuit for normal operation.  
Analog ground 3 for the DACs.  
Test pin. Connected to digital ground for normal operation.  
Test pin. Connected to digital ground for normal operation.  
digital ground 8  
SP  
VSSD8  
VDDD8  
VSSD9  
VDDD9  
SCL  
digital supply voltage 8 (3.3 V)  
digital ground 9  
digital supply voltage 9 (5 V)  
I2C-bus serial clock input.  
I2C-bus serial data input/output.  
SDA  
1996 Oct 02  
8
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
DP2 12  
DP3 13  
74  
73  
72  
71  
70  
69  
68  
67  
66  
65  
64  
63  
62  
61  
60  
59  
58  
57  
56  
55  
54  
V
DDA9  
CVBS  
V
14  
15  
V
DDD2  
DDA8  
V
Y
SSD2  
DP4 16  
DP5 17  
DP6 18  
DP7 19  
V
DDA7  
CHROMA  
V
DDA6  
V
SSA2  
TTXRQ 20  
TTX 21  
n.c.  
SELI  
SAA7182A  
SAA7183A  
V
22  
V
DDD3  
DDA5  
n.c. 23  
24  
V
DDA4  
V
RI  
SSD3  
MP7 25  
MP6 26  
MP5 27  
MP4 28  
29  
RED  
V
DDA3  
GI  
GREEN  
V
V
DDD4  
DDA2  
V
30  
BI  
SSD4  
MP3 31  
BLUE  
MP2  
32  
V
DDA1  
MGD669  
Fig.3 Pin configuration; PLCC84.  
9
1996 Oct 02  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
DP0  
DP1  
DP2  
DP3  
1
2
3
4
5
6
7
8
9
64 V  
SSA3  
63 TESTB  
62  
V
DDA9  
61 CVBS  
V
60  
59  
58  
V
Y
V
DDD2  
DDA8  
DDA7  
V
SSD1  
DP4  
DP5  
DP6  
57 CHROMA  
56  
55  
V
V
DDA6  
SSA2  
DP7 10  
TTXRQ 11  
TTX 12  
54 SELI  
53  
52  
V
V
DDA5  
DDA4  
SAA7182A  
SAA7183A  
V
13  
14  
DDD1  
V
51 RI  
SSD2  
MP7 15  
MP6 16  
MP5 17  
MP4 18  
19  
50 RED  
49  
V
DDA3  
48 GI  
47 GREEN  
V
46 V  
DDA2  
DDD4  
V
20  
45 BI  
SSD3  
MP3 21  
MP2 22  
MP1 23  
MP0 24  
44 BLUE  
43  
V
DDA1  
42 TESTC  
41  
V
SSA1  
MGD671  
Fig.4 Pin configuration; QFP80.  
10  
1996 Oct 02  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
actual subcarrier, PAL-ID, and if connected to SAA7111,  
definite subcarrier phase can be inserted.  
FUNCTIONAL DESCRIPTION  
The digital video encoder (EURO-DENC2) encodes digital  
luminance and colour difference signals into analog CVBS  
and simultaneously S-Video signals. NTSC-M, PAL B/G,  
SECAM standards and sub-standards are supported.  
The EURO-DENC2 synthesizes all necessary internal  
signals, colour subcarrier frequency, and synchronization  
signals, from that clock.  
European teletext encoding is supported if an appropriate  
teletext bitstream is applied to the TTX pin.  
Both interlaced and non-interlaced operation is possible  
for all standards.  
Wide screen signalling data can be loaded via the I2C-bus,  
and is inserted into line 23 for standards using 50 Hz field  
rate.  
In addition, the de-matrixed Y, Cb, and Cr input is  
available on three separate analog outputs as RED,  
GREEN and BLUE. Under software control the dematrix  
can be by-passed to output digital-to-analog converted Cr,  
Y, and Cb signals on RGB outputs. Separate digital gain  
adjustment for luminance and colour difference signals is  
available.  
The IC also contains Closed Caption and Extended Data  
Services Encoding (Line 21), and supports anti-taping  
signal generation in accordance with Macrovision; it also  
supports overlay via KEY and three control bits by a  
24 × 8 LUT.  
Analog on-chip multiplexing between internal  
digital-to-analog converted RGB and external RI, GI and  
BI signals is also supported.  
A number of possibilities are provided for setting different  
video parameters such as:  
The basic encoder function consists of subcarrier  
generation, colour modulation and insertion of  
synchronization signals. Luminance and chrominance  
signals are filtered in accordance with the standard  
requirements of “RS-170-A” and “CCIR 624”.  
Black and blanking level control  
Colour subcarrier frequency  
Variable burst amplitude etc.  
During reset (RESET = LOW) and after reset is released,  
all digital I/O stages are set to input mode. A reset forces  
the I2C-bus interface to abort any running bus transfer and  
sets register 3A to 03H, register 61 to 06H and  
registers 6BH and 6EH to 00H. All other control registers  
are not influenced by a reset.  
For ease of analog post filtering the signals are twice  
oversampled with respect to the pixel clock before  
digital-to-analog conversion.  
For total filter transfer characteristics see  
Figs 5, 6, 7, 8, 9 and 10. The DACs for Y, C, and CVBS  
are realized with full 10-bit resolution, DACs for RGB are  
with 9-bit resolution.  
Data manager  
In the data manager, real time arbitration on the data  
stream to be encoded is performed.  
The MPEG port (MP) accept 8 line multiplexed Cb, Y, Cr  
data.  
Depending on the polarity of pin KEY, the MP input  
(or MP/DP input) or OVL input are selected to be encoded  
to CVBS and Y/C signals, and output as RGB.  
The 8-bit multiplexed Cb-Y-Cr formats are “CCIR 656”  
(D1 format) compatible, but the SAV and EAV codes can  
be decoded optionally, when the device is to operate in  
slave mode.  
KEY controls OVL entries of a programmable LUT for  
encoded signals and for RGB output. The common KEY  
switching signal can be disabled by software for the  
signals to be encoded (Y, C and CVBS), such that OVL will  
appear on RGB outputs, but not on Y, C and CVBS.  
Alternatively, 8-bits Y on MP port and 8-bit multiplexed Cb,  
Cr on DP port can be chosen as input.  
A crystal-stable master clock (LLC) of 27 MHz, which is  
twice the CCIR line-locked pixel clock of 13.5 MHz, needs  
to be supplied externally. Optionally, a crystal oscillator  
input/output pair of pins and an on-chip clock driver is  
provided.  
OVL input under control of KEY can be also used to insert  
decoded teletext information or other on-screen data.  
Optionally, the OVL colour LUTs located in this block, can  
be read out in a pre-defined sequence (8 steps per active  
video line), achieving, for example, a colour bar test  
pattern generator without need for an external data  
source. The colour bar function is only under software  
control.  
It is also possible to connect a Philips Digital Video  
Decoder (SAA7111 or SAA7151B) in conjunction with a  
CREF clock qualifier to EURO-DENC2. Via the RTCI pin,  
connected to RTCO of a decoder, information concerning  
1996 Oct 02  
11  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
Encoder  
CLOSED CAPTION ENCODER  
Using this circuit, data in accordance with the specification  
of Closed Caption or Extended Data Service, delivered by  
the control interface, can be encoded (Line 21).  
Two dedicated pairs of bytes (two bytes per field), each  
pair preceded by run-in clocks and framing code, are  
possible.  
VIDEO PATH  
The encoder generates out of Y, U and V baseband  
signals luminance and colour subcarrier output signals,  
suitable for use as CVBS or separate Y and C signals.  
Luminance is modified in gain and in offset (latter  
programmable in a certain range to enable different black  
level set-ups). After having been inserted a fixed  
synchronization level, in accordance with standard  
composite synchronization schemes, and blanking level,  
programmable also in a certain range to allow for  
manipulations with Macrovision anti-taping, additional  
insertion of AGC super-white pulses, programmable in  
height, is supported.  
The actual line number where data is to be encoded in, can  
be modified in a certain range.  
Data clock frequency is in accordance with definition for  
NTSC-M standard 32 times horizontal line frequency.  
Data LOW at the output of the DACs corresponds to 0 IRE,  
data HIGH at the output of the DACs corresponds to  
approximately 50 IRE.  
In order to enable easy analog post filtering, luminance is  
interpolated from 13.5 MHz data rate to 27 MHz data rate,  
providing luminance in 10-bit resolution. This filter is also  
used to define smoothed transients for synchronization  
pulses and blanking period. For transfer characteristic of  
the luminance interpolation filter see Figs 7 and 8.  
It is also possible to encode Closed Caption Data for 50 Hz  
field frequencies at 32 times horizontal line frequency.  
ANTI-TAPING (SAA7183A ONLY)  
For more information contact your nearest Philips  
Semiconductors sales office.  
Chrominance is modified in gain (programmable  
separately for U and V), standard dependent burst is  
inserted, before baseband colour signals are interpolated  
from 6.75 MHz data rate to 27 MHz data rate. One of the  
interpolation stages can be bypassed, thus providing a  
higher colour bandwidth, which can be made use of for  
Y and C output. For transfer characteristics of the  
chrominance interpolation filter see Figs 5 and 6.  
RGB processor  
This block contains a dematrix in order to produce RED,  
GREEN and BLUE signals to be fed to a SCART plug.  
Before Y, Cb and Cr signals are de-matrixed, individual  
gain adjustment for Y and colour difference signals and  
2 times oversampling for luminance and 4 times  
oversampling for colour difference signals is performed.  
For transfer curves of luminance and colour difference  
components of RGB see Figs 9 and 10.  
The amplitude of inserted burst is programmable in a  
certain range, suitable for standard signals and for special  
effects. Behind the succeeding quadrature modulator,  
colour in 10-bit resolution is provided on subcarrier.  
SECAM processor  
The numeric ratio between Y and C outputs is in  
accordance with set standards.  
SECAM specific pre-processing is achieved in this block  
by a pre-emphasis of colour difference signals (for gain  
and phase see Figs 11 and 12).  
TELETEXT INSERTION AND ENCODING  
Pin TTX receives a teletext bitstream sampled at the LLC  
clock, each teletext bit is carried by four or three LLC  
samples.  
A baseband frequency modulator with a reference  
frequency shifted from 4.286 MHz to DC carries out  
SECAM modulation in accordance with appropriate  
standard or optionally wide clipping limits.  
Phase variant interpolation is achieved on this bitstream in  
the internal teletext encoder, providing sufficient small  
phase jitter on the output text lines.  
After the HF pre-emphasis, also applied on a DC reference  
carrier (anti-Cloche filter; see Figs 13 and 14), line-by-line  
sequential carriers with black reference of 4.25 MHz (Db)  
and 4.40625 MHz (Dr) are generated using specified  
values for FSC programming bytes.  
TTXRQ provides a fully programmable request signal to  
the teletext source, indicating the insertion period of  
bitstream at lines selectable independently for both fields.  
The internal insertion window for text is set to 360 teletext  
bits including clock run-in bits. For protocol and timing  
see Fig.19.  
Alternating phase reset in accordance with SECAM  
standard is carried out automatically. During vertical  
blanking the so-called bottle pulses are not provided.  
1996 Oct 02  
12  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
A Vertical Sync signal (VS) with 3 or 2.5 lines duration,  
Output interface/DACs  
or;  
In the output interface encoded both Y and C signals are  
converted from digital-to-analog in 10-bit resolution.  
Y and C signals are also combined to a 10-bit CVBS  
signal.  
An ODD/EVEN signal which is LOW in odd fields, or;  
A field sequence signal (FSEQ) which is HIGH in the first  
of 4, 8, 12 fields respectively.  
The CVBS output occurs with the same processing delay  
as the Y and C outputs. Absolute amplitudes at the input  
On the RCV2 port, the IC can provide a horizontal pulse  
with programmable start and stop phase; this pulse can be  
inhibited in the vertical blanking period to build up, for  
example, a composite blanking signal.  
of the DAC for CVBS is reduced by 15  
16 with respect to  
Y and C DACs to make maximum use of conversion  
ranges.  
The polarity of both RCV1 and RCV2 is selectable by  
software control.  
RED, GREEN and BLUE signals are also converted from  
digital-to-analog, each providing a 9-bit resolution. It is  
also possible to feed through three external analog RGB  
signals at pins RI, BI and GI when pin SELI = HIGH  
The length of a field and the start and end of its active part  
can be programmed. The active part of a field always  
starts at the beginning of a line.  
Outputs of the DACs can be set together in two groups via  
software control to minimum output voltage for either  
purpose.  
I2C-bus interface  
The I2C-bus interface is a standard slave transceiver,  
supporting 7-bit slave addresses and 400 kbits/s  
guaranteed transfer rate. It uses 8-bit subaddressing with  
an auto-increment function. All registers are write only,  
except one readable status byte.  
Synchronization  
Synchronization of the EURO-DENC2 is able to operate in  
two modes; slave mode and master mode.  
Two I2C-bus slave addresses are selected:  
88H: LOW at pin SA  
In the slave mode, the circuit accepts synchronization  
pulses at the bidirectional RCV1 port. The timing and  
trigger behaviour related to RCV1 can be influenced by  
programming the polarity and on-chip delay of RCV1.  
Active slope of RCV1 defines the vertical phase and  
optionally the odd/even and colour frame phase to be  
initialized, it can be also used to set the horizontal phase.  
8CH: HIGH at pin SA.  
Input levels and formats  
EURO-DENC2 expects digital Y, Cb, Cr data with levels  
(digital codes) in accordance with “CCIR 601”.  
If the horizontal phase is not to be influenced by RCV1, a  
horizontal pulse needs to be supplied at the RCV2 pin.  
Timing and trigger behaviour can also be influenced for  
RCV2.  
For C and CVBS outputs, deviating amplitudes of the  
colour difference signals can be compensated by  
independent gain control setting, while gain for luminance  
is set to predefined values, distinguishable for 7.5 IRE  
set-up or without set-up.  
If there are missing pulses at RCV1 and/or RCV2, the time  
base of EURO-DENC2 runs free, thus an arbitrary number  
of synchronization slopes may miss, but no additional  
pulses (with the incorrect phase) must occur.  
For RGB outputs variable amplification of the Y, Cb and Cr  
components is provided, enabling adjustment of contrast  
and colour saturation in certain range.  
If the vertical and horizontal phase is derived from RCV1,  
RCV2 can be used for horizontal or composite blanking  
input or output.  
Reference levels are measured with a colour bar,  
100% white, 100% amplitude and 100% saturation.  
Alternatively, the device can be triggered by auxiliary  
codes in a CCIR 656 data stream at the MP port  
In the master mode, the time base of the circuit  
continuously runs free. On the RCV1 port, the IC can  
output:  
1996 Oct 02  
13  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
Table 1 “CCIR 601” signal component levels  
SIGNALS(1)  
COLOUR  
Y
Cb  
Cr  
R(2)  
G(2)  
B(2)  
White  
Yellow  
Cyan  
235  
210  
170  
145  
106  
81  
128  
16  
128  
146  
16  
235  
235  
16  
235  
235  
235  
235  
16  
235  
16  
166  
54  
235  
16  
Green  
Magenta  
Red  
34  
16  
202  
90  
222  
240  
110  
128  
235  
235  
16  
235  
16  
16  
Blue  
41  
240  
128  
16  
235  
16  
Black  
16  
16  
16  
Notes  
1. Transformation:  
a) R = Y + 1.3707 × (Cr 128)  
b) G = Y 0.3365 × (Cb 128) 0.6982 × (Cr 128)  
c) B = Y + 1.7324 × (Cb 128).  
2. Representation of R, G and B (or Cr, Y and Cb) at the output is 9 bits at 27 MHz.  
Table 2 8-bit multiplexed format (similar to “CCIR 601”)  
BITS  
TIME  
0
1
2
2
4
5
6
7
Sample  
Cb0  
Y0  
Cr0  
Y1  
Cb2  
Y2  
Cr2  
Y3  
Luminance pixel number  
Colour pixel number  
0
1
2
3
0
2
Table 3 16-bit multiplexed format (DTV2 format)  
BITS  
TIME  
0
1
2
3
4
5
6
7
Sample Y line  
Y0  
Cb0  
0
Y1  
Cr0  
1
Y2  
Cb2  
2
Y3  
Cr2  
3
Sample UV line  
Luminance pixel number  
Colour pixel number  
0
2
1996 Oct 02  
14  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
1996 Oct 02  
15  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
1996 Oct 02  
16  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
I2C-bus format  
Table 5 I2C-bus address; see Table 6  
S
SLAVE ADDRESS ACK SUBADDRESS ACK DATA 0 ACK  
--------  
DATA n ACK  
P
Table 6 Explanation of Table 5  
PART  
DESCRIPTION  
S
START condition  
Slave address  
1 0 0 0 1 0 0 X or 1 0 0 0 1 1 0 X (note 1)  
acknowledge, generated by the slave  
subaddress byte  
ACK  
Subaddress (note 2)  
DATA  
--------  
P
data byte  
continued data bytes and ACKs  
STOP condition  
Notes  
1. X is the read/write control bit; X = logic 0 is order to write; X = logic 1 is order to read, no subaddressing with read.  
2. If more than 1 byte DATA is transmitted, then auto-increment of the subaddress is performed.  
Slave Receiver  
Table 7 Subaddress 26 and 27  
DATA BYTE  
LOGIC LEVEL  
DESCRIPTION  
WSS0 to WSS13  
Wide Screen Signalling bits  
3 to 0 = aspect ratio  
7 to 4 = enhanced services  
10 to 8 = subtitles  
13 to 11 = reserved  
WSSON  
0
1
wide screen signalling output is disabled  
wide screen signalling output is enabled  
Table 8 Subaddress 38 and 39  
DATA BYTE  
DESCRIPTION  
GY0 to GY4  
Gain luminance of RGB (Cr, Y and Cb) output, ranging from (1 1632) to (1 + 1532). Suggested  
nominal value = 6 (11010b), depending on external application.  
GCD0 to GCD4 Gain Colour Difference of RGB (Cr, Y and Cb) output, ranging from (1 - 1632) to (1 + 1532).  
Suggested nominal value = 6 (11010b), depending on external application.  
1996 Oct 02  
17  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
Table 9 Subaddress 3A  
DATA  
BYTE  
LOGIC  
LEVEL  
DESCRIPTION  
UV2C  
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
Cb, Cr data are two’s complement.  
Cb, Cr data are straight binary. Default after reset.  
Y data is two’s complement.  
Y2C  
Y data is straight binary. Default after reset.  
FMT16  
DEMOFF  
SYMP  
Selects Cb, Y, Cr and Y on 8 lines on MP port (“CCIR 656” compatible). Default after reset.  
Selects Cb and Cr on DP port and Y on MP port.  
Y, Cb and Cr for RGB dematrix is active. Default after reset.  
Y, Cb and Cr for RGB dematrix is bypassed.  
Horizontal and vertical trigger is taken from RCV2 and RCV1 respectively. Default after reset.  
Horizontal and vertical trigger is decoded out of “CCIR 656” compatible data at MP port.  
Normal polarity of CREF for DIG-TV2 compatible input signals.  
Inverted polarity of CREF for DIG-TV2 compatible input signals.  
OVL keying enabled for Y, C and CVBS outputs. Default after reset.  
OVL keying disabled for Y, C and CVBS outputs.  
PCREF  
DISKEY  
CBENB  
Data from input ports is encoded. Default after reset.  
Colour bar with programmable colours (entries of OVL_LUTs) is encoded. The LUTs are read in  
upward order from index 0 to index 7.  
1996 Oct 02  
18  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
Table 10 Subaddress 42 to 59  
DATA BYTE(1)  
COLOUR  
INDEX(2)  
OVLY  
OVLU  
OVLV  
White  
107 (6BH)  
107 (6BH)  
82 (52H)  
0 (00H)  
0 (00H)  
0 (00H)  
0 (00H)  
0
1
2
3
4
5
6
7
Yellow  
Cyan  
144 (90H)  
172 (ACH)  
38 (26H)  
29 (1DH)  
182 (B6H)  
200 (C8H)  
74 (4AH)  
56 (38H)  
218 (DAH)  
227 (E3H)  
112 (70H)  
84 (54H)  
0 (00H)  
18 (12H)  
14 (0EH)  
144 (90H)  
172 (ACH)  
162 (A2H)  
185 (B9H)  
94 (5EH)  
71 (47H)  
112 (70H)  
84 (54H)  
238 (EEH)  
242 (F2H)  
0 (00H)  
34 (22H)  
42 (2AH)  
03 (03H)  
Green  
Magenta  
Red  
17 (11H)  
240 (F0H)  
234 (EAH)  
212 (D4H)  
209 (D1H)  
193 (C1H)  
169 (A9H)  
163 (A3H)  
144 (90H)  
144 (90H)  
Blue  
Black  
0 (00H)  
0 (00H)  
Notes  
1. Contents of OVL look-up tables. All 8 entries are 8-bits. Data representation is in accordance with “CCIR 601”  
(Y, Cb and Cr), but two’s complement, e.g. for a 100  
100 (upper number) or 100  
75 (lower number) colour bar.  
2. For normal colour bar with CBENB = logic 1.  
Table 11 Subaddress 5A  
DATA BYTE(1)  
VALUE  
RESULT  
CHPS  
tbf  
tbf  
tbf  
tbf  
PAL-B/G and data from input ports  
PAL-B/G and data from look-up table  
NTSC-M and data from input ports  
NTSC-M and data from look-up table  
Note  
1. Phase of encoded colour subcarrier (including burst) relative to horizontal sync. Can be adjusted in steps of 360/256  
degrees.  
1996 Oct 02  
19  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
Table 12 Subaddress 5B and 5D  
DATA BYTE  
DESCRIPTION  
CONDITIONS  
REMARKS  
GAINU  
variable gain for Cb signal; white-to-black = 92.5 IRE(1)  
input representation  
accordance with  
GAINU = 118 (76H)  
“CCIR 601”  
GAINU = 0  
output subcarrier of U contribution = 0  
output subcarrier of U contribution = nominal  
white-to-black = 100 IRE(2)  
GAINU = 0  
output subcarrier of U contribution = 0  
GAINU = 125 (7DH)  
output subcarrier of U contribution = nominal  
nominal GAINU for  
SECAM encoding  
value = 106 (6AH)  
Notes  
1. GAINU = 2.17 × nominal to +2.16 × nominal.  
2. GAINU = 2.05 × nominal to +2.04 × nominal.  
Table 13 Subaddress 5C and 5E  
DATA BYTE  
DESCRIPTION  
CONDITIONS  
REMARKS  
GAINV  
variable gain for Cr signal; white-to-black = 92.5 IRE(1)  
input representation  
accordance with  
GAINV = 165 (A5H)  
“CCIR 601”  
GAINV = 0  
output subcarrier of V contribution = 0  
output subcarrier of V contribution = nominal  
white-to-black = 100 IRE(2)  
GAINV = 0  
output subcarrier of V contribution = 0  
GAINV = 175 (AFH)  
output subcarrier of V contribution = nominal  
nominal GAINV for  
SECAM encoding  
value = 129 (17FH)  
Notes  
1. GAINV = 1.55 × nominal to +1.55 × nominal.  
2. GAINV = 1.46 × nominal to +1.46 × nominal.  
Table 14 Subaddress 5D  
DATA BYTE  
DESCRIPTION  
CONDITIONS  
REMARKS  
BLCKL  
variable black level; input white-to-sync = 140 IRE(1)  
representationaccordance  
with “CCIR 601”  
BLCKL = 0  
output black level = 24 IRE  
output black level = 49 IRE  
BLCKL = 63 (3FH)  
white-to-sync = 143 IRE(2)  
BLCKL = 0  
output black level = 24 IRE  
output black level = 50 IRE  
BLCKL = 63 (3FH)  
Notes  
1. Output black level/IRE = BLCKL × 25/63 + 24; recommended value: BLCKL = 60 (3CH) normal.  
2. Output black level/IRE = BLCKL × 26/63 + 24; recommended value: BLCKL = 45 (2DH) normal.  
1996 Oct 02  
20  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
Table 15 Subaddress 5E  
DATA BYTE  
DESCRIPTION  
CONDITIONS  
REMARKS  
BLNNL  
variable blanking level  
white-to-sync = 140 IRE(1)  
BLNNL = 0  
output blanking level = 17 IRE  
BLNNL = 63 (3FH)  
white-to-sync = 143 IRE(2)  
BLNNL = 0  
output blanking level = 42 IRE  
output blanking level = 17 IRE  
BLNNL = 63 (3FH)  
logic 0  
output blanking level = 43 IRE  
DECTYP  
RTCI  
real time control input from SAA7151B  
real time control input from SAA7111  
logic 1  
Notes  
1. Output black level/IRE = BLNNL × 25/63 + 17; recommended value: BLNNL = 58 (3AH) normal.  
2. Output black level/IRE = BLNNL × 26/63 + 17; recommended value: BLNNL = 63 (3FH) normal.  
Table 16 Subaddress 5F  
DATA BYTE  
BLNVB  
DESCRIPTION  
variable blanking level during vertical blanking interval is typically identical to value of BLNNL  
select cross colour reduction filter in luminance; see Table 17  
CCRS  
Table 17 Logic levels and function of CCRS  
CCRS1  
CCRS0  
FUNCTION  
0
0
1
1
0
1
0
1
no cross colour reduction; for overall transfer characteristic of luminance see Fig.7  
cross colour reduction #1 active; for overall transfer characteristic see Fig.7  
cross colour reduction #2 active; for overall transfer characteristic see Fig.7  
cross colour reduction #3 active; for overall transfer characteristic see Fig.7  
1996 Oct 02  
21  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
Table 18 Subaddress 61:  
DATA BYTE LOGIC LEVEL  
DESCRIPTION  
FISE  
0
1
0
1
0
864 total pixel clocks per line; default after reset  
858 total pixel clocks per line  
PAL  
NTSC encoding (non-alternating V component)  
PAL encoding (alternating V component); default after reset  
SCBW  
enlarged bandwidth for chrominance encoding (for overall transfer characteristic of  
chrominance in baseband representation see Figs 5 and 6); wide clipping for  
SECAM  
1
standard bandwidth for chrominance encoding (for overall transfer characteristic of  
chrominance in baseband representation see Figs 5 and 6); default after reset  
SECAM  
YGS  
0
1
0
1
0
1
0
1
0
1
no SECAM encoding; default after reset  
SECAM encoding activated  
luminance gain for white black 100 IRE; default after reset  
luminance gain for white black 92.5 IRE including 7.5 IRE set-up of black  
PAL switch phase is nominal; default after reset  
INPI  
PAL switch phase is inverted compared to nominal  
DACs for CVBS, Y and C in normal operational mode; default after reset  
DACs for CVBS, Y and C forced to lowest output voltage  
DACs for R, G and B in normal operational mode; default after reset  
DACs for R, G and B forced to lowest output voltage  
DOWNA  
DOWNB  
1996 Oct 02  
22  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
Table 19 Subaddress 62A  
DATA BYTE LOGIC LEVEL  
DESCRIPTION  
RTCE  
0
1
no real time control of generated subcarrier frequency  
real time control of generated subcarrier frequency through SAA7151B or SAA7111  
(timing see Fig.18)  
Table 20 Subaddress 62B  
DATA BYTE  
DESCRIPTION  
CONDITIONS  
REMARKS  
BSTA  
amplitude of colour burst; white-to-black = 92.5 IRE;  
input representation in  
accordance with  
“CCIR 601”  
burst = 40 IRE; NTSC encoding  
BSTA = 0 to 1.25 × nominal(1)  
white-to-black = 92.5 IRE;  
burst = 40 IRE; PAL encoding  
BSTA = 0 to 1.76 × nominal(2)  
white-to-black = 100 IRE;  
burst = 43 IRE; NTSC encoding  
BSTA = 0 to 1.20 × nominal(3)  
white-to-black = 100 IRE;  
burst = 43 IRE; PAL encoding  
BSTA = 0 to 1.67 × nominal(4)  
fixed burst amplitude with SECAM encoding  
Notes  
1. Recommended value: BSTA = 102 (66H).  
2. Recommended value: BSTA = 72 (48H).  
3. Recommended value: BSTA = 106 (6AH).  
4. Recommended value: BSTA = 75 (4BH).  
Table 21 Subaddress 63 to 66 (four bytes to program subcarrier frequency)  
DATA BYTE  
DESCRIPTION  
CONDITIONS  
REMARKS  
FSC0 to FSC3 ffsc = subcarrier frequency  
(in multiples of line  
frequency);  
FSC3 = most significant byte  
f
FSC = round fsc × 2 32  
-------  
fllc  
FSC0 = least significant byte  
fllc = clock frequency (in  
multiples of line  
see note 1  
frequency)  
Note  
1. Examples:  
a) NTSC-M: ffsc = 227.5, fllc = 1716 FSC = 569408543 (21F07C1FH).  
b) PAL-B/G: ffsc = 283.7516, fllc = 1728 FSC = 705268427 (2A098ACBH).  
c) SECAM: ffsc = 274.304, fllc = 1728 FSC = 681786290 (28A33BB2H).  
1996 Oct 02  
23  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
Table 22 Subaddress 67 to 6A  
DATA BYTE(1)  
DESCRIPTION  
L21O0  
L21O1  
L21E0  
L21E1  
first byte of captioning data, odd field  
second byte of captioning data, odd field  
first byte of extended data, even field  
second byte of extended data, even field  
Note  
1. LSBs of the respective bytes are encoded immediately after run-in and framing code, the MSBs of the respective  
bytes have to carry the parity bit, in accordance with the definition of Line 21 encoding format.  
Table 23 Subaddress 6B  
DATA BYTE LOGIC LEVEL  
DESCRIPTION  
PRCV2  
0
polarity of RCV2 as output is active HIGH, rising edge is taken when input,  
respectively; default after reset  
1
polarity of RCV2 as output is active LOW, falling edge is taken when input,  
respectively  
ORCV2  
CBLF  
0
1
0
pin RCV2 is switched to input; default after reset  
pin RCV2 is switched to output  
if ORCV2 = HIGH, pin RCV2 provides an HREF signal (Horizontal Reference pulse  
that is defined by RCV2S and RCV2E, also during vertical blanking Interval); default  
after reset  
if ORCV2 = LOW and bit SYMP = LOW, signal input to RCV2 is used for horizontal  
synchronization only (if TRCV2 = 1); default after reset  
1
if ORCV2 = HIGH, pin RCV2 provides a ‘Composite-Blanking-Not’ signal, this is a  
reference pulse that is defined by RCV2S and RCV2E, excluding Vertical Blanking  
Interval, which is defined by FAL and LAL  
if ORCV2 = LOW and bit SYMP = LOW, signal input to RCV2 is used for horizontal  
synchronization (if TRCV2 = 1) and as an internal blanking signal  
PRCV1  
0
polarity of RCV1 as output is active HIGH, rising edge is taken when input; default  
after reset  
1
0
1
0
polarity of RCV1 as output is active LOW, falling edge is taken when input  
pin RCV1 is switched to input; default after reset  
pin RCV1 is switched to output  
ORCV1  
TRCV2  
horizontal synchronization is taken from RCV1 port (at bit SYMP = LOW) or from  
decoded frame sync of CCIR 656 input (at bit SYMP = HIGH); default after reset  
1
horizontal synchronization is taken from RCV2 port (at bit SYMP = LOW)  
defines signal type on pin RCV1; see Table 24  
SRCV1  
1996 Oct 02  
24  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
Table 24 Logic levels and function of SRCV1  
DATA BYTE  
AS OUTPUT  
AS INPUT  
FUNCTION  
SRCV11  
SRCV10  
0
0
1
0
1
0
VS  
FS  
VS  
FS  
vertical sync each field; default after reset  
frame sync (odd/even)  
FSEQ  
FSEQ  
field sequence, vertical sync every fourth field  
(PAL = 0), eighth field (PAL = 1) or twelfth field  
(SECAM = 1)  
1
1
not applicable not applicable  
Table 25 Subaddress 6C and 6D  
DATA BYTE  
DESCRIPTION  
HTRIG  
sets the horizontal trigger phase related to signal on RCV1 or RCV2 input  
values above 1715 (FISE = 1) or 1727 (FISE = 0) are not allowed  
increasing HTRIG decreases delays of all internally generated timing signals  
reference mark: analog output horizontal sync (leading slope) coincides with active edge of RCV  
used for triggering at HTRIG = tbf (tbf)  
Table 26 Subaddress 6D  
DATA BYTE LOGIC LEVEL  
DESCRIPTION  
VTRIG  
sets the vertical trigger phase related to signal on RCV1 input  
increasing VTRIG decreases delays of all internally generated timing signals,  
measured in half lines  
variation range of VTRIG = 0 to 31 (1FH)  
Table 27 Subaddress 6E  
DATA BYTE LOGIC LEVEL  
DESCRIPTION  
SBLBN  
0
1
vertical blanking is defined by programming of FAL and LAL; default after reset  
vertical blanking is forced in accordance with “CCIR 624” (50 Hz) or RS170A (60 Hz)  
selects the phase reset mode of the colour subcarrier generator; see Table 28  
field length control; see Table 29  
PHRES  
FLC  
Table 28 Logic levels and function of PHRES  
DATA BYTE  
FUNCTION  
PHRES1  
PHRES0  
0
0
1
1
0
1
0
1
no reset or reset via RTCI from SAA7111 if bit RTCE = 1; default after reset  
reset every two lines or SECAM-specific if bit SECAM = 1  
reset every eight fields  
reset every four fields  
1996 Oct 02  
25  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
Table 29 Logic levels and function of FLC  
DATA BYTE  
FUNCTION  
FLC1  
FLC0  
0
0
1
1
0
1
0
1
interlaced 312.5 lines/field at 50 Hz, 262.5 lines/field at 60 Hz; default after reset  
non-interlaced 312 lines/field at 50 Hz, 262 lines/field at 60 Hz  
non-interlaced 313 lines/field at 50 Hz, 263 lines/field at 60 Hz  
non-interlaced 313 lines/field at 50 Hz, 263 lines/field at 60 Hz  
Table 30 Subaddress 6F  
DATA BYTE LOGIC LEVEL  
DESCRIPTION  
CCEN  
0
1
enables individual Line 21 encoding; see Table 31  
disables teletext insertion  
TTXEN  
enables teletext insertion  
SCCLN  
selects the actual line, where closed caption or extended data are encoded  
line = (SCCLN + 4) for M-systems  
line = (SCCLN + 1) for other systems  
Table 31 Logic levels and function of CCEN  
DATA BYTE  
FUNCTION  
CCEN1  
CCEN0  
0
0
1
1
0
1
0
1
Line 21 encoding off  
enables encoding in field 1 (odd)  
enables encoding in field 2 (even)  
enables encoding in both fields  
Table 32 Subaddress 70 to 72  
DATA BYTE  
DESCRIPTION  
RCV2S  
start of output signal on RCV2 pin  
values above 1715 (FISE = 1) or 1727 (FISE = 0) are not allowed  
first active pixel at analog outputs (corresponding input pixel coinciding with RCV2) at  
RCV2S = tbfH (tbfH)  
RCV2E  
end of output signal on RCV2 pin  
values above 1715 (FISE = 1) or 1727 (FISE = 0) are not allowed  
last active pixel at analog outputs (corresponding input pixel coinciding with RCV2) at  
RCV2E = tbfH (tbfH)  
1996 Oct 02  
26  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
Table 33 Subaddress 73 to 75  
DATA BYTE  
DESCRIPTION  
start of signal on pin TTXRQ (standard for 50 Hz field rate = tbf)  
TTXHS  
values above 1715 (FISE = 1) or 1727 (FISE = 0) are not allowed  
end of signal on pin TTXRQ (standard for 50 Hz field rate = TTXHS + 1402)  
values above 1715 (FISE = 1) or 1727 (FISE = 0) are not allowed  
TTXHE  
Table 34 Subaddress 76, 77 and 7C  
DATA BYTE  
DESCRIPTION  
TTXOVS  
TTXOVE  
first line of occurrence of signal on pin TTXRQ in odd field = TTXOVS + 1 (50 Hz field rate)  
last line of occurrence of signal on pin TTXRQ in odd field = TTXOVE (50 Hz field rate)  
Table 35 Subaddress 78, 79 and 7C  
DATA BYTE  
DESCRIPTION  
TTXEVS  
TTXEVE  
first line of occurrence of signal on pin TTXRQ in even field = TTXEVS + 1 (50 Hz field rate)  
last line of occurrence of signal on pin TTXRQ in even field = TTXEVE (50 Hz field rate)  
Table 36 Subaddress 7A to 7C  
DATA BYTE  
DESCRIPTION  
FAL  
first active line = FAL + 4 for M-systems, = FAL + 1 for other systems, measured in lines  
FAL = 0 coincides with the first field synchronization pulse  
LAL  
last active line = LAL + 3 for M-systems, = LAL for other system, measured in lines  
LAL = 0 coincides with the first field synchronization pulse  
Table 37 Subaddress 7A to 7C  
DATA BYTE  
DESCRIPTION  
LINE  
individual lines in both fields (PAL counting) can be disabled for insertion of teletext by the respective  
bits, disabled line = LINExx (50 Hz field rate)  
this bit mask is effective only, if the lines are enabled by TTXOVS/TTXOVE and TTXEVS/TTXEVE  
SUBADDRESSES  
In subaddresses 5B, 5C, 5D, 5E and 62 all IRE values are rounded up.  
1996 Oct 02  
27  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
Slave Transmitter  
Table 38 Slave transmitter (slave address 89H or 8DH)  
DATA BYTE  
D4 D3  
REGISTER  
SUBADDRESS  
FUNCTION  
D7  
D6  
D5  
D2  
D1  
D0  
Status byte  
VER2  
VER1  
VER0 CCRDO CCRDE  
0
FSEQ  
O_E  
Table 39 No subaddress  
DATA BYTE LOGIC LEVEL  
DESCRIPTION  
VER  
Version identification of the device. It will be changed with all versions of the IC that  
have different programming models. Current Version is 001 binary.  
CCRDO  
1
0
Closed caption bytes of the odd field have been encoded.  
The bit is reset after information has been written to the subaddresses 67 and 68. It  
is set immediately after the data has been encoded.  
CCRDE  
FSEQ  
O_E  
1
0
Closed caption bytes of the even field have been encoded.  
The bit is reset after information has been written to the subaddresses 69 and 6A.  
It is set immediately after the data has been encoded.  
1
During first field of a sequence (repetition rate: NTSC = 4 fields, PAL = 8 fields,  
SECAM = 12 fields.  
0
1
0
Not first field of a sequence.  
During even field.  
During odd field.  
1996 Oct 02  
28  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
MBE737  
6
G
v
(dB)  
0
6  
12  
18  
24  
(1)  
(2)  
30  
36  
42  
48  
54  
0
2
4
6
8
10  
12  
14  
f (MHz)  
(1) SCBW = 1.  
(2) SCBW = 0.  
Fig.5 Chrominance transfer characteristic 1.  
MBE735  
handbook, halfpage  
2
G
v
(dB)  
0
(1)  
(2)  
2  
4  
6  
0
0.4  
0.8  
1.2  
1.6  
f (MHz)  
(1) SCBW = 1.  
(2) SCBW = 0.  
Fig.6 Chrominance transfer characteristic 2.  
29  
1996 Oct 02  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
MGD672  
6
G
v
(dB)  
(4)  
0
(2)  
(3)  
6  
12  
18  
(1)  
24  
30  
36  
42  
48  
54  
0
2
4
6
8
10  
12  
14  
f (MHz)  
(1) CCRS1 = 0; CCRS0 = 1.  
(2) CCRS1 = 1; CCRS0 = 0.  
(3) CCRS1 = 1; CCRS0 = 1.  
(4) CCRS1 = 0; CCRS0 = 0.  
Fig.7 Luminance transfer characteristic 1.  
MBE736  
handbook, halfpage  
1
G
v
(dB)  
(1)  
0
1  
2  
3  
4  
5  
0
2
4
6
f (MHz)  
(1) CCRS1 = 0; CCRS0 = 0.  
Fig.8 Luminance transfer characteristic 2.  
30  
1996 Oct 02  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
MGB708  
6
G
v
(dB)  
0
6  
12  
18  
24  
30  
36  
42  
48  
54  
0
2
4
6
8
10  
12  
14  
f (MHz)  
Fig.9 Luminance transfer characteristic in RGB.  
MGB706  
6
G
v
(dB)  
0
6  
12  
18  
24  
30  
36  
42  
48  
54  
0
2
4
6
8
10  
12  
14  
f (MHz)  
Fig.10 Colour difference transfer characteristic in RGB.  
31  
1996 Oct 02  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
MGB705  
10  
G
v
(dB)  
8
6
4
2
0
0
0.2  
0.4  
0.6  
0.8  
1
1.2  
1.4  
1.6  
f (MHz)  
Fig.11 Gain of SECAM pre-emphasis.  
MGB704  
30  
ϕ
(deg)  
20  
10  
0
0
0.2  
0.4  
0.6  
0.8  
1
1.2  
1.4  
1.6  
f (MHz)  
Fig.12 Phase of SECAM pre-emphasis.  
32  
1996 Oct 02  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
MGB703  
20  
G
v
(dB)  
16  
12  
8
4
0
0
0.2  
0.4  
0.6  
0.8  
1
1.2  
1.4  
1.6  
f (MHz)  
Fig.13 Gain of SECAM anti-Cloche.  
MGB702  
80  
ϕ
(deg)  
60  
40  
20  
0
0
0.2  
0.4  
0.6  
0.8  
1
1.2  
1.4  
1.6  
f (MHz)  
Fig.14 Phase of SECAM anti-Cloche.  
33  
1996 Oct 02  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
CHARACTERISTICS  
VDDD(3) = 3.0 to 3.6 V; VDDD(5) = 4.75 to 5.25 V; Tamb = 0 to +70 °C; unless otherwise specified.  
SYMBOL  
PARAMETER  
CONDITIONS  
MIN.  
MAX.  
UNIT  
Supply  
VDDA(3)  
VDDD(3)  
VDDD(5)  
IDDA  
analog supply voltage (3.3 V)  
digital supply voltage (3.3 V)  
digital supply voltage (5 V)  
analog supply current  
3.1  
3.5  
3.6  
V
3.0  
4.75  
V
5.25  
110  
80  
V
note 1  
mA  
mA  
mA  
IDDD(3)  
IDDD(5)  
digital supply current (3.3 V)  
digital supply current (5 V)  
note 1  
note 1  
10  
Inputs  
VIL  
LOW level input voltage  
0.5  
+0.8  
V
(except SDA, SCL, AP, SP and XTALI)  
VIH  
HIGH level input voltage  
2.0  
VDDD(5) + 0.5 V  
(except LLC, SDA, SCL, AP, SP and XTALI)  
HIGH level input voltage (LLC)  
input leakage current  
input capacitance  
2.4  
VDDD(5) + 0.5 V  
ILI  
Ci  
1
µA  
clocks  
10  
8
pF  
pF  
pF  
data  
I/Os at high impedance  
8
Outputs  
VOL  
LOW level output voltage  
(except SDA and XTALO)  
note 2  
note 2  
note 2  
0
0.6  
V
VOH  
HIGH level output voltage  
(except LLC, SDA, and XTALO)  
2.4  
2.6  
VDDD(5) + 0.5 V  
VDDD(5) + 0.5 V  
HIGH level output voltage (LLC)  
I2C-bus; SDA and SCL  
VIL  
VIH  
Ii  
LOW level input voltage  
0.5  
3.0  
10  
+1.5  
V
HIGH level input voltage  
input current  
VDDD(5) + 0.5 V  
Vi = LOW or HIGH  
IOL = 3 mA  
+10  
0.4  
µA  
VOL  
Io  
LOW level output voltage (SDA)  
output current  
V
during acknowledge  
3
mA  
Clock timing (LLC)  
TLLC  
cycle time  
note 3  
note 4  
note 3  
note 3  
34  
40  
41  
60  
5
ns  
%
δ
tr  
tf  
duty factor tHIGH/TLLC  
rise time  
ns  
ns  
fall time  
6
Input timing  
tSU;DAT  
input data set-up time (any other except  
CDIR, SCL, SDA, RESET, AP and SP)  
6
3
ns  
ns  
tHD;DAT  
input data hold time (any other except  
CDIR, SCL, SDA, RESET, AP and SP)  
1996 Oct 02  
34  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
SYMBOL  
PARAMETER  
CONDITIONS  
MIN.  
MAX.  
UNIT  
Crystal oscillator  
fn  
nominal frequency (usually 27 MHz)  
3rd harmonic  
30  
MHz  
106  
f/fn  
permissible deviation of nominal frequency note 5  
50  
+50  
CRYSTAL SPECIFICATION  
Tamb  
CL  
operating ambient temperature  
0
8
70  
°C  
pF  
load capacitance  
RS  
C1  
series resistance  
80  
motional capacitance (typical)  
parallel capacitance (typical)  
1.5 20% 1.5 +20%  
3.5 20% 3.5 +20%  
fF  
C0  
pF  
Data and reference signal output timing  
CL  
th  
output load capacitance  
output hold time  
7.5  
4
40  
pF  
ns  
ns  
td  
output delay time  
25  
CHROMA, Y, CVBS and RGB outputs  
Vo(p-p)  
Rint  
RL  
output signal voltage (peak-to-peak value)  
note 6  
1.35  
1
1.45  
3
V
internal serial resistance  
output load resistance  
75  
10  
300  
B
output signal bandwidth of DACs  
LF integral linearity error of DACs  
LF differential linearity error of DACs  
3 dB  
MHz  
LSB  
LSB  
ILE  
DLE  
±2  
±1  
Notes  
1. At maximum supply voltage with highly active input signals.  
2. The levels have to be measured with load circuits of 1.2 kto 3.0 V (standard TTL load) and CL = 25 pF.  
3. The data is for both input and output direction.  
4. With LLC in input mode. In output mode, with a crystal connected to XTALO/XTALI duty factor is typically 50%.  
5. If an internal oscillator is used, crystal deviation of nominal frequency is directly proportional to the deviation of  
subcarrier frequency and line/field frequency.  
6. For full digital range, without load, VDDA = 3.3 V. The typical voltage swing is 1.4 V, the typical minimum output  
voltage (digital zero at DAC) is 0.2 V.  
1996 Oct 02  
35  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
T
LLC  
t
HIGH  
2.6 V  
1.5 V  
0.6 V  
LLC clock output  
t
t
t
HD; DAT  
f
r
T
LLC  
t
HIGH  
2.4 V  
1.5 V  
0.8 V  
LLC clock input  
t
t
f
t
HD; DAT  
r
t
SU; DAT  
2.0 V  
valid  
input data  
valid  
not valid  
0.8 V  
t
d
t
HD; DAT  
2.4 V  
valid  
output data  
valid  
not valid  
0.6 V  
MBE742  
Fig.15 Clock data timing.  
LLC  
MP(n)  
RCV2  
Cb(0)  
Y(0)  
Cr(0)  
Y(1)  
Cb(2)  
MGB699  
The data demultiplexing phase is coupled to the internal horizontal phase.  
The phase of the RCV2 signal is programmed to tbf (tbf for 50 Hz) in this example in output mode (RCV2S).  
Fig.16 Functional timing.  
1996 Oct 02  
36  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
LLC  
CREF  
MP(n)  
Y(0)  
Y(1)  
Y(2)  
Y(3)  
Y(4)  
DP(n)  
RCV2  
Cb(0)  
Cr(0)  
Cb(2)  
Cr(2)  
Cb(4)  
MBE739  
The data demultiplexing phase is coupled to the internal horizontal phase.  
The CREF signal applies only for the 16 line digital TV format, because these signals are only valid in 13.5 MHz.  
The phase of the RCV2 signal is programmed to tbf (tbf for 50 Hz) in this example in output mode (RCV2S).  
Fig.17 Digital TV timing.  
sequence reserved (2)  
bit (1)  
H/L transition  
4 bits  
5 bits  
reset  
count start  
reserved  
reserved  
bit (3)  
LOW  
HPLL  
increment  
FSCPLL increment (4)  
128  
13  
0
21  
0
RTCI  
time slot:  
01  
14  
19  
67 68  
not used in  
SAA7182A/83A  
valid  
sample sample  
invalid  
8/LLC  
MGD673  
(1) Sequence bit:  
PAL = logic 0 then (R Y) line normal; PAL = logic 1 then (R Y) line inverted.  
NTSC = logic 0 then no change.  
(2) Reserved bits: 235 with 50 Hz systems; 232 with 60 Hz systems.  
(3) Only from SAA7111 decoder.  
(4) SAAA7111 provides (22 : 0) bits, resulting in 3 reserved bits before sequence bit.  
Fig.18 RTCI timing.  
1996 Oct 02  
37  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
Thus 37 TTX bits correspond to 144 LLC clocks, each bit  
has a duration of nearly 4 LLC clocks. The chip-internal  
sequencer and variable phase interpolation filter  
minimizes the phase jitter, and thus generates a  
bandwidth limited signal, which is digital-to-analog  
converted for the CVBS and Y outputs.  
Teletext timing  
Time tFD is the time needed to interpolate input data TTX  
and inserting it into the CVBS and Y output signal, such  
that it appears at tTTX = 10.2 µs after the leading edge of  
the horizontal synchronization pulse.  
Time tPD is the pipeline delay time introduced by the  
source that is gated by TTXRQ in order to deliver TTX  
data.  
At the TTX input, bit duration scheme repeats after 37 TTX  
bits or 144 LLC clocks. The protocol demands that TXX  
bits 10, 19, 28 and 37 are carried by three LLC samples,  
all others by four LLC samples. After a cycle of 37 TTX  
bits, the next bits with three LLC samples are bits 47, 56,  
65 and 74; this scheme holds for all succeeding cycles of  
37 TTX bits, until 360 TTX bits (including 16 run-in bits)  
are completed. For every additional line with TTX data, the  
bit duration scheme starts in the same way.  
Since the pulse representing the TTXRQ signal is fully  
programmable in duration and rising/falling edges (TTXHS  
and TTXHE), the TTX data is always inserted at the correct  
position of 10.2 µs after the leading edge of outgoing  
horizontal synchronization pulse.  
Time tTTXWin is the internally used insertion window for  
TTX data; it has a constant length that allows insertion of  
360 teletext bits (maximum) at a text data rate of  
6.9375 Mbits/s. The insertion window is not opened if the  
control bit TTXEN is zero.  
Using appropriate programming, all suitable lines of the  
odd field (TTXOVS and TTXOVE) plus all suitable lines of  
the even field (TTXEVS and TTXEVE) can be used for  
teletext insertion.  
TELETEXT PROTOCOL  
The frequency relationship between TTX bit clock and the  
system clock LLC for 50 Hz field rate is given by the  
relationship of line frequency multiples, which means  
1728/444.  
CVBS/Y  
t
t
TTX  
TTXWin  
textbit #:  
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24  
TTX  
4
3
4
1/LLC  
4
3
4
1/LLC  
t
t
PD  
FD  
TTXRQ  
MGB701  
Fig.19 Teletext timing diagram.  
38  
1996 Oct 02  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
APPLICATION INFORMATION  
GM6D74  
o o k , f u l l p a g e w i d t h  
1996 Oct 02  
39  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
GM0D7  
b o o k , f u l l p a g e w i d t h  
1996 Oct 02  
40  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
Values for the external series resistors result from a 75 Ω  
Analog output voltages  
load (see Figs 20 and 21).  
The analog output voltages are dependent on the open  
loop voltage of the operational amplifiers for full-scale  
conversion (typical value 1.4 V), the internal series resistor  
(typical value 2 ), the external series resistor and the  
external load impedance.  
The analog inputs RI, GI, and BI are shifted first by an  
offset of 0.16 V (typical value), followed by an amplification  
of 1.72 (typical value). For an input voltage of 0 to 0.7 V an  
open loop output voltage of 0.28 to 1.48 V is achieved,  
resulting in Vo = 0.86 V (p-p) with an internal series  
resistor of 2 , an external series resistor of 27 at a 75 Ω  
load impedance.  
The digital output signals in front of the DACs under  
nominal conditions occupy different conversion ranges, as  
indicated in Table 40 for a 100  
100 colour bar signal.  
Table 40 Digital output signals conversion range  
CONVERSION RANGE (peak-to-peak  
CVBS, SYNC  
TIP-TO-PEAK CARRIER  
(digits)  
RGB (Y)  
BLACK-TO-WHITE AT GDY = GDC = 6  
(digits)  
Y (VBS) SYNC TIP-TO-WHITE  
(digits)  
1023  
888  
712  
1996 Oct 02  
41  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
PACKAGE OUTLINES  
PLCC84: plastic leaded chip carrier; 84 leads  
SOT189-2  
e
e
D
E
y
X
A
74  
54  
Z
53  
E
75  
b
p
b
1
w
M
84  
1
H
E
E
pin 1 index  
e
A
A
1
A
4
(A )  
3
k
L
1
p
11  
β
33  
k
detail X  
12  
32  
v
M
A
e
Z
D
D
B
H
v
M
B
D
0
5
10 mm  
scale  
DIMENSIONS (millimetre dimensions are derived from the original inch dimensions)  
(1)  
(1)  
A
min.  
A
max.  
k
1
max.  
Z
Z
E
(1)  
(1)  
1
4
D
UNIT  
mm  
A
A
b
D
E
e
e
e
H
H
k
L
v
w
y
β
b
D
E
D
E
p
3
p
1
max. max.  
4.57  
4.19  
0.81 29.41 29.41  
0.66 29.21 29.21  
28.70 28.70 30.35 30.35 1.22  
27.69 27.69 30.10 30.10 1.07  
1.44  
1.02  
0.53  
0.33  
0.51  
0.51 0.25 3.30  
0.020 0.01 0.13  
1.27  
0.05  
0.18 0.18 0.10 2.16 2.16  
o
45  
0.180  
0.165  
0.032 1.158 1.158  
0.026 1.150 1.150  
1.130 1.130 1.195 1.195 0.048  
1.090 1.090 1.185 1.185 0.042  
0.057  
0.040  
0.021  
0.013  
inches  
0.020  
0.007 0.007 0.004 0.085 0.085  
Note  
1. Plastic or metal protrusions of 0.01 inches maximum per side are not included.  
REFERENCES  
OUTLINE  
EUROPEAN  
PROJECTION  
ISSUE DATE  
VERSION  
IEC  
JEDEC  
EIAJ  
92-11-17  
95-03-11  
SOT189-2  
1996 Oct 02  
42  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
QFP80: plastic quad flat package; 80 leads (lead length 1.95 mm); body 14 x 20 x 2.8 mm  
SOT318-2  
y
X
A
64  
65  
41  
40  
Z
E
e
Q
A
2
H
A
E
(A )  
3
E
A
1
w M  
p
θ
pin 1 index  
L
p
b
L
80  
25  
detail X  
1
24  
w M  
Z
v
M
M
D
A
B
b
p
e
D
B
H
v
D
0
5
scale  
10 mm  
DIMENSIONS (mm are the original dimensions)  
A
(1)  
(1)  
(1)  
(1)  
UNIT  
A
A
A
b
c
D
E
e
H
D
H
L
L
Q
v
w
y
Z
Z
E
θ
1
2
3
p
E
p
D
max.  
7o  
0o  
0.25 2.90  
0.05 2.65  
0.45 0.25 20.1 14.1  
0.30 0.14 19.9 13.9  
24.2 18.2  
23.6 17.6  
1.0  
0.6  
1.4  
1.2  
1.0  
0.6  
1.2  
0.8  
mm  
3.2  
0.25  
0.8  
1.95  
0.2  
0.2  
0.1  
Note  
1. Plastic or metal protrusions of 0.25 mm maximum per side are not included.  
REFERENCES  
OUTLINE  
EUROPEAN  
PROJECTION  
ISSUE DATE  
VERSION  
IEC  
JEDEC  
EIAJ  
92-12-15  
95-02-04  
SOT318-2  
1996 Oct 02  
43  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
The package footprint must incorporate solder thieves at  
the downstream corners.  
SOLDERING  
Introduction  
QFP  
There is no soldering method that is ideal for all IC  
packages. Wave soldering is often preferred when  
through-hole and surface mounted components are mixed  
on one printed-circuit board. However, wave soldering is  
not always suitable for surface mounted ICs, or for  
printed-circuits with high population densities. In these  
situations reflow soldering is often used.  
Wave soldering is not recommended for QFP packages.  
This is because of the likelihood of solder bridging due to  
closely-spaced leads and the possibility of incomplete  
solder penetration in multi-lead devices.  
If wave soldering cannot be avoided, the following  
conditions must be observed:  
This text gives a very brief insight to a complex technology.  
A more in-depth account of soldering ICs can be found in  
our “IC Package Databook” (order code 9398 652 90011).  
A double-wave (a turbulent wave with high upward  
pressure followed by a smooth laminar wave)  
soldering technique should be used.  
Reflow soldering  
The footprint must be at an angle of 45° to the board  
direction and must incorporate solder thieves  
downstream and at the side corners.  
Reflow soldering techniques are suitable for all PLCC and  
QFP packages.  
Even with these conditions, do not consider wave  
soldering the following packages: QFP52 (SOT379-1),  
QFP100 (SOT317-1), QFP100 (SOT317-2),  
The choice of heating method may be influenced by larger  
PLCC or QFP packages (44 leads, or more). If infrared or  
vapour phase heating is used and the large packages are  
not absolutely dry (less than 0.1% moisture content by  
weight), vaporization of the small amount of moisture in  
them can cause cracking of the plastic body. For more  
information, refer to the Drypack chapter in our “Quality  
Reference Handbook” (order code 9397 750 00192).  
QFP100 (SOT382-1) or QFP160 (SOT322-1).  
METHOD (PLCC AND QFP)  
During placement and before soldering, the package must  
be fixed with a droplet of adhesive. The adhesive can be  
applied by screen printing, pin transfer or syringe  
dispensing. The package can be soldered after the  
adhesive is cured.  
Reflow soldering requires solder paste (a suspension of  
fine solder particles, flux and binding agent) to be applied  
to the printed-circuit board by screen printing, stencilling or  
pressure-syringe dispensing before package placement.  
Maximum permissible solder temperature is 260 °C, and  
maximum duration of package immersion in solder is  
10 seconds, if cooled to less than 150 °C within  
Several techniques exist for reflowing; for example,  
thermal conduction by heated belt. Dwell times vary  
between 50 and 300 seconds depending on heating  
method. Typical reflow temperatures range from  
215 to 250 °C.  
6 seconds. Typical dwell time is 4 seconds at 250 °C.  
A mildly-activated flux will eliminate the need for removal  
of corrosive residues in most applications.  
Preheating is necessary to dry the paste and evaporate  
the binding agent. Preheating duration: 45 minutes at  
45 °C.  
Repairing soldered joints  
Fix the component by first soldering two diagonally-  
opposite end leads. Use only a low voltage soldering iron  
(less than 24 V) applied to the flat part of the lead. Contact  
time must be limited to 10 seconds at up to 300 °C. When  
using a dedicated tool, all other leads can be soldered in  
one operation within 2 to 5 seconds between  
270 and 320 °C.  
Wave soldering  
PLCC  
Wave soldering techniques can be used for all PLCC  
packages if the following conditions are observed:  
A double-wave (a turbulent wave with high upward  
pressure followed by a smooth laminar wave) soldering  
technique should be used.  
The longitudinal axis of the package footprint must be  
parallel to the solder flow.  
1996 Oct 02  
44  
Philips Semiconductors  
Preliminary specification  
Digital Video Encoder (EURO-DENC2)  
SAA7182A; SAA7183A  
DEFINITIONS  
Data sheet status  
Objective specification  
Preliminary specification  
Product specification  
This data sheet contains target or goal specifications for product development.  
This data sheet contains preliminary data; supplementary data may be published later.  
This data sheet contains final product specifications.  
Limiting values  
Limiting values given are in accordance with the Absolute Maximum Rating System (IEC 134). Stress above one or  
more of the limiting values may cause permanent damage to the device. These are stress ratings only and operation  
of the device at these or at any other conditions above those given in the Characteristics sections of the specification  
is not implied. Exposure to limiting values for extended periods may affect device reliability.  
Application information  
Where application information is given, it is advisory and does not form part of the specification.  
LIFE SUPPORT APPLICATIONS  
These products are not designed for use in life support appliances, devices, or systems where malfunction of these  
products can reasonably be expected to result in personal injury. Philips customers using or selling these products for  
use in such applications do so at their own risk and agree to fully indemnify Philips for any damages resulting from such  
improper use or sale.  
PURCHASE OF PHILIPS I2C COMPONENTS  
Purchase of Philips I2C components conveys a license under the Philips’ I2C patent to use the  
components in the I2C system provided the system conforms to the I2C specification defined by  
Philips. This specification can be ordered using the code 9398 393 40011.  
1996 Oct 02  
45  

相关型号:

SAA7165WP

IC SPECIALTY CONSUMER CIRCUIT, PQCC44, PLASTIC, LCC-44, Consumer IC:Other
NXP

SAA7167

YUV-to-RGB Digital-to-Analog Converter DAC
NXP

SAA7167A

D/A Converter, 8-Bit, 3 Func, CMOS, PQFP48
PHILIPS

SAA7167AH

IC SPECIALTY CONSUMER CIRCUIT, PQFP48, 7 X 7 MM, 1.40 MM HEIGHT, PLASTIC, MS-026, SOT-313-2, LQFP-48, Consumer IC:Other
NXP

SAA7167AH/00

IC SPECIALTY CONSUMER CIRCUIT, PQFP48, PLASTIC, SOT-313-2, LQFP-48, Consumer IC:Other
NXP

SAA7167AHB

IC SPECIALTY CONSUMER CIRCUIT, PQFP48, PLASTIC, SOT-313-2, LQFP-48, Consumer IC:Other
NXP

SAA7167AHB-S

IC SPECIALTY CONSUMER CIRCUIT, PQFP48, PLASTIC, SOT-313-2, LQFP-48, Consumer IC:Other
NXP

SAA7167H/00

IC SPECIALTY CONSUMER CIRCUIT, PQFP48, PLASTIC, SOT-313-2, LQFP-48, Consumer IC:Other
NXP

SAA7174

Analog TV video and stereo decoder for computing and consumer applications in Europe
NXP

SAA7182

Digital Video Encoder EURO-DENC
NXP

SAA7182A

Digital Video Encoder EURO-DENC2
NXP

SAA7182AGP

IC COLOR SIGNAL ENCODER, PQFP80, 14 X 20 MM, 2.80 MM HEIGHT, PLASTIC, SOT-318-2, QFP-80, Color Signal Converter
NXP